{"passed":false,"scope":"Real HTTPS / ASR / ECAPA / local+API LLM / TTS diagnostics.\n\nDigital PCM injection uses the actual browser playback clock. Not a physical\nmicrophone/acoustic echo certification. Every run is archived, including failures.\nPublic ESC-50 inputs stay local; the receipt includes provenance, never raw audio.\n","started":"20261027T132148Z","group":"all","source_sha256":"50fa9e83a04a033e86a9717a2d321dba92a778529cba6e8f09c32cf3782ad639","injections":[{"browser_start_ms":4482.700000762939,"browser_end_ms":9486.70000076294,"packets":250,"max_schedule_lag_ms":9.299999237060547,"mean_schedule_lag_ms":0.9500002746582031,"stream_start_sample":38400,"samples":80000,"sha256":"474c118b913bd8c001d69ff27613cee094b8afecd9cff9c1a044bba901e41189"},{"browser_start_ms":1849.6000022888184,"browser_end_ms":17849.10000228882,"packets":800,"max_schedule_lag_ms":62.29999923706055,"mean_schedule_lag_ms":1.713374500274658,"stream_start_sample":6400,"samples":256000,"sha256":"3225a2bc288de755d3bdc17787d198b6e61fc8463ae5a76d7c775ee957fda44d"},{"browser_start_ms":4300.799999237061,"browser_end_ms":9300.199996948242,"packets":250,"max_schedule_lag_ms":7.899997711181641,"mean_schedule_lag_ms":0.5779998474121094,"stream_start_sample":38400,"samples":80000,"sha256":"7d3b3ad57d3c01aeecde256dd1656177f845391c1b11f9ee4e501537fe7d0bab"},{"browser_start_ms":3778.2999992370605,"browser_end_ms":8794.79999923706,"packets":250,"max_schedule_lag_ms":55.20000076293945,"mean_schedule_lag_ms":3.777200607299805,"stream_start_sample":38400,"samples":80000,"sha256":"38d153ad9d1a314e615fd8c84459cef4b000b3c044af3397e12529e02b16a028"},{"browser_start_ms":3887.2999992370605,"browser_end_ms":8886.5,"packets":250,"max_schedule_lag_ms":94.29999923706055,"mean_schedule_lag_ms":4.969999710083008,"stream_start_sample":38400,"samples":80000,"sha256":"2d035982c15f176648ffe3107236fcb1df5e864c9b0bace93ff89c63bbe1f6ce"},{"browser_start_ms":3777.5,"browser_end_ms":8777,"packets":250,"max_schedule_lag_ms":7.200000762939453,"mean_schedule_lag_ms":1.062400405883789,"stream_start_sample":38400,"samples":80000,"sha256":"048b79fbc94b62cce2f131873bf1bace053c75b9eb6cc3942e509ad6fa1275e7"},{"browser_start_ms":4288.299999237061,"browser_end_ms":7287.600002288818,"packets":150,"max_schedule_lag_ms":6,"mean_schedule_lag_ms":0.7926670328776042,"stream_start_sample":41600,"samples":48000,"sha256":"917ef60d2530297772bfcf56a58dc10c27a6188586bea9316f0e5875fefb3cb4"},{"browser_start_ms":4240.599998474121,"browser_end_ms":7244.599998474121,"packets":150,"max_schedule_lag_ms":9.10000228881836,"mean_schedule_lag_ms":0.8226671346028646,"stream_start_sample":41600,"samples":48000,"sha256":"917ef60d2530297772bfcf56a58dc10c27a6188586bea9316f0e5875fefb3cb4"},{"browser_start_ms":4074.1000022888184,"browser_end_ms":8513.5,"packets":222,"max_schedule_lag_ms":7.799999237060547,"mean_schedule_lag_ms":1.1693689500963367,"stream_start_sample":38400,"samples":70880,"sha256":"5d665c310e221c7e3af7ac9ed35c4762517c8a14d93056ea055ebafca2b766b7"},{"browser_start_ms":24711.20000076294,"browser_end_ms":28672.79999923706,"packets":198,"max_schedule_lag_ms":5.099998474121094,"mean_schedule_lag_ms":0.41666655106977984,"stream_start_sample":362080,"samples":63200,"sha256":"dd2ca90a85488b86997a05c74e6fe3017ea091f6832a627d073e141fc3738249"},{"browser_start_ms":39579.60000228882,"browser_end_ms":44578.79999923706,"packets":250,"max_schedule_lag_ms":3.799999237060547,"mean_schedule_lag_ms":0.42519961547851565,"stream_start_sample":594880,"samples":79840,"sha256":"4dcb7fd02b14b2e29cbc064502468df8c18d7b862839701a500f21f6821559fd"},{"browser_start_ms":57152,"browser_end_ms":61111.70000076294,"packets":198,"max_schedule_lag_ms":5.400001525878906,"mean_schedule_lag_ms":0.6095962139091107,"stream_start_sample":869920,"samples":63200,"sha256":"dd2ca90a85488b86997a05c74e6fe3017ea091f6832a627d073e141fc3738249"}],"cases":[{"case":"music_onset","events":[{"type":"interaction","enabled":true,"helpful_enabled":false,"rules":[],"decision_model":"convaiinnovations/laya-multilingual (Stop) · dialogue model (context)","decision_interval_ms":320,"checks":0,"last_decision":null,"waiting_for":"No agreed rules yet","attempts":0,"semantic_error":null,"scope":"session-only conversational rules; acoustic verification is separate","at":2436.7000007629395},{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a consistent rhythmic pulse.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":3.44,"tokens":35,"preprocess_ms":9.06,"encoder_gpu_ms":11.95,"decoder_gpu_ms":294,"inference_ms":306.18,"total_ms":315.25,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":328.43,"capture_ms":3240,"event_age_ms":3568.43,"kind":"music","eligible":true,"suppression_reason":null,"at":8087.200000762939},{"type":"environment_observation","kind":"music","status":"responding","at":8187.100002288818},{"type":"latency_start","reply_id":1,"input":{"kind":"environment","sound_caption_ms":328.43},"at":8187.400001525879},{"type":"reply","text":"Oh, da läuft gerade ein ziemlich treibender elektronischer Beat.","reply_id":1,"performance":{"sentences":[{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Oh, da läuft gerade ein ziemlich treibender elektronischer Beat.","start_s":0,"end_s":3.06}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Oh, da läuft gerade ein ziemlich treibender elektronischer Beat.","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":577.51,"finish_reason":"stop","usage":{"prompt_tokens":1681,"completion_tokens":20,"total_tokens":1701,"cost":0.0016770501,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00169399,"upstream_inference_prompt_cost":0.00166419,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":630.78,"generation_calls":1,"llm_total_ms":630.81,"response_complete":true,"sentence_ready_ms":644.94,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":8833},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":577.51,"finish_reason":"stop","usage":{"prompt_tokens":1681,"completion_tokens":20,"total_tokens":1701,"cost":0.0016770501,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00169399,"upstream_inference_prompt_cost":0.00166419,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":630.78,"generation_calls":1,"llm_total_ms":630.81,"response_complete":true,"answer_first_audio_ms":1411.48,"tts_first_audio_ms":766.44,"first_audio_ms":1411.49},"at":9598.100002288818},{"type":"pipeline_metrics","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":577.51,"finish_reason":"stop","usage":{"prompt_tokens":1681,"completion_tokens":20,"total_tokens":1701,"cost":0.0016770501,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00169399,"upstream_inference_prompt_cost":0.00166419,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":630.78,"generation_calls":1,"llm_total_ms":630.81,"response_complete":true,"answer_first_audio_ms":1411.48,"tts_first_audio_ms":766.44,"first_audio_ms":1411.49,"tts_server_stages":[{"generation_ms":1509,"alignment_ms":46,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1680,"performance":{"device_seconds":1.2196214051246643,"raw_audio_seconds":4.24,"aggregate_rtf":0.2876465578124208,"audio_seconds_per_second":3.476488672783343,"batches":[{"batch_size":1,"device_seconds":1.2196214051246643,"audio_seconds":[4.24],"model_wall_seconds":1.3903858065605164,"first_audio_seconds":0.4711086120605469,"graph_capture_seconds":0,"live_callback_seconds":0.16608211398124695}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1693.34,"audio_s":4.03}],"total_ms":2338.4,"audio_s":4.03,"sentences":1,"generation_rtf":0.5802},"at":10533.20000076294},{"type":"turn_done","barged":false,"at":14072.5}],"reply":"Oh, da läuft gerade ein ziemlich treibender elektronischer Beat.","captions":[{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a consistent rhythmic pulse.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":3.44,"tokens":35,"preprocess_ms":9.06,"encoder_gpu_ms":11.95,"decoder_gpu_ms":294,"inference_ms":306.18,"total_ms":315.25,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":328.43,"capture_ms":3240,"event_age_ms":3568.43,"kind":"music","eligible":true,"suppression_reason":null,"at":8087.200000762939}],"pcm_chunks":5,"page_errors":[],"passed":true},{"case":"music_already_playing","events":[{"type":"interaction","enabled":true,"helpful_enabled":false,"rules":[],"decision_model":"convaiinnovations/laya-multilingual (Stop) · dialogue model (context)","decision_interval_ms":320,"checks":0,"last_decision":null,"waiting_for":"No agreed rules yet","attempts":0,"semantic_error":null,"scope":"session-only conversational rules; acoustic verification is separate","at":1733.900001525879},{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a consistent rhythmic pulse.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":3.44,"tokens":35,"preprocess_ms":7.71,"encoder_gpu_ms":8.11,"decoder_gpu_ms":271.94,"inference_ms":280.27,"total_ms":287.98,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":299.44,"capture_ms":3240,"event_age_ms":3539.45,"kind":"music","eligible":true,"suppression_reason":null,"at":6684.200000762939},{"type":"environment_observation","kind":"music","status":"responding","at":6711.299999237061},{"type":"latency_start","reply_id":1,"input":{"kind":"environment","sound_caption_ms":299.44},"at":6711.5},{"type":"reply","text":"Oh, das klingt nach einem treibenden Beat!","reply_id":1,"performance":{"sentences":[{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Oh, das klingt nach einem treibenden Beat!","start_s":0,"end_s":2.38}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Oh, das klingt nach einem treibenden Beat!","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":412.26,"finish_reason":"stop","usage":{"prompt_tokens":1680,"completion_tokens":17,"total_tokens":1697,"cost":0.0016716447,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00168853,"upstream_inference_prompt_cost":0.0016632,"upstream_inference_completions_cost":2.533e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":454.29,"generation_calls":1,"llm_total_ms":454.32,"response_complete":true,"sentence_ready_ms":455.39,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":7164.799999237061},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":412.26,"finish_reason":"stop","usage":{"prompt_tokens":1680,"completion_tokens":17,"total_tokens":1697,"cost":0.0016716447,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00168853,"upstream_inference_prompt_cost":0.0016632,"upstream_inference_completions_cost":2.533e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":454.29,"generation_calls":1,"llm_total_ms":454.32,"response_complete":true,"answer_first_audio_ms":1202.67,"tts_first_audio_ms":747.09,"first_audio_ms":1202.68},"at":7912.100002288818},{"type":"pipeline_metrics","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":412.26,"finish_reason":"stop","usage":{"prompt_tokens":1680,"completion_tokens":17,"total_tokens":1697,"cost":0.0016716447,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00168853,"upstream_inference_prompt_cost":0.0016632,"upstream_inference_completions_cost":2.533e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":454.29,"generation_calls":1,"llm_total_ms":454.32,"response_complete":true,"answer_first_audio_ms":1202.67,"tts_first_audio_ms":747.09,"first_audio_ms":1202.68,"tts_server_stages":[{"generation_ms":1053,"alignment_ms":38,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1204,"performance":{"device_seconds":0.8442615194320678,"raw_audio_seconds":2.88,"aggregate_rtf":0.29314636091391244,"audio_seconds_per_second":3.411265269957308,"batches":[{"batch_size":1,"device_seconds":0.8442615194320678,"audio_seconds":[2.88],"model_wall_seconds":0.9400555863976479,"first_audio_seconds":0.46793728637695314,"graph_capture_seconds":0,"live_callback_seconds":0.0955239087343216}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1217.49,"audio_s":2.67}],"total_ms":1673.09,"audio_s":2.67,"sentences":1,"generation_rtf":0.6266},"at":8392.900001525879},{"type":"turn_done","barged":false,"at":11027},{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a simple, repetitive drum pattern.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":3.2,"tokens":37,"preprocess_ms":7.38,"encoder_gpu_ms":9.27,"decoder_gpu_ms":320.61,"inference_ms":330.09,"total_ms":337.47,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":350.04,"capture_ms":2960,"event_age_ms":3310.05,"kind":"music","eligible":false,"suppression_reason":"music_cooldown","at":18526.40000152588}],"reply":"Oh, das klingt nach einem treibenden Beat!","captions":[{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a consistent rhythmic pulse.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":3.44,"tokens":35,"preprocess_ms":7.71,"encoder_gpu_ms":8.11,"decoder_gpu_ms":271.94,"inference_ms":280.27,"total_ms":287.98,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":299.44,"capture_ms":3240,"event_age_ms":3539.45,"kind":"music","eligible":true,"suppression_reason":null,"at":6684.200000762939},{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a simple, repetitive drum pattern.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":3.2,"tokens":37,"preprocess_ms":7.38,"encoder_gpu_ms":9.27,"decoder_gpu_ms":320.61,"inference_ms":330.09,"total_ms":337.47,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":350.04,"capture_ms":2960,"event_age_ms":3310.05,"kind":"music","eligible":false,"suppression_reason":"music_cooldown","at":18526.40000152588}],"pcm_chunks":4,"page_errors":[],"passed":true},{"case":"dog","events":[{"type":"interaction","enabled":true,"helpful_enabled":false,"rules":[],"decision_model":"convaiinnovations/laya-multilingual (Stop) · dialogue model (context)","decision_interval_ms":320,"checks":0,"last_decision":null,"waiting_for":"No agreed rules yet","attempts":0,"semantic_error":null,"scope":"session-only conversational rules; acoustic verification is separate","at":2261.199996948242},{"status":"available","caption":"The audio contains a single, distinct bark. The bark is relatively short in duration and has a medium pitch.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":1.84,"tokens":23,"preprocess_ms":7.74,"encoder_gpu_ms":9.31,"decoder_gpu_ms":178.79,"inference_ms":188.35,"total_ms":196.09,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":208.63,"capture_ms":1640,"event_age_ms":1848.63,"kind":"sound","eligible":true,"suppression_reason":null,"at":8422.79999923706},{"type":"environment_observation","kind":"sound","status":"responding","at":8441},{"type":"latency_start","reply_id":1,"input":{"kind":"environment","sound_caption_ms":208.63},"at":8441.099998474121},{"type":"reply","text":"Kann ich fragen, ob gerade ein kleiner Hund in der Nähe bellt?","reply_id":1,"performance":{"sentences":[{"emotion":"contentment","secondary":null,"blend":0,"intensity":0.55,"direction":"curiosity, soft","text":"Kann ich fragen, ob gerade ein kleiner Hund in der Nähe bellt?","start_s":0,"end_s":4.08}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Kann ich fragen, ob gerade ein kleiner Hund in der Nähe bellt?","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":321.12,"finish_reason":"stop","usage":{"prompt_tokens":1646,"completion_tokens":21,"total_tokens":1667,"cost":0.0016442217,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00166083,"upstream_inference_prompt_cost":0.00162954,"upstream_inference_completions_cost":3.129e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":368.58,"generation_calls":1,"llm_total_ms":368.61,"response_complete":true,"sentence_ready_ms":369.48,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"delivery.soft.soft"},"sources":[],"lookups":[],"at":8811.899997711182},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":321.12,"finish_reason":"stop","usage":{"prompt_tokens":1646,"completion_tokens":21,"total_tokens":1667,"cost":0.0016442217,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00166083,"upstream_inference_prompt_cost":0.00162954,"upstream_inference_completions_cost":3.129e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":368.58,"generation_calls":1,"llm_total_ms":368.61,"response_complete":true,"answer_first_audio_ms":1158.23,"tts_first_audio_ms":788.64,"first_audio_ms":1158.23},"at":9603.899997711182},{"type":"pipeline_metrics","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":321.12,"finish_reason":"stop","usage":{"prompt_tokens":1646,"completion_tokens":21,"total_tokens":1667,"cost":0.0016442217,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00166083,"upstream_inference_prompt_cost":0.00162954,"upstream_inference_completions_cost":3.129e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":368.58,"generation_calls":1,"llm_total_ms":368.61,"response_complete":true,"answer_first_audio_ms":1158.23,"tts_first_audio_ms":788.64,"first_audio_ms":1158.23,"tts_server_stages":[{"generation_ms":1618,"alignment_ms":38,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1795,"performance":{"device_seconds":1.325214720726013,"raw_audio_seconds":4.64,"aggregate_rtf":0.2856066208461235,"audio_seconds_per_second":3.5013193918174985,"batches":[{"batch_size":1,"device_seconds":1.325214720726013,"audio_seconds":[4.64],"model_wall_seconds":1.4945457726716995,"first_audio_seconds":0.46859561157226565,"graph_capture_seconds":0,"live_callback_seconds":0.16900340467691422}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1805.77,"audio_s":3.95}],"total_ms":2175.37,"audio_s":3.95,"sentences":1,"generation_rtf":0.5507},"at":10641.599998474121},{"type":"turn_done","barged":false,"at":14007.699996948242}],"reply":"Kann ich fragen, ob gerade ein kleiner Hund in der Nähe bellt?","captions":[{"status":"available","caption":"The audio contains a single, distinct bark. The bark is relatively short in duration and has a medium pitch.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":1.84,"tokens":23,"preprocess_ms":7.74,"encoder_gpu_ms":9.31,"decoder_gpu_ms":178.79,"inference_ms":188.35,"total_ms":196.09,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":208.63,"capture_ms":1640,"event_age_ms":1848.63,"kind":"sound","eligible":true,"suppression_reason":null,"at":8422.79999923706}],"pcm_chunks":5,"page_errors":[],"passed":true},{"case":"car_horn","events":[{"type":"interaction","enabled":true,"helpful_enabled":false,"rules":[],"decision_model":"convaiinnovations/laya-multilingual (Stop) · dialogue model (context)","decision_interval_ms":320,"checks":0,"last_decision":null,"waiting_for":"No agreed rules yet","attempts":0,"semantic_error":null,"scope":"session-only conversational rules; acoustic verification is separate","at":1734.7000007629395},{"status":"available","caption":"The audio contains a single, distinct, and loud honking sound. The sound is short in duration and has a clear, resonant quality.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":1.84,"tokens":30,"preprocess_ms":7.89,"encoder_gpu_ms":8.53,"decoder_gpu_ms":213.54,"inference_ms":222.31,"total_ms":230.2,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":240.77,"capture_ms":1640,"event_age_ms":1880.77,"kind":"sound","eligible":true,"suppression_reason":null,"at":5690.900001525879},{"type":"environment_observation","kind":"sound","status":"responding","at":5692.200000762939},{"type":"latency_start","reply_id":1,"input":{"kind":"environment","sound_caption_ms":240.77},"at":5692.799999237061},{"type":"reply","text":"Huch, ein lautes Horn?","reply_id":1,"performance":{"sentences":[{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Huch, ein lautes Horn?","start_s":0,"end_s":1.36}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Huch, ein lautes Horn?","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":325.3,"llm_first_sentence_ms":348.3,"sentence_ready_ms":348.85,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":6045},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":325.3,"llm_first_sentence_ms":348.3,"finish_reason":"stop","usage":{"prompt_tokens":1653,"completion_tokens":20,"total_tokens":1673,"cost":0.0016496073,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00166627,"upstream_inference_prompt_cost":0.00163647,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":404.05,"response_complete":true,"answer_first_audio_ms":1082.98,"tts_first_audio_ms":734.08,"first_audio_ms":1082.99},"at":6777.400001525879},{"type":"reply","text":"Huch, ein lautes Horn? Hörst du das auch?","reply_id":1,"performance":{"sentences":[{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Huch, ein lautes Horn?","start_s":0,"end_s":2.03},{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Hörst du das auch?","start_s":2.03,"end_s":3.3899999999999997}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":1,"sentence_text":"Hörst du das auch?","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":325.3,"llm_first_sentence_ms":348.3,"finish_reason":"stop","usage":{"prompt_tokens":1653,"completion_tokens":20,"total_tokens":1673,"cost":0.0016496073,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00166627,"upstream_inference_prompt_cost":0.00163647,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":404.05,"response_complete":true,"answer_first_audio_ms":1082.98,"tts_first_audio_ms":734.08,"first_audio_ms":1082.99,"tts_server_stages":[{"generation_ms":1000,"alignment_ms":36,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1155,"performance":{"device_seconds":0.7884904770851136,"raw_audio_seconds":2.72,"aggregate_rtf":0.2898862048107035,"audio_seconds_per_second":3.4496294870361384,"batches":[{"batch_size":1,"device_seconds":0.7884904770851136,"audio_seconds":[2.72],"model_wall_seconds":0.8876305818557739,"first_audio_seconds":0.4509798278808594,"graph_capture_seconds":0,"live_callback_seconds":0.09893150627613068}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1163.81,"audio_s":2.03}],"sentence_ready_ms":1513.48,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":7269.400001525879},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":325.3,"llm_first_sentence_ms":348.3,"finish_reason":"stop","usage":{"prompt_tokens":1653,"completion_tokens":20,"total_tokens":1673,"cost":0.0016496073,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00166627,"upstream_inference_prompt_cost":0.00163647,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":404.05,"response_complete":true,"answer_first_audio_ms":1082.98,"tts_first_audio_ms":734.08,"first_audio_ms":1082.99,"tts_server_stages":[{"generation_ms":1000,"alignment_ms":36,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1155,"performance":{"device_seconds":0.7884904770851136,"raw_audio_seconds":2.72,"aggregate_rtf":0.2898862048107035,"audio_seconds_per_second":3.4496294870361384,"batches":[{"batch_size":1,"device_seconds":0.7884904770851136,"audio_seconds":[2.72],"model_wall_seconds":0.8876305818557739,"first_audio_seconds":0.4509798278808594,"graph_capture_seconds":0,"live_callback_seconds":0.09893150627613068}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1163.81,"audio_s":2.03}]},"at":7929.900001525879},{"type":"pipeline_metrics","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":325.3,"llm_first_sentence_ms":348.3,"finish_reason":"stop","usage":{"prompt_tokens":1653,"completion_tokens":20,"total_tokens":1673,"cost":0.0016496073,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00166627,"upstream_inference_prompt_cost":0.00163647,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":404.05,"response_complete":true,"answer_first_audio_ms":1082.98,"tts_first_audio_ms":734.08,"first_audio_ms":1082.99,"tts_server_stages":[{"generation_ms":1000,"alignment_ms":36,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1155,"performance":{"device_seconds":0.7884904770851136,"raw_audio_seconds":2.72,"aggregate_rtf":0.2898862048107035,"audio_seconds_per_second":3.4496294870361384,"batches":[{"batch_size":1,"device_seconds":0.7884904770851136,"audio_seconds":[2.72],"model_wall_seconds":0.8876305818557739,"first_audio_seconds":0.4509798278808594,"graph_capture_seconds":0,"live_callback_seconds":0.09893150627613068}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}},{"generation_ms":972,"alignment_ms":35,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1123,"performance":{"device_seconds":0.7814245929718018,"raw_audio_seconds":2.72,"aggregate_rtf":0.28728845329845654,"audio_seconds_per_second":3.4808221093422294,"batches":[{"batch_size":1,"device_seconds":0.7814245929718018,"audio_seconds":[2.72],"model_wall_seconds":0.8640152364969254,"first_audio_seconds":0.44655206298828126,"graph_capture_seconds":0,"live_callback_seconds":0.08027100563049316}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1163.81,"audio_s":2.03},{"sentence":1,"total_ms":1130.29,"audio_s":1.87}],"total_ms":2643.89,"audio_s":3.9,"sentences":2,"generation_rtf":0.6779},"at":8340.79999923706},{"type":"turn_done","barged":false,"at":11123.400001525879}],"reply":"Huch, ein lautes Horn? Hörst du das auch?","captions":[{"status":"available","caption":"The audio contains a single, distinct, and loud honking sound. The sound is short in duration and has a clear, resonant quality.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":1.84,"tokens":30,"preprocess_ms":7.89,"encoder_gpu_ms":8.53,"decoder_gpu_ms":213.54,"inference_ms":222.31,"total_ms":230.2,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":240.77,"capture_ms":1640,"event_age_ms":1880.77,"kind":"sound","eligible":true,"suppression_reason":null,"at":5690.900001525879}],"pcm_chunks":7,"page_errors":[],"passed":true},{"case":"glass_breaking","events":[{"type":"interaction","enabled":true,"helpful_enabled":false,"rules":[],"decision_model":"convaiinnovations/laya-multilingual (Stop) · dialogue model (context)","decision_interval_ms":320,"checks":0,"last_decision":null,"waiting_for":"No agreed rules yet","attempts":0,"semantic_error":null,"scope":"session-only conversational rules; acoustic verification is separate","at":1842.6999969482422},{"status":"available","caption":"The audio contains a single, sharp, percussive sound. It sounds like a single, forceful impact, possibly a drum hit or a similar sound.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":1.86,"tokens":33,"preprocess_ms":10.31,"encoder_gpu_ms":15.77,"decoder_gpu_ms":248.76,"inference_ms":264.77,"total_ms":275.08,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":287.37,"capture_ms":1660,"event_age_ms":1947.37,"kind":"sound","eligible":true,"suppression_reason":null,"at":5977.399997711182},{"type":"environment_observation","kind":"sound","status":"responding","at":6007},{"type":"latency_start","reply_id":1,"input":{"kind":"environment","sound_caption_ms":287.37},"at":6007.199996948242},{"type":"reply","text":"Oh, das war ein kräftiges Pochen!","reply_id":1,"performance":{"sentences":[{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Oh, das war ein kräftiges Pochen!","start_s":0,"end_s":2.04}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Oh, das war ein kräftiges Pochen!","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":338.06,"llm_first_sentence_ms":365.14,"sentence_ready_ms":366.22,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":6375.5},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":338.06,"llm_first_sentence_ms":365.14,"finish_reason":"stop","usage":{"prompt_tokens":1656,"completion_tokens":30,"total_tokens":1686,"cost":0.0016672986,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00168414,"upstream_inference_prompt_cost":0.00163944,"upstream_inference_completions_cost":4.47e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":391.27,"response_complete":true,"answer_first_audio_ms":1113.02,"tts_first_audio_ms":746.72,"first_audio_ms":1113.03},"at":7123.099998474121},{"type":"reply","text":"Oh, das war ein kräftiges Pochen! Hat da jemand gerade rhythmisch auf den Tisch geklopft?","reply_id":1,"performance":{"sentences":[{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Oh, das war ein kräftiges Pochen!","start_s":0,"end_s":2.35},{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Hat da jemand gerade rhythmisch auf den Tisch geklopft?","start_s":2.35,"end_s":5.41}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":1,"sentence_text":"Hat da jemand gerade rhythmisch auf den Tisch geklopft?","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":338.06,"llm_first_sentence_ms":365.14,"finish_reason":"stop","usage":{"prompt_tokens":1656,"completion_tokens":30,"total_tokens":1686,"cost":0.0016672986,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00168414,"upstream_inference_prompt_cost":0.00163944,"upstream_inference_completions_cost":4.47e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":391.27,"response_complete":true,"answer_first_audio_ms":1113.02,"tts_first_audio_ms":746.72,"first_audio_ms":1113.03,"tts_server_stages":[{"generation_ms":1058,"alignment_ms":37,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1207,"performance":{"device_seconds":0.8496933155059814,"raw_audio_seconds":2.96,"aggregate_rtf":0.28705855253580453,"audio_seconds_per_second":3.4836098460270435,"batches":[{"batch_size":1,"device_seconds":0.8496933155059814,"audio_seconds":[2.96],"model_wall_seconds":0.9447122290730476,"first_audio_seconds":0.46331185913085937,"graph_capture_seconds":0,"live_callback_seconds":0.09280992299318314}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1219.72,"audio_s":2.35}],"sentence_ready_ms":1586.9,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":7638.099998474121},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":338.06,"llm_first_sentence_ms":365.14,"finish_reason":"stop","usage":{"prompt_tokens":1656,"completion_tokens":30,"total_tokens":1686,"cost":0.0016672986,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00168414,"upstream_inference_prompt_cost":0.00163944,"upstream_inference_completions_cost":4.47e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":391.27,"response_complete":true,"answer_first_audio_ms":1113.02,"tts_first_audio_ms":746.72,"first_audio_ms":1113.03,"tts_server_stages":[{"generation_ms":1058,"alignment_ms":37,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1207,"performance":{"device_seconds":0.8496933155059814,"raw_audio_seconds":2.96,"aggregate_rtf":0.28705855253580453,"audio_seconds_per_second":3.4836098460270435,"batches":[{"batch_size":1,"device_seconds":0.8496933155059814,"audio_seconds":[2.96],"model_wall_seconds":0.9447122290730476,"first_audio_seconds":0.46331185913085937,"graph_capture_seconds":0,"live_callback_seconds":0.09280992299318314}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1219.72,"audio_s":2.35}]},"at":8348.899997711182},{"type":"pipeline_metrics","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":338.06,"llm_first_sentence_ms":365.14,"finish_reason":"stop","usage":{"prompt_tokens":1656,"completion_tokens":30,"total_tokens":1686,"cost":0.0016672986,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00168414,"upstream_inference_prompt_cost":0.00163944,"upstream_inference_completions_cost":4.47e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":391.27,"response_complete":true,"answer_first_audio_ms":1113.02,"tts_first_audio_ms":746.72,"first_audio_ms":1113.03,"tts_server_stages":[{"generation_ms":1058,"alignment_ms":37,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1207,"performance":{"device_seconds":0.8496933155059814,"raw_audio_seconds":2.96,"aggregate_rtf":0.28705855253580453,"audio_seconds_per_second":3.4836098460270435,"batches":[{"batch_size":1,"device_seconds":0.8496933155059814,"audio_seconds":[2.96],"model_wall_seconds":0.9447122290730476,"first_audio_seconds":0.46331185913085937,"graph_capture_seconds":0,"live_callback_seconds":0.09280992299318314}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}},{"generation_ms":1357,"alignment_ms":37,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1512,"performance":{"device_seconds":1.073373800754547,"raw_audio_seconds":3.6,"aggregate_rtf":0.2981593890984853,"audio_seconds_per_second":3.353910816035678,"batches":[{"batch_size":1,"device_seconds":1.073373800754547,"audio_seconds":[3.6],"model_wall_seconds":1.2414954751729965,"first_audio_seconds":0.4613570556640625,"graph_capture_seconds":0,"live_callback_seconds":0.16526293754577637}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1219.72,"audio_s":2.35},{"sentence":1,"total_ms":1525.82,"audio_s":3.6}],"total_ms":3112.85,"audio_s":5.95,"sentences":2,"generation_rtf":0.5232},"at":9130},{"type":"turn_done","barged":false,"at":13524.199996948242}],"reply":"Oh, das war ein kräftiges Pochen! Hat da jemand gerade rhythmisch auf den Tisch geklopft?","captions":[{"status":"available","caption":"The audio contains a single, sharp, percussive sound. It sounds like a single, forceful impact, possibly a drum hit or a similar sound.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":1.86,"tokens":33,"preprocess_ms":10.31,"encoder_gpu_ms":15.77,"decoder_gpu_ms":248.76,"inference_ms":264.77,"total_ms":275.08,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":287.37,"capture_ms":1660,"event_age_ms":1947.37,"kind":"sound","eligible":true,"suppression_reason":null,"at":5977.399997711182}],"pcm_chunks":9,"page_errors":[],"passed":true},{"case":"cat","events":[{"type":"interaction","enabled":true,"helpful_enabled":false,"rules":[],"decision_model":"convaiinnovations/laya-multilingual (Stop) · dialogue model (context)","decision_interval_ms":320,"checks":0,"last_decision":null,"waiting_for":"No agreed rules yet","attempts":0,"semantic_error":null,"scope":"session-only conversational rules; acoustic verification is separate","at":1729.900001525879},{"status":"available","caption":"The audio contains a single, clear meow. The meow is relatively high-pitched and has a slightly nasal quality.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":3.44,"tokens":25,"preprocess_ms":10.27,"encoder_gpu_ms":9.15,"decoder_gpu_ms":221.76,"inference_ms":231.17,"total_ms":241.45,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":254.16,"capture_ms":3240,"event_age_ms":3494.16,"kind":"sound","eligible":true,"suppression_reason":null,"at":7309.5},{"type":"environment_observation","kind":"sound","status":"responding","at":7410.100002288818},{"type":"latency_start","reply_id":1,"input":{"kind":"environment","sound_caption_ms":254.16},"at":7410.200000762939},{"type":"reply","text":"Hörst du auch dieses kleine Miauen?","reply_id":1,"performance":{"sentences":[{"emotion":"surprise","secondary":null,"blend":0,"intensity":0.55,"direction":"surprise, soft","text":"Hörst du auch dieses kleine Miauen?","start_s":0,"end_s":2.04}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Hörst du auch dieses kleine Miauen?","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":490.72,"finish_reason":"stop","usage":{"prompt_tokens":1650,"completion_tokens":16,"total_tokens":1666,"cost":0.0016407666,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00165734,"upstream_inference_prompt_cost":0.0016335,"upstream_inference_completions_cost":2.384e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":554.55,"generation_calls":1,"llm_total_ms":554.58,"response_complete":true,"sentence_ready_ms":555.66,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"surprise.soft"},"sources":[],"lookups":[],"at":7966.799999237061},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":490.72,"finish_reason":"stop","usage":{"prompt_tokens":1650,"completion_tokens":16,"total_tokens":1666,"cost":0.0016407666,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00165734,"upstream_inference_prompt_cost":0.0016335,"upstream_inference_completions_cost":2.384e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":554.55,"generation_calls":1,"llm_total_ms":554.58,"response_complete":true,"answer_first_audio_ms":1292.09,"tts_first_audio_ms":736.32,"first_audio_ms":1292.09},"at":8695.70000076294},{"type":"pipeline_metrics","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":490.72,"finish_reason":"stop","usage":{"prompt_tokens":1650,"completion_tokens":16,"total_tokens":1666,"cost":0.0016407666,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00165734,"upstream_inference_prompt_cost":0.0016335,"upstream_inference_completions_cost":2.384e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":554.55,"generation_calls":1,"llm_total_ms":554.58,"response_complete":true,"answer_first_audio_ms":1292.09,"tts_first_audio_ms":736.32,"first_audio_ms":1292.09,"tts_server_stages":[{"generation_ms":1024,"alignment_ms":39,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1176,"performance":{"device_seconds":0.8168344292640687,"raw_audio_seconds":2.8,"aggregate_rtf":0.29172658188002454,"audio_seconds_per_second":3.4278672637766685,"batches":[{"batch_size":1,"device_seconds":0.8168344292640687,"audio_seconds":[2.8],"model_wall_seconds":0.9122566357254982,"first_audio_seconds":0.45677560424804686,"graph_capture_seconds":0,"live_callback_seconds":0.09518687427043915}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1187.05,"audio_s":2.35}],"total_ms":1742.83,"audio_s":2.35,"sentences":1,"generation_rtf":0.7416},"at":9172.400001525879},{"type":"turn_done","barged":false,"at":11493.400001525879}],"reply":"Hörst du auch dieses kleine Miauen?","captions":[{"status":"available","caption":"The audio contains a single, clear meow. The meow is relatively high-pitched and has a slightly nasal quality.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":3.44,"tokens":25,"preprocess_ms":10.27,"encoder_gpu_ms":9.15,"decoder_gpu_ms":221.76,"inference_ms":231.17,"total_ms":241.45,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":254.16,"capture_ms":3240,"event_age_ms":3494.16,"kind":"sound","eligible":true,"suppression_reason":null,"at":7309.5}],"pcm_chunks":4,"page_errors":[],"passed":true},{"case":"after_question_music_bridge_False","events":[{"type":"environment_observation","kind":"requested_sound","status":"listening_after_question","at":4090.7999992370605},{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a consistent rhythmic pulse.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.4,"tokens":35,"preprocess_ms":7.9,"encoder_gpu_ms":18.98,"decoder_gpu_ms":292.79,"inference_ms":312.03,"total_ms":319.93,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":332.39,"capture_ms":2400,"event_age_ms":2732.4,"kind":"music","eligible":true,"suppression_reason":null,"requested":true,"at":6858.400001525879},{"type":"latency_start","reply_id":1,"input":{"kind":"typed"},"at":6859.299999237061},{"type":"reply","text":"Ich höre hier gerade eine schnelle, elektronische Musik mit einem treibenden Beat und einer repetitiven Synth-Melodie.","reply_id":1,"performance":{"sentences":[{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Ich höre hier gerade eine schnelle, elektronische Musik mit einem treibenden Beat und einer repetitiven Synth-Melodie.","start_s":0,"end_s":5.44}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Ich höre hier gerade eine schnelle, elektronische Musik mit einem treibenden Beat und einer repetitiven Synth-Melodie.","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":374.73,"finish_reason":"stop","usage":{"prompt_tokens":1739,"completion_tokens":33,"total_tokens":1772,"cost":0.0017530722,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00177078,"upstream_inference_prompt_cost":0.00172161,"upstream_inference_completions_cost":4.917e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":442.22,"generation_calls":1,"llm_total_ms":442.24,"response_complete":true,"sentence_ready_ms":443.88,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":7325},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":374.73,"finish_reason":"stop","usage":{"prompt_tokens":1739,"completion_tokens":33,"total_tokens":1772,"cost":0.0017530722,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00177078,"upstream_inference_prompt_cost":0.00172161,"upstream_inference_completions_cost":4.917e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":442.22,"generation_calls":1,"llm_total_ms":442.24,"response_complete":true,"answer_first_audio_ms":1207.63,"tts_first_audio_ms":763.64,"first_audio_ms":1207.64},"at":8068.5},{"type":"pipeline_metrics","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":374.73,"finish_reason":"stop","usage":{"prompt_tokens":1739,"completion_tokens":33,"total_tokens":1772,"cost":0.0017530722,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00177078,"upstream_inference_prompt_cost":0.00172161,"upstream_inference_completions_cost":4.917e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":442.22,"generation_calls":1,"llm_total_ms":442.24,"response_complete":true,"answer_first_audio_ms":1207.63,"tts_first_audio_ms":763.64,"first_audio_ms":1207.64,"tts_server_stages":[{"generation_ms":2140,"alignment_ms":39,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":2291,"performance":{"device_seconds":1.7756217284202576,"raw_audio_seconds":6.32,"aggregate_rtf":0.2809528051297876,"audio_seconds_per_second":3.5593166600989976,"batches":[{"batch_size":1,"device_seconds":1.7756217284202576,"audio_seconds":[6.32],"model_wall_seconds":2.0228959769010544,"first_audio_seconds":0.4780031433105469,"graph_capture_seconds":0,"live_callback_seconds":0.24692555516958237}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":2307.8,"audio_s":6.32}],"total_ms":2751.8,"audio_s":6.32,"sentences":1,"generation_rtf":0.4354},"at":9619.400001525879},{"type":"turn_done","barged":false,"at":14827.900001525879}],"reply":"Ich höre hier gerade eine schnelle, elektronische Musik mit einem treibenden Beat und einer repetitiven Synth-Melodie.","captions":[{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a consistent rhythmic pulse.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.4,"tokens":35,"preprocess_ms":7.9,"encoder_gpu_ms":18.98,"decoder_gpu_ms":292.79,"inference_ms":312.03,"total_ms":319.93,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":332.39,"capture_ms":2400,"event_age_ms":2732.4,"kind":"music","eligible":true,"suppression_reason":null,"requested":true,"at":6858.400001525879}],"pcm_chunks":6,"page_errors":[],"passed":true},{"case":"after_question_music_bridge_True","events":[{"type":"quick_reply_settings","enabled":true,"at":4012.400001525879},{"type":"environment_observation","kind":"requested_sound","status":"listening_after_question","at":4044.7999992370605},{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a consistent rhythmic pulse.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.4,"tokens":35,"preprocess_ms":11.04,"encoder_gpu_ms":10.56,"decoder_gpu_ms":273.96,"inference_ms":284.77,"total_ms":295.81,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":307.48,"capture_ms":2400,"event_age_ms":2707.48,"kind":"music","eligible":true,"suppression_reason":null,"requested":true,"at":6781.599998474121},{"type":"latency_start","reply_id":1,"input":{"kind":"typed"},"at":6781.799999237061},{"type":"reply","text":"Das klingt nach Musik.","reply_id":1,"performance":{"sentences":[{"emotion":"contentment","secondary":"affection","blend":0.25,"intensity":0.55,"direction":"contentment, soft, warm","text":"Das klingt nach Musik.","start_s":0,"end_s":1.36}],"source":"Contextual reaction + informed Qwen + shared TTS batch","timing":"Sentence starts anchored to preceding PCM; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Das klingt nach Musik.","sentence_role":"reaction","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"selected","tts_backend":"continuous_batch","tts_weight_instances":1,"tts_temperature":0.8,"informed_bridge_status":"replaced_by_informed_main","main_informed_by_reaction":true,"main_launch_ms":159.59,"llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","quick_reply_generation_ms":154.47,"quick_reply_model":"nemotron-bolt","quick_reply_memory_entries":0,"quick_reply_memory_person_matched":false,"quick_reply_words":4,"sentence_ready_ms":162.07,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"contentment.soft"},"sources":[],"lookups":[],"at":6948.400001525879},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"played","tts_backend":"continuous_batch","tts_weight_instances":1,"tts_temperature":0.8,"informed_bridge_status":"replaced_by_informed_main","main_informed_by_reaction":true,"main_launch_ms":159.59,"llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","quick_reply_generation_ms":154.47,"quick_reply_model":"nemotron-bolt","quick_reply_memory_entries":0,"quick_reply_memory_person_matched":false,"quick_reply_words":4,"first_audio_ms":332.03,"tts_first_audio_ms":169.86},"at":7115.599998474121},{"type":"reply","text":"Das klingt nach Musik. Oh, das klingt nach einem echten elektronischen Beat!","reply_id":1,"performance":{"sentences":[{"emotion":"contentment","secondary":"affection","blend":0.25,"intensity":0.55,"direction":"contentment, soft, warm","text":"Das klingt nach Musik.","start_s":0,"end_s":2.72},{"emotion":"amusement","secondary":null,"blend":0,"intensity":0.72,"direction":"amusement, playful","text":"Oh, das klingt nach einem echten elektronischen Beat!","start_s":2.72,"end_s":5.44}],"source":"Contextual reaction + informed Qwen + shared TTS batch","timing":"Sentence starts anchored to preceding PCM; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":1,"sentence_text":"Oh, das klingt nach einem echten elektronischen Beat!","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"played","tts_backend":"continuous_batch","tts_weight_instances":1,"tts_temperature":0.8,"informed_bridge_status":"replaced_by_informed_main","main_informed_by_reaction":true,"main_launch_ms":159.59,"llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","quick_reply_generation_ms":154.47,"quick_reply_model":"nemotron-bolt","quick_reply_memory_entries":0,"quick_reply_memory_person_matched":false,"quick_reply_words":4,"first_audio_ms":332.03,"tts_first_audio_ms":169.86,"llm_actual_provider":"Cerebras","llm_ttft_ms":371.96,"llm_first_sentence_ms":394.48,"finish_reason":"stop","usage":{"prompt_tokens":1823,"completion_tokens":69,"total_tokens":1892,"cost":0.0018885042,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00190758,"upstream_inference_prompt_cost":0.00180477,"upstream_inference_completions_cost":0.00010281},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":435.13,"response_complete":true,"tts_server_stages":[{"generation_ms":1987.55,"alignment_ms":0.01,"alignment":null,"frames":34,"live_alignment":{"enabled":false,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":0}}],"quick_reply_audio_s":2.72,"quick_reply_tts_ms":1990.5,"sentence_tts":[{"sentence":0,"role":"reaction","audio_s":2.72,"total_ms":1990.5,"prefetched":true}],"sentence_ready_ms":2152.65,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"amusement.soft"},"sources":[],"lookups":[],"at":8946.29999923706},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"played","tts_backend":"continuous_batch","tts_weight_instances":1,"tts_temperature":0.8,"informed_bridge_status":"replaced_by_informed_main","main_informed_by_reaction":true,"main_launch_ms":159.59,"llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","quick_reply_generation_ms":154.47,"quick_reply_model":"nemotron-bolt","quick_reply_memory_entries":0,"quick_reply_memory_person_matched":false,"quick_reply_words":4,"first_audio_ms":332.03,"tts_first_audio_ms":169.86,"llm_actual_provider":"Cerebras","llm_ttft_ms":371.96,"llm_first_sentence_ms":394.48,"finish_reason":"stop","usage":{"prompt_tokens":1823,"completion_tokens":69,"total_tokens":1892,"cost":0.0018885042,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00190758,"upstream_inference_prompt_cost":0.00180477,"upstream_inference_completions_cost":0.00010281},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":435.13,"response_complete":true,"tts_server_stages":[{"generation_ms":1987.55,"alignment_ms":0.01,"alignment":null,"frames":34,"live_alignment":{"enabled":false,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":0}}],"quick_reply_audio_s":2.72,"quick_reply_tts_ms":1990.5,"sentence_tts":[{"sentence":0,"role":"reaction","audio_s":2.72,"total_ms":1990.5,"prefetched":true}],"answer_first_audio_ms":2154.74,"answer_audio_ready_ms":1704.31},"at":8947.099998474121},{"type":"reply","text":"Das klingt nach Musik. Oh, das klingt nach einem echten elektronischen Beat!\nIch höre eine schnelle, treibende Synthesizer-Melodie mit einem starken Bass und einem festen Rhythmus.","reply_id":1,"performance":{"sentences":[{"emotion":"contentment","secondary":"affection","blend":0.25,"intensity":0.55,"direction":"contentment, soft, warm","text":"Das klingt nach Musik.","start_s":0,"end_s":2.72},{"emotion":"amusement","secondary":null,"blend":0,"intensity":0.72,"direction":"amusement, playful","text":"Oh, das klingt nach einem echten elektronischen Beat!","start_s":2.72,"end_s":6.59},{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Ich höre eine schnelle, treibende Synthesizer-Melodie mit einem starken Bass und einem festen Rhythmus.","start_s":6.59,"end_s":11.35}],"source":"Contextual reaction + informed Qwen + shared TTS batch","timing":"Sentence starts anchored to preceding PCM; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":2,"sentence_text":"Ich höre eine schnelle, treibende Synthesizer-Melodie mit einem starken Bass und einem festen Rhythmus.","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"played","tts_backend":"continuous_batch","tts_weight_instances":1,"tts_temperature":0.8,"informed_bridge_status":"replaced_by_informed_main","main_informed_by_reaction":true,"main_launch_ms":159.59,"llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","quick_reply_generation_ms":154.47,"quick_reply_model":"nemotron-bolt","quick_reply_memory_entries":0,"quick_reply_memory_person_matched":false,"quick_reply_words":4,"first_audio_ms":332.03,"tts_first_audio_ms":169.86,"llm_actual_provider":"Cerebras","llm_ttft_ms":371.96,"llm_first_sentence_ms":394.48,"finish_reason":"stop","usage":{"prompt_tokens":1823,"completion_tokens":69,"total_tokens":1892,"cost":0.0018885042,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00190758,"upstream_inference_prompt_cost":0.00180477,"upstream_inference_completions_cost":0.00010281},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":435.13,"response_complete":true,"tts_server_stages":[{"generation_ms":1987.55,"alignment_ms":0.01,"alignment":null,"frames":34,"live_alignment":{"enabled":false,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":0}},{"generation_ms":3657.04,"alignment_ms":37.15,"alignment":{"applied":true,"lead_s":0.24,"cut_s":3.72,"removed_s":0.69,"words":8,"note":"","spans":[["Oh",0.24,0.4,1],["das",0.96,1.2,1],["klingt",1.2,1.44,1],["nach",1.44,1.6,1],["einem",1.6,1.84,1],["echten",1.92,2.32,1],["elektronischen",2.32,3.2,1],["Beat",3.2,3.6,1]],"enabled":true,"available":true,"raw_seconds":4.56,"seconds":3.87,"lead_ramp_s":0.1,"tail_ramp_s":0.15,"score_kind":"structural plausibility, not model confidence"},"frames":57,"live_alignment":{"enabled":true,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":3.87}}],"quick_reply_audio_s":2.72,"quick_reply_tts_ms":1990.5,"sentence_tts":[{"sentence":0,"role":"reaction","audio_s":2.72,"total_ms":1990.5,"prefetched":true},{"sentence":1,"role":"answer","audio_s":3.87,"total_ms":3675.14,"prefetched":true}],"answer_first_audio_ms":2154.74,"answer_audio_ready_ms":1704.31,"sentence_ready_ms":4231.38,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":11027.5},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"played","tts_backend":"continuous_batch","tts_weight_instances":1,"tts_temperature":0.8,"informed_bridge_status":"replaced_by_informed_main","main_informed_by_reaction":true,"main_launch_ms":159.59,"llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","quick_reply_generation_ms":154.47,"quick_reply_model":"nemotron-bolt","quick_reply_memory_entries":0,"quick_reply_memory_person_matched":false,"quick_reply_words":4,"first_audio_ms":332.03,"tts_first_audio_ms":169.86,"llm_actual_provider":"Cerebras","llm_ttft_ms":371.96,"llm_first_sentence_ms":394.48,"finish_reason":"stop","usage":{"prompt_tokens":1823,"completion_tokens":69,"total_tokens":1892,"cost":0.0018885042,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00190758,"upstream_inference_prompt_cost":0.00180477,"upstream_inference_completions_cost":0.00010281},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":435.13,"response_complete":true,"tts_server_stages":[{"generation_ms":1987.55,"alignment_ms":0.01,"alignment":null,"frames":34,"live_alignment":{"enabled":false,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":0}},{"generation_ms":3657.04,"alignment_ms":37.15,"alignment":{"applied":true,"lead_s":0.24,"cut_s":3.72,"removed_s":0.69,"words":8,"note":"","spans":[["Oh",0.24,0.4,1],["das",0.96,1.2,1],["klingt",1.2,1.44,1],["nach",1.44,1.6,1],["einem",1.6,1.84,1],["echten",1.92,2.32,1],["elektronischen",2.32,3.2,1],["Beat",3.2,3.6,1]],"enabled":true,"available":true,"raw_seconds":4.56,"seconds":3.87,"lead_ramp_s":0.1,"tail_ramp_s":0.15,"score_kind":"structural plausibility, not model confidence"},"frames":57,"live_alignment":{"enabled":true,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":3.87}}],"quick_reply_audio_s":2.72,"quick_reply_tts_ms":1990.5,"sentence_tts":[{"sentence":0,"role":"reaction","audio_s":2.72,"total_ms":1990.5,"prefetched":true},{"sentence":1,"role":"answer","audio_s":3.87,"total_ms":3675.14,"prefetched":true}],"answer_first_audio_ms":2154.74,"answer_audio_ready_ms":1704.31},"at":11027.900001525879},{"type":"reply","text":"Das klingt nach Musik. Oh, das klingt nach einem echten elektronischen Beat!\nIch höre eine schnelle, treibende Synthesizer-Melodie mit einem starken Bass und einem festen Rhythmus.\nKlingt nach guter Musik zum Tanzen oder Konzentrieren!","reply_id":1,"performance":{"sentences":[{"emotion":"contentment","secondary":"affection","blend":0.25,"intensity":0.55,"direction":"contentment, soft, warm","text":"Das klingt nach Musik.","start_s":0,"end_s":2.72},{"emotion":"amusement","secondary":null,"blend":0,"intensity":0.72,"direction":"amusement, playful","text":"Oh, das klingt nach einem echten elektronischen Beat!","start_s":2.72,"end_s":6.59},{"emotion":"interest","secondary":null,"blend":0,"intensity":0.55,"direction":"interest, soft","text":"Ich höre eine schnelle, treibende Synthesizer-Melodie mit einem starken Bass und einem festen Rhythmus.","start_s":6.59,"end_s":12.62},{"emotion":"affection","secondary":null,"blend":0,"intensity":0.72,"direction":"curiosity, warm","text":"Klingt nach guter Musik zum Tanzen oder Konzentrieren!","start_s":12.62,"end_s":15.34}],"source":"Contextual reaction + informed Qwen + shared TTS batch","timing":"Sentence starts anchored to preceding PCM; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":3,"sentence_text":"Klingt nach guter Musik zum Tanzen oder Konzentrieren!","sentence_role":"answer","speaker_id":null,"speaker_name":null,"speaker_update":null,"timing":{"quick_reply_status":"played","tts_backend":"continuous_batch","tts_weight_instances":1,"tts_temperature":0.8,"informed_bridge_status":"replaced_by_informed_main","main_informed_by_reaction":true,"main_launch_ms":159.59,"llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","quick_reply_generation_ms":154.47,"quick_reply_model":"nemotron-bolt","quick_reply_memory_entries":0,"quick_reply_memory_person_matched":false,"quick_reply_words":4,"first_audio_ms":332.03,"tts_first_audio_ms":169.86,"llm_actual_provider":"Cerebras","llm_ttft_ms":371.96,"llm_first_sentence_ms":394.48,"finish_reason":"stop","usage":{"prompt_tokens":1823,"completion_tokens":69,"total_tokens":1892,"cost":0.0018885042,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00190758,"upstream_inference_prompt_cost":0.00180477,"upstream_inference_completions_cost":0.00010281},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":435.13,"response_complete":true,"tts_server_stages":[{"generation_ms":1987.55,"alignment_ms":0.01,"alignment":null,"frames":34,"live_alignment":{"enabled":false,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":0}},{"generation_ms":3657.04,"alignment_ms":37.15,"alignment":{"applied":true,"lead_s":0.24,"cut_s":3.72,"removed_s":0.69,"words":8,"note":"","spans":[["Oh",0.24,0.4,1],["das",0.96,1.2,1],["klingt",1.2,1.44,1],["nach",1.44,1.6,1],["einem",1.6,1.84,1],["echten",1.92,2.32,1],["elektronischen",2.32,3.2,1],["Beat",3.2,3.6,1]],"enabled":true,"available":true,"raw_seconds":4.56,"seconds":3.87,"lead_ramp_s":0.1,"tail_ramp_s":0.15,"score_kind":"structural plausibility, not model confidence"},"frames":57,"live_alignment":{"enabled":true,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":3.87}},{"generation_ms":4950.15,"alignment_ms":41.31,"alignment":{"applied":true,"lead_s":0.32,"cut_s":5.88,"removed_s":0.29,"words":15,"note":"","spans":[["Ich",0.32,0.48,1],["höre",0.48,0.56,1],["eine",0.56,0.8,1],["schnelle",0.8,1.28,1],["treibende",1.44,1.92,1],["Synthesizer",1.92,2.48,1],["Melodie",2.48,2.96,1],["mit",2.96,2.96,1],["einem",3.52,3.68,1],["starken",3.68,4.08,1],["Bass",4.08,4.48,1],["und",4.64,4.8,1],["einem",4.8,4.96,1],["festen",4.96,5.28,1],["Rhythmus",5.28,5.76,1]],"enabled":true,"available":true,"raw_seconds":6.32,"seconds":6.03,"lead_ramp_s":0.1,"tail_ramp_s":0.15,"score_kind":"structural plausibility, not model confidence"},"frames":79,"live_alignment":{"enabled":true,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":6.03}}],"quick_reply_audio_s":2.72,"quick_reply_tts_ms":1990.5,"sentence_tts":[{"sentence":0,"role":"reaction","audio_s":2.72,"total_ms":1990.5,"prefetched":true},{"sentence":1,"role":"answer","audio_s":3.87,"total_ms":3675.14,"prefetched":true},{"sentence":2,"role":"answer","audio_s":6.03,"total_ms":6534.81,"prefetched":true}],"answer_first_audio_ms":2154.74,"answer_audio_ready_ms":1704.31,"sentence_ready_ms":7106.09,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"delivery.warm.soft"},"sources":[],"lookups":[],"at":13902.400001525879},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"played","tts_backend":"continuous_batch","tts_weight_instances":1,"tts_temperature":0.8,"informed_bridge_status":"replaced_by_informed_main","main_informed_by_reaction":true,"main_launch_ms":159.59,"llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","quick_reply_generation_ms":154.47,"quick_reply_model":"nemotron-bolt","quick_reply_memory_entries":0,"quick_reply_memory_person_matched":false,"quick_reply_words":4,"first_audio_ms":332.03,"tts_first_audio_ms":169.86,"llm_actual_provider":"Cerebras","llm_ttft_ms":371.96,"llm_first_sentence_ms":394.48,"finish_reason":"stop","usage":{"prompt_tokens":1823,"completion_tokens":69,"total_tokens":1892,"cost":0.0018885042,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00190758,"upstream_inference_prompt_cost":0.00180477,"upstream_inference_completions_cost":0.00010281},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":435.13,"response_complete":true,"tts_server_stages":[{"generation_ms":1987.55,"alignment_ms":0.01,"alignment":null,"frames":34,"live_alignment":{"enabled":false,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":0}},{"generation_ms":3657.04,"alignment_ms":37.15,"alignment":{"applied":true,"lead_s":0.24,"cut_s":3.72,"removed_s":0.69,"words":8,"note":"","spans":[["Oh",0.24,0.4,1],["das",0.96,1.2,1],["klingt",1.2,1.44,1],["nach",1.44,1.6,1],["einem",1.6,1.84,1],["echten",1.92,2.32,1],["elektronischen",2.32,3.2,1],["Beat",3.2,3.6,1]],"enabled":true,"available":true,"raw_seconds":4.56,"seconds":3.87,"lead_ramp_s":0.1,"tail_ramp_s":0.15,"score_kind":"structural plausibility, not model confidence"},"frames":57,"live_alignment":{"enabled":true,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":3.87}},{"generation_ms":4950.15,"alignment_ms":41.31,"alignment":{"applied":true,"lead_s":0.32,"cut_s":5.88,"removed_s":0.29,"words":15,"note":"","spans":[["Ich",0.32,0.48,1],["höre",0.48,0.56,1],["eine",0.56,0.8,1],["schnelle",0.8,1.28,1],["treibende",1.44,1.92,1],["Synthesizer",1.92,2.48,1],["Melodie",2.48,2.96,1],["mit",2.96,2.96,1],["einem",3.52,3.68,1],["starken",3.68,4.08,1],["Bass",4.08,4.48,1],["und",4.64,4.8,1],["einem",4.8,4.96,1],["festen",4.96,5.28,1],["Rhythmus",5.28,5.76,1]],"enabled":true,"available":true,"raw_seconds":6.32,"seconds":6.03,"lead_ramp_s":0.1,"tail_ramp_s":0.15,"score_kind":"structural plausibility, not model confidence"},"frames":79,"live_alignment":{"enabled":true,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":6.03}}],"quick_reply_audio_s":2.72,"quick_reply_tts_ms":1990.5,"sentence_tts":[{"sentence":0,"role":"reaction","audio_s":2.72,"total_ms":1990.5,"prefetched":true},{"sentence":1,"role":"answer","audio_s":3.87,"total_ms":3675.14,"prefetched":true},{"sentence":2,"role":"answer","audio_s":6.03,"total_ms":6534.81,"prefetched":true}],"answer_first_audio_ms":2154.74,"answer_audio_ready_ms":1704.31},"at":13903.099998474121},{"type":"pipeline_metrics","reply_id":1,"metrics":{"quick_reply_status":"played","tts_backend":"continuous_batch","tts_weight_instances":1,"tts_temperature":0.8,"informed_bridge_status":"replaced_by_informed_main","main_informed_by_reaction":true,"main_launch_ms":159.59,"llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","quick_reply_generation_ms":154.47,"quick_reply_model":"nemotron-bolt","quick_reply_memory_entries":0,"quick_reply_memory_person_matched":false,"quick_reply_words":4,"first_audio_ms":332.03,"tts_first_audio_ms":169.86,"llm_actual_provider":"Cerebras","llm_ttft_ms":371.96,"llm_first_sentence_ms":394.48,"finish_reason":"stop","usage":{"prompt_tokens":1823,"completion_tokens":69,"total_tokens":1892,"cost":0.0018885042,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00190758,"upstream_inference_prompt_cost":0.00180477,"upstream_inference_completions_cost":0.00010281},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":435.13,"response_complete":true,"tts_server_stages":[{"generation_ms":1987.55,"alignment_ms":0.01,"alignment":null,"frames":34,"live_alignment":{"enabled":false,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":0}},{"generation_ms":3657.04,"alignment_ms":37.15,"alignment":{"applied":true,"lead_s":0.24,"cut_s":3.72,"removed_s":0.69,"words":8,"note":"","spans":[["Oh",0.24,0.4,1],["das",0.96,1.2,1],["klingt",1.2,1.44,1],["nach",1.44,1.6,1],["einem",1.6,1.84,1],["echten",1.92,2.32,1],["elektronischen",2.32,3.2,1],["Beat",3.2,3.6,1]],"enabled":true,"available":true,"raw_seconds":4.56,"seconds":3.87,"lead_ramp_s":0.1,"tail_ramp_s":0.15,"score_kind":"structural plausibility, not model confidence"},"frames":57,"live_alignment":{"enabled":true,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":3.87}},{"generation_ms":4950.15,"alignment_ms":41.31,"alignment":{"applied":true,"lead_s":0.32,"cut_s":5.88,"removed_s":0.29,"words":15,"note":"","spans":[["Ich",0.32,0.48,1],["höre",0.48,0.56,1],["eine",0.56,0.8,1],["schnelle",0.8,1.28,1],["treibende",1.44,1.92,1],["Synthesizer",1.92,2.48,1],["Melodie",2.48,2.96,1],["mit",2.96,2.96,1],["einem",3.52,3.68,1],["starken",3.68,4.08,1],["Bass",4.08,4.48,1],["und",4.64,4.8,1],["einem",4.8,4.96,1],["festen",4.96,5.28,1],["Rhythmus",5.28,5.76,1]],"enabled":true,"available":true,"raw_seconds":6.32,"seconds":6.03,"lead_ramp_s":0.1,"tail_ramp_s":0.15,"score_kind":"structural plausibility, not model confidence"},"frames":79,"live_alignment":{"enabled":true,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":6.03}},{"generation_ms":3010.28,"alignment_ms":38.97,"alignment":{"applied":true,"lead_s":0.32,"cut_s":3.16,"removed_s":0.37,"words":8,"note":"","spans":[["Klingt",0.32,0.64,1],["nach",0.64,0.8,1],["guter",0.8,1.04,1],["Musik",1.04,1.52,1],["zum",1.52,1.68,1],["Tanzen",1.68,2.16,1],["oder",2.16,2.32,1],["Konzentrieren",2.32,3.04,1]],"enabled":true,"available":true,"raw_seconds":3.68,"seconds":3.31,"lead_ramp_s":0.1,"tail_ramp_s":0.15,"score_kind":"structural plausibility, not model confidence"},"frames":46,"live_alignment":{"enabled":true,"backend":"Parakeet v3 word timestamps + final Qwen forced alignment","lead_s":null,"cut_s":null,"stopped":false,"lookahead_seconds":1,"emitted_seconds":3.31}}],"quick_reply_audio_s":2.72,"quick_reply_tts_ms":1990.5,"sentence_tts":[{"sentence":0,"role":"reaction","audio_s":2.72,"total_ms":1990.5,"prefetched":true},{"sentence":1,"role":"answer","audio_s":3.87,"total_ms":3675.14,"prefetched":true},{"sentence":2,"role":"answer","audio_s":6.03,"total_ms":6534.81,"prefetched":true},{"sentence":3,"role":"answer","audio_s":3.31,"total_ms":6647.49,"prefetched":true}],"answer_first_audio_ms":2154.74,"answer_audio_ready_ms":1704.31,"total_ms":7242.65,"audio_s":15.93,"sentences":4,"continuous_tts":{"scheduler":"continuous CFG batch","weight_instances":1,"batch_histogram":{"1":10,"2":105},"prefill_ms":[46.15,41.13,42.47,40.54],"frame_ms":[9.03,113.36,18.47,17.82,17.9,17.52,113.91,18.4,17.22,17.43,19.38,210.51,20.86,19.69,19.73,19.69,205.98,20.31,19.94,20.39,20.27,205.79,20.38,19.84,20.06,19.78,206.66,21.13,20.27,19.96,19.94,207.19,20.75,19.96,115.15,20.4,212.88,20.85,20.47,20.33,20.33,205.02,20.57,20.68,20.4,19.94,209.33,20.8,20.03,19.98,20.11,215.99,21.36,20.04,19.96,19.92,205.24,20.73,20.05,20.75,20.39,207.14,21.03,20.22,20.17,20.18,205.03,59.01,20.24,116.33,20.83,114.48,21.53,20.14,112.01,20.76,113.19,20.79,20.3,114.38,21.37,113.28,20.88,20.28,114.09,20.85,113.33,21.5,20.74,113.42,20.92,113.71,23.99,20.36,118.8,22.02,114.83,21.16,20.47,113.93,21.19,114.3,21.15,21.21,113.62,21.13,113.61,21.25,20.44,125.1,21.54,119.14,21.74,20.5,297.17],"max_active":2,"jobs":4,"cancelled_jobs":0,"total_ms":7226.89,"gpu_allocated_mb":11404.9},"generation_rtf":0.4547},"at":14031.29999923706},{"type":"turn_done","barged":false,"at":23392.70000076294}],"reply":"Das klingt nach Musik. Oh, das klingt nach einem echten elektronischen Beat!\nIch höre eine schnelle, treibende Synthesizer-Melodie mit einem starken Bass und einem festen Rhythmus.\nKlingt nach guter Musik zum Tanzen oder Konzentrieren!","captions":[{"status":"available","caption":"The audio features a fast-paced electronic track with a prominent, repetitive synth melody. A driving beat is present, with a strong bassline and a consistent rhythmic pulse.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.4,"tokens":35,"preprocess_ms":11.04,"encoder_gpu_ms":10.56,"decoder_gpu_ms":273.96,"inference_ms":284.77,"total_ms":295.81,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":307.48,"capture_ms":2400,"event_age_ms":2707.48,"kind":"music","eligible":true,"suppression_reason":null,"requested":true,"at":6781.599998474121}],"pcm_chunks":45,"page_errors":[],"passed":true},{"case":"christoph-intro","events":[{"type":"asr_preview","text":"","status":"sentence_stream","source":"local_sentence_asr","at":4607.200000762939},{"type":"asr_preview","text":"Hey Rao.","status":"sentence_stream","source":"local_sentence_asr","at":5094.700000762939},{"type":"asr_preview","text":"Hey Rob, ich","status":"sentence_stream","source":"local_sentence_asr","at":5674.799999237061},{"type":"asr_preview","text":"Hey Robbie, bin.","status":"sentence_stream","source":"native_asr","at":6227.600002288818},{"status":"available","caption":"The audio contains a single male voice speaking in German. The speaker's tone is conversational and relatively neutral, with a moderate pace.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.12,"tokens":28,"preprocess_ms":10.92,"encoder_gpu_ms":8.33,"decoder_gpu_ms":217.98,"inference_ms":226.56,"total_ms":237.47,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":250.15,"capture_ms":1920,"event_age_ms":2170.16,"kind":"voice","eligible":false,"suppression_reason":"speech_or_quiet","at":6657.5},{"type":"asr_preview","text":"Hey Rabbi, ich bin Christoph.","status":"sentence_stream","source":"local_sentence_asr","at":6988},{"type":"asr_finalizing","model":"parakeet_v3","label":"Transcribing the completed speech segment…","at":7045.400001525879},{"type":"asr_preview","text":"Hey Robbie, ich bin Christoph.","status":"sentence_stream","source":"local_sentence_asr","at":7110.200000762939},{"type":"asr_sentence","text":"Hey Robbie, ich bin Christoph.","audio_s":5.12,"queued":false,"asr":{"text":"Hey Robbie, ich bin Christoph.","model":"parakeet_v3","chunks":1,"transcribe_ms":50.42,"speech_pause_ms":700,"observed_quiet_ms":24},"reason":"neural speech pause","speech_pause_ms":700,"observed_quiet_ms":24,"at":7113.100002288818},{"type":"identity","enabled":false,"mode":"open-sentences","sentence_mode":true,"open_enrollment":true,"speakers":[{"id":"speaker-1","name":"Christoph","named":true,"reference_ready":true,"voice_variants":1,"admitted":true}],"users":["Christoph"],"threshold":0.38,"conversation_focus":{"enabled":false,"threshold":0.28,"timeout_s":180,"speaker_id":null,"name":null,"primary_id":"speaker-1","participants":[{"id":"speaker-1","name":"Christoph"}]},"handover_pending":false,"storage":"Session RAM","prediction":{"name":"Christoph","similarity":-1},"at":7113.299999237061},{"type":"voice_caption","status":"available","caption":"Possibly slightly smooth, slightly steady, very liberation","uncertainty":"Uncalibrated acoustic estimates; noise, other speakers and acting may be misleading.","voice_impression_only":false,"window_s":2.12,"inference_ms":26.73,"model":"laion/humaneness-ears-base-medium","model_revision":"818506970d93c809a3295a0a02c40dee8ff2bfc3","rules_revision":"7ebd4a9b7da96df989af8fc6ad0ce34f4e8aa3ed","device":"cuda:1","precision":"bfloat16","selected":{"voicenet":[{"dim":"DFLU","name":"Disfluency","value":0.397,"z":-0.9,"direction":"below","intensity":"Somewhat","always_on":false,"phrase":"somewhat smooth and flawless","tag":"slightly smooth","reliability":1,"rank_score":0.898},{"dim":"VOLT","name":"Volatility","value":1.314,"z":-0.76,"direction":"below","intensity":"Somewhat","always_on":false,"phrase":"somewhat emotionally steady and static","tag":"slightly steady","reliability":1,"rank_score":0.763},{"dim":"TEMP","name":"Tempo","value":2.828,"z":-0.25,"direction":"below","intensity":"About-average","always_on":true,"phrase":"moderate in tempo","tag":null,"reliability":1,"rank_score":0.254}],"emotions":[{"emotion":"Relief","value":1.095,"z":1.85,"direction":"above","intensity":"Very","synonym":"liberation","phrase":"very carrying liberation","tag":"very liberation"}]},"normalization":"speech reference z-scores; not probabilities","scope":"this turn","request_ms":38.89,"added_wait_ms":0,"at":7113.799999237061},{"type":"endpoint","transcript":"Hey Robbie, ich bin Christoph.","speaker_id":"speaker-1","asr":{"text":"Hey Robbie, ich bin Christoph.","model":"parakeet_v3","chunks":1,"transcribe_ms":50.42,"speech_pause_ms":700,"observed_quiet_ms":24},"speaker_name":"Christoph","by":"asr_complete_sentence","timing":{"speaker_assignment_ms":1.012,"speaker_embedding_wait_ms":0.003,"speaker_lookup_ms":1.009,"speaker_prefetch_used":true,"speaker_prefetch_compute_ms":75.98,"sentence_queue_ms":1.27,"sentence_audio_s":5.12},"at":7113.799999237061},{"type":"latency_start","reply_id":1,"input":{"kind":"voice","asr":{"text":"Hey Robbie, ich bin Christoph.","model":"parakeet_v3","chunks":1,"transcribe_ms":50.42,"speech_pause_ms":700,"observed_quiet_ms":24},"voice_caption_ms":38.89,"voice_caption_status":"available","speaker_assignment_ms":1.012,"speaker_prefetch_compute_ms":75.98,"speaker_prefetch_used":true,"sentence_queue_ms":1.26},"at":7114.5},{"type":"reply","text":"Hallo Christoph!","reply_id":1,"performance":{"sentences":[{"emotion":"affection","secondary":null,"blend":0,"intensity":0.72,"direction":"happy, warm","text":"Hallo Christoph!","start_s":0,"end_s":0.68}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Hallo Christoph!","sentence_role":"answer","speaker_id":"speaker-1","speaker_name":"Christoph","speaker_update":{"speaker_id":"speaker-1","display_name":"Christoph","is_self_introduction":true,"source":"validated_acoustic_transcript"},"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":401.56,"llm_first_sentence_ms":433.64,"sentence_ready_ms":434.91,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"elation.soft"},"sources":[],"lookups":[],"at":7548.100002288818},{"type":"asr_preview","text":"","status":"sentence_stream","source":"local_sentence_asr","at":7579.900001525879},{"type":"asr_preview","text":"","status":"sentence_stream","source":"local_sentence_asr","at":8074.100002288818},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":401.56,"llm_first_sentence_ms":433.64,"finish_reason":"stop","usage":{"prompt_tokens":1708,"completion_tokens":44,"total_tokens":1752,"cost":0.0017389152,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00175648,"upstream_inference_prompt_cost":0.00169092,"upstream_inference_completions_cost":6.556e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":492.3,"response_complete":true,"answer_first_audio_ms":1176.31,"tts_first_audio_ms":741.24,"first_audio_ms":1176.32},"at":8288.400001525879},{"type":"asr_preview","text":"schön dich kennenzulernen.","status":"sentence_stream","source":"local_sentence_asr","at":8574.29999923706},{"type":"reply","text":"Hallo Christoph! Schön, dass du da bist.","reply_id":1,"performance":{"sentences":[{"emotion":"affection","secondary":null,"blend":0,"intensity":0.72,"direction":"happy, warm","text":"Hallo Christoph!","start_s":0,"end_s":2.19},{"emotion":"affection","secondary":null,"blend":0,"intensity":0.72,"direction":"happy, warm","text":"Schön, dass du da bist.","start_s":2.19,"end_s":3.8899999999999997}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":1,"sentence_text":"Schön, dass du da bist.","sentence_role":"answer","speaker_id":"speaker-1","speaker_name":"Christoph","speaker_update":{"speaker_id":"speaker-1","display_name":"Christoph","is_self_introduction":true,"source":"validated_acoustic_transcript"},"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":401.56,"llm_first_sentence_ms":433.64,"finish_reason":"stop","usage":{"prompt_tokens":1708,"completion_tokens":44,"total_tokens":1752,"cost":0.0017389152,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00175648,"upstream_inference_prompt_cost":0.00169092,"upstream_inference_completions_cost":6.556e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":492.3,"response_complete":true,"answer_first_audio_ms":1176.31,"tts_first_audio_ms":741.24,"first_audio_ms":1176.32,"tts_server_stages":[{"generation_ms":1075,"alignment_ms":44,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1237,"performance":{"device_seconds":0.8726874685287476,"raw_audio_seconds":3.04,"aggregate_rtf":0.2870682462265617,"audio_seconds_per_second":3.483492211851164,"batches":[{"batch_size":1,"device_seconds":0.8726874685287476,"audio_seconds":[3.04],"model_wall_seconds":0.9614429473876953,"first_audio_seconds":0.4565145263671875,"graph_capture_seconds":0,"live_callback_seconds":0.08843447268009186}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1248.24,"audio_s":2.19}],"sentence_ready_ms":1684.29,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"elation.soft"},"sources":[],"lookups":[],"at":8799.20000076294},{"type":"asr_finalizing","model":"parakeet_v3","label":"Transcribing the completed speech segment…","at":8924.20000076294},{"type":"asr_preview","text":"Schön, dich kennenzulernen.","status":"sentence_stream","source":"local_sentence_asr","at":8996.400001525879},{"type":"asr_sentence","text":"Schön, dich kennenzulernen.","audio_s":7.04,"queued":true,"asr":{"text":"Schön, dich kennenzulernen.","model":"parakeet_v3","chunks":1,"transcribe_ms":47.12,"speech_pause_ms":700,"observed_quiet_ms":598},"reason":"neural speech pause","speech_pause_ms":700,"observed_quiet_ms":598,"at":9000.20000076294},{"type":"identity","enabled":false,"mode":"open-sentences","sentence_mode":true,"open_enrollment":true,"speakers":[{"id":"speaker-1","name":"Christoph","named":true,"reference_ready":true,"voice_variants":1,"admitted":true}],"users":["Christoph"],"threshold":0.38,"conversation_focus":{"enabled":false,"threshold":0.28,"timeout_s":180,"speaker_id":null,"name":null,"primary_id":"speaker-1","participants":[{"id":"speaker-1","name":"Christoph"}]},"handover_pending":false,"storage":"Session RAM","prediction":{"name":"Christoph","similarity":0.3914},"at":9000.5},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":401.56,"llm_first_sentence_ms":433.64,"finish_reason":"stop","usage":{"prompt_tokens":1708,"completion_tokens":44,"total_tokens":1752,"cost":0.0017389152,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00175648,"upstream_inference_prompt_cost":0.00169092,"upstream_inference_completions_cost":6.556e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":492.3,"response_complete":true,"answer_first_audio_ms":1176.31,"tts_first_audio_ms":741.24,"first_audio_ms":1176.32,"tts_server_stages":[{"generation_ms":1075,"alignment_ms":44,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1237,"performance":{"device_seconds":0.8726874685287476,"raw_audio_seconds":3.04,"aggregate_rtf":0.2870682462265617,"audio_seconds_per_second":3.483492211851164,"batches":[{"batch_size":1,"device_seconds":0.8726874685287476,"audio_seconds":[3.04],"model_wall_seconds":0.9614429473876953,"first_audio_seconds":0.4565145263671875,"graph_capture_seconds":0,"live_callback_seconds":0.08843447268009186}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1248.24,"audio_s":2.19}]},"at":9557.600002288818},{"type":"reply","text":"Hallo Christoph! Schön, dass du da bist. Ich bin Robby, übrigens, nicht Robbie, aber das ist ein Detail, das wir übersehen können.","reply_id":1,"performance":{"sentences":[{"emotion":"affection","secondary":null,"blend":0,"intensity":0.72,"direction":"happy, warm","text":"Hallo Christoph!","start_s":0,"end_s":2.19},{"emotion":"affection","secondary":null,"blend":0,"intensity":0.72,"direction":"happy, warm","text":"Schön, dass du da bist.","start_s":2.19,"end_s":4.86},{"emotion":"amusement","secondary":null,"blend":0,"intensity":0.72,"direction":"amused, playful","text":"Ich bin Robby, übrigens, nicht Robbie, aber das ist ein Detail, das wir übersehen können.","start_s":4.86,"end_s":9.96}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":2,"sentence_text":"Ich bin Robby, übrigens, nicht Robbie, aber das ist ein Detail, das wir übersehen können.","sentence_role":"answer","speaker_id":"speaker-1","speaker_name":"Christoph","speaker_update":{"speaker_id":"speaker-1","display_name":"Christoph","is_self_introduction":true,"source":"validated_acoustic_transcript"},"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":401.56,"llm_first_sentence_ms":433.64,"finish_reason":"stop","usage":{"prompt_tokens":1708,"completion_tokens":44,"total_tokens":1752,"cost":0.0017389152,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00175648,"upstream_inference_prompt_cost":0.00169092,"upstream_inference_completions_cost":6.556e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":492.3,"response_complete":true,"answer_first_audio_ms":1176.31,"tts_first_audio_ms":741.24,"first_audio_ms":1176.32,"tts_server_stages":[{"generation_ms":1075,"alignment_ms":44,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1237,"performance":{"device_seconds":0.8726874685287476,"raw_audio_seconds":3.04,"aggregate_rtf":0.2870682462265617,"audio_seconds_per_second":3.483492211851164,"batches":[{"batch_size":1,"device_seconds":0.8726874685287476,"audio_seconds":[3.04],"model_wall_seconds":0.9614429473876953,"first_audio_seconds":0.4565145263671875,"graph_capture_seconds":0,"live_callback_seconds":0.08843447268009186}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}},{"generation_ms":1029,"alignment_ms":39,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1179,"performance":{"device_seconds":0.8219381828308105,"raw_audio_seconds":2.8,"aggregate_rtf":0.29354935101100377,"audio_seconds_per_second":3.4065822205224863,"batches":[{"batch_size":1,"device_seconds":0.8219381828308105,"audio_seconds":[2.8],"model_wall_seconds":0.906829223036766,"first_audio_seconds":0.4643337097167969,"graph_capture_seconds":0,"live_callback_seconds":0.08464871346950531}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1248.24,"audio_s":2.19},{"sentence":1,"total_ms":1186.93,"audio_s":2.67}],"sentence_ready_ms":2872.34,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"delivery.playful.soft"},"sources":[],"lookups":[],"at":9988.5},{"type":"latency_update","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":401.56,"llm_first_sentence_ms":433.64,"finish_reason":"stop","usage":{"prompt_tokens":1708,"completion_tokens":44,"total_tokens":1752,"cost":0.0017389152,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00175648,"upstream_inference_prompt_cost":0.00169092,"upstream_inference_completions_cost":6.556e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":492.3,"response_complete":true,"answer_first_audio_ms":1176.31,"tts_first_audio_ms":741.24,"first_audio_ms":1176.32,"tts_server_stages":[{"generation_ms":1075,"alignment_ms":44,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1237,"performance":{"device_seconds":0.8726874685287476,"raw_audio_seconds":3.04,"aggregate_rtf":0.2870682462265617,"audio_seconds_per_second":3.483492211851164,"batches":[{"batch_size":1,"device_seconds":0.8726874685287476,"audio_seconds":[3.04],"model_wall_seconds":0.9614429473876953,"first_audio_seconds":0.4565145263671875,"graph_capture_seconds":0,"live_callback_seconds":0.08843447268009186}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}},{"generation_ms":1029,"alignment_ms":39,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1179,"performance":{"device_seconds":0.8219381828308105,"raw_audio_seconds":2.8,"aggregate_rtf":0.29354935101100377,"audio_seconds_per_second":3.4065822205224863,"batches":[{"batch_size":1,"device_seconds":0.8219381828308105,"audio_seconds":[2.8],"model_wall_seconds":0.906829223036766,"first_audio_seconds":0.4643337097167969,"graph_capture_seconds":0,"live_callback_seconds":0.08464871346950531}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1248.24,"audio_s":2.19},{"sentence":1,"total_ms":1186.93,"audio_s":2.67}]},"at":10740.79999923706},{"type":"pipeline_metrics","reply_id":1,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":401.56,"llm_first_sentence_ms":433.64,"finish_reason":"stop","usage":{"prompt_tokens":1708,"completion_tokens":44,"total_tokens":1752,"cost":0.0017389152,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00175648,"upstream_inference_prompt_cost":0.00169092,"upstream_inference_completions_cost":6.556e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":492.3,"response_complete":true,"answer_first_audio_ms":1176.31,"tts_first_audio_ms":741.24,"first_audio_ms":1176.32,"tts_server_stages":[{"generation_ms":1075,"alignment_ms":44,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1237,"performance":{"device_seconds":0.8726874685287476,"raw_audio_seconds":3.04,"aggregate_rtf":0.2870682462265617,"audio_seconds_per_second":3.483492211851164,"batches":[{"batch_size":1,"device_seconds":0.8726874685287476,"audio_seconds":[3.04],"model_wall_seconds":0.9614429473876953,"first_audio_seconds":0.4565145263671875,"graph_capture_seconds":0,"live_callback_seconds":0.08843447268009186}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}},{"generation_ms":1029,"alignment_ms":39,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1179,"performance":{"device_seconds":0.8219381828308105,"raw_audio_seconds":2.8,"aggregate_rtf":0.29354935101100377,"audio_seconds_per_second":3.4065822205224863,"batches":[{"batch_size":1,"device_seconds":0.8219381828308105,"audio_seconds":[2.8],"model_wall_seconds":0.906829223036766,"first_audio_seconds":0.4643337097167969,"graph_capture_seconds":0,"live_callback_seconds":0.08464871346950531}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}},{"generation_ms":2054,"alignment_ms":40,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":2215,"performance":{"device_seconds":1.707684470176697,"raw_audio_seconds":6.08,"aggregate_rtf":0.280869156279062,"audio_seconds_per_second":3.5603767008380025,"batches":[{"batch_size":1,"device_seconds":1.707684470176697,"audio_seconds":[6.08],"model_wall_seconds":1.943635381758213,"first_audio_seconds":0.4609619445800781,"graph_capture_seconds":0,"live_callback_seconds":0.23214926570653915}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1248.24,"audio_s":2.19},{"sentence":1,"total_ms":1186.93,"audio_s":2.67},{"sentence":2,"total_ms":2228.92,"audio_s":6.08}],"total_ms":5101.49,"audio_s":10.94,"sentences":3,"generation_rtf":0.4663},"at":12225.70000076294},{"type":"turn_done","barged":false,"at":19689.70000076294},{"type":"voice_caption","status":"available","caption":"Possibly very fading, very smooth, slow, very liberation","uncertainty":"Uncalibrated acoustic estimates; noise, other speakers and acting may be misleading.","voice_impression_only":false,"window_s":1.62,"inference_ms":56.09,"model":"laion/humaneness-ears-base-medium","model_revision":"818506970d93c809a3295a0a02c40dee8ff2bfc3","rules_revision":"7ebd4a9b7da96df989af8fc6ad0ce34f4e8aa3ed","device":"cuda:1","precision":"bfloat16","selected":{"voicenet":[{"dim":"DARC","name":"Dynamic Arc","value":2.357,"z":-1.67,"direction":"below","intensity":"Very","always_on":false,"phrase":"very fading in loudness","tag":"very fading","reliability":1,"rank_score":1.671},{"dim":"DFLU","name":"Disfluency","value":-0.258,"z":-1.56,"direction":"below","intensity":"Very","always_on":false,"phrase":"very smooth and flawless","tag":"very smooth","reliability":1,"rank_score":1.563},{"dim":"TEMP","name":"Tempo","value":2.38,"z":-1.11,"direction":"below","intensity":"Notably","always_on":true,"phrase":"notably slow and deliberate in tempo","tag":"slow","reliability":1,"rank_score":1.112}],"emotions":[{"emotion":"Relief","value":1.929,"z":3.7,"direction":"above","intensity":"Extremely","synonym":"liberation","phrase":"extremely carrying liberation","tag":"very liberation"}]},"normalization":"speech reference z-scores; not probabilities","scope":"this turn","request_ms":66.17,"added_wait_ms":0,"at":19690.90000152588},{"type":"endpoint","transcript":"Schön, dich kennenzulernen.","speaker_id":"speaker-1","asr":{"text":"Schön, dich kennenzulernen.","model":"parakeet_v3","chunks":1,"transcribe_ms":47.12,"speech_pause_ms":700,"observed_quiet_ms":598},"speaker_name":"Christoph","by":"asr_complete_sentence","timing":{"speaker_assignment_ms":0.481,"speaker_embedding_wait_ms":0.003,"speaker_lookup_ms":0.478,"speaker_prefetch_used":true,"speaker_prefetch_compute_ms":48.68,"sentence_queue_ms":10717.31,"sentence_audio_s":7.04},"at":19691},{"type":"latency_start","reply_id":2,"input":{"kind":"voice","asr":{"text":"Schön, dich kennenzulernen.","model":"parakeet_v3","chunks":1,"transcribe_ms":47.12,"speech_pause_ms":700,"observed_quiet_ms":598},"voice_caption_ms":66.17,"voice_caption_status":"available","speaker_assignment_ms":0.481,"speaker_prefetch_compute_ms":48.68,"speaker_prefetch_used":true,"sentence_queue_ms":10717.3},"at":19691.5},{"type":"reply","text":"Mir auch, Christoph!","reply_id":2,"performance":{"sentences":[{"emotion":"interest","secondary":"affection","blend":0.25,"intensity":0.72,"direction":"interest, warm","text":"Mir auch, Christoph!","start_s":0,"end_s":1.02}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Mir auch, Christoph!","sentence_role":"answer","speaker_id":"speaker-1","speaker_name":"Christoph","speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":1118.37,"finish_reason":"stop","usage":{"prompt_tokens":1792,"completion_tokens":11,"total_tokens":1803,"cost":0.0017725653,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00179047,"upstream_inference_prompt_cost":0.00177408,"upstream_inference_completions_cost":1.639e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":1175.04,"generation_calls":1,"llm_total_ms":1175.06,"response_complete":true,"sentence_ready_ms":1176.16,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":20873.70000076294},{"type":"latency_update","reply_id":2,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":1118.37,"finish_reason":"stop","usage":{"prompt_tokens":1792,"completion_tokens":11,"total_tokens":1803,"cost":0.0017725653,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00179047,"upstream_inference_prompt_cost":0.00177408,"upstream_inference_completions_cost":1.639e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":1175.04,"generation_calls":1,"llm_total_ms":1175.06,"response_complete":true,"answer_first_audio_ms":1925.54,"tts_first_audio_ms":749.23,"first_audio_ms":1925.55},"at":21617.5},{"type":"pipeline_metrics","reply_id":2,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":1118.37,"finish_reason":"stop","usage":{"prompt_tokens":1792,"completion_tokens":11,"total_tokens":1803,"cost":0.0017725653,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00179047,"upstream_inference_prompt_cost":0.00177408,"upstream_inference_completions_cost":1.639e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":1175.04,"generation_calls":1,"llm_total_ms":1175.06,"response_complete":true,"answer_first_audio_ms":1925.54,"tts_first_audio_ms":749.23,"first_audio_ms":1925.55,"tts_server_stages":[{"generation_ms":624,"alignment_ms":36,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":776,"performance":{"device_seconds":0.463185085773468,"raw_audio_seconds":1.6,"aggregate_rtf":0.2894906786084175,"audio_seconds_per_second":3.4543426572731213,"batches":[{"batch_size":1,"device_seconds":0.463185085773468,"audio_seconds":[1.6],"model_wall_seconds":0.5075870901346207,"first_audio_seconds":0.4563240966796875,"graph_capture_seconds":0,"live_callback_seconds":0.044230952858924866}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":784.46,"audio_s":1.55}],"total_ms":1960.8,"audio_s":1.55,"sentences":1,"generation_rtf":1.265},"at":21653.79999923706},{"type":"turn_done","barged":false,"at":23486.70000076294}],"reply":"Mir auch, Christoph!","captions":[{"status":"available","caption":"The audio contains a single male voice speaking in German. The speaker's tone is conversational and relatively neutral, with a moderate pace.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.12,"tokens":28,"preprocess_ms":10.92,"encoder_gpu_ms":8.33,"decoder_gpu_ms":217.98,"inference_ms":226.56,"total_ms":237.47,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":250.15,"capture_ms":1920,"event_age_ms":2170.16,"kind":"voice","eligible":false,"suppression_reason":"speech_or_quiet","at":6657.5}],"pcm_chunks":15,"page_errors":[],"passed":true,"speakers":["Christoph","Christoph"]},{"case":"selma-followup-clear","events":[{"type":"asr_preview","text":"","status":"sentence_stream","source":"local_sentence_asr","at":26541.5},{"type":"asr_preview","text":"","status":"sentence_stream","source":"local_sentence_asr","at":27028.70000076294},{"type":"asr_preview","text":"Was können wir","status":"sentence_stream","source":"native_asr","at":27440},{"type":"asr_preview","text":"Was können wir heute zu","status":"sentence_stream","source":"local_sentence_asr","at":28173.29999923706},{"type":"asr_preview","text":"Was können wir heute machen?","status":"sentence_stream","source":"native_asr","at":28406.20000076294},{"status":"available","caption":"A female voice speaks in German. The speaker is asking a question, and the tone is neutral and conversational.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.42,"tokens":24,"preprocess_ms":10.64,"encoder_gpu_ms":21.17,"decoder_gpu_ms":210.29,"inference_ms":231.72,"total_ms":242.36,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":255.19,"capture_ms":2220,"event_age_ms":2475.2,"kind":"voice","eligible":false,"suppression_reason":"speech_or_quiet","at":28940.29999923706},{"type":"asr_preview","text":"Was können wir heute zusammen machen","status":"sentence_stream","source":"local_sentence_asr","at":29208.5},{"type":"asr_finalizing","model":"parakeet_v3","label":"Transcribing the completed speech segment…","at":29259.40000152588},{"type":"asr_preview","text":"Was können wir heute zusammen machen?","status":"sentence_stream","source":"local_sentence_asr","at":29320.29999923706},{"type":"asr_sentence","text":"Was können wir heute zusammen machen?","audio_s":26.88,"queued":false,"asr":{"text":"Was können wir heute zusammen machen?","model":"parakeet_v3","chunks":1,"transcribe_ms":48.11,"speech_pause_ms":700,"observed_quiet_ms":760},"reason":"neural speech pause","speech_pause_ms":700,"observed_quiet_ms":760,"at":29320.70000076294},{"type":"identity","enabled":false,"mode":"open-sentences","sentence_mode":true,"open_enrollment":true,"speakers":[{"id":"speaker-1","name":"Christoph","named":true,"reference_ready":true,"voice_variants":1,"admitted":true},{"id":"speaker-2","name":"Speaker 2","named":false,"reference_ready":true,"voice_variants":1,"admitted":true}],"users":["Christoph","Speaker 2"],"threshold":0.38,"conversation_focus":{"enabled":false,"threshold":0.28,"timeout_s":180,"speaker_id":null,"name":null,"primary_id":"speaker-1","participants":[{"id":"speaker-1","name":"Christoph"},{"id":"speaker-2","name":"Speaker 2"}]},"handover_pending":false,"storage":"Session RAM","prediction":{"name":"Speaker 2","similarity":0.0497},"at":29391.70000076294},{"type":"voice_caption","status":"available","caption":"Possibly smooth, slightly fading, very hope","uncertainty":"Uncalibrated acoustic estimates; noise, other speakers and acting may be misleading.","voice_impression_only":false,"window_s":2.64,"inference_ms":21.72,"model":"laion/humaneness-ears-base-medium","model_revision":"818506970d93c809a3295a0a02c40dee8ff2bfc3","rules_revision":"7ebd4a9b7da96df989af8fc6ad0ce34f4e8aa3ed","device":"cuda:1","precision":"bfloat16","selected":{"voicenet":[{"dim":"DFLU","name":"Disfluency","value":0.259,"z":-1.04,"direction":"below","intensity":"Notably","always_on":false,"phrase":"notably smooth and flawless","tag":"smooth","reliability":1,"rank_score":1.039},{"dim":"DARC","name":"Dynamic Arc","value":2.682,"z":-0.78,"direction":"below","intensity":"Somewhat","always_on":false,"phrase":"somewhat fading in loudness","tag":"slightly fading","reliability":1,"rank_score":0.775},{"dim":"TEMP","name":"Tempo","value":2.847,"z":-0.22,"direction":"below","intensity":"About-average","always_on":true,"phrase":"moderate in tempo","tag":null,"reliability":1,"rank_score":0.218}],"emotions":[{"emotion":"Hope/Optimism","value":2.139,"z":1.97,"direction":"above","intensity":"Very","synonym":"hope","phrase":"very carrying hope","tag":"very hope"}]},"normalization":"speech reference z-scores; not probabilities","scope":"this turn","request_ms":33.46,"added_wait_ms":0,"at":29393.5},{"type":"endpoint","transcript":"Was können wir heute zusammen machen?","speaker_id":"speaker-2","asr":{"text":"Was können wir heute zusammen machen?","model":"parakeet_v3","chunks":1,"transcribe_ms":48.11,"speech_pause_ms":700,"observed_quiet_ms":760},"speaker_name":"Speaker 2","by":"asr_complete_sentence","timing":{"speaker_assignment_ms":71.123,"speaker_embedding_wait_ms":70.552,"speaker_lookup_ms":0.571,"speaker_prefetch_used":false,"speaker_prefetch_compute_ms":null,"sentence_queue_ms":71.39,"sentence_audio_s":26.88},"at":29393.60000228882},{"type":"latency_start","reply_id":3,"input":{"kind":"voice","asr":{"text":"Was können wir heute zusammen machen?","model":"parakeet_v3","chunks":1,"transcribe_ms":48.11,"speech_pause_ms":700,"observed_quiet_ms":760},"voice_caption_ms":33.46,"voice_caption_status":"available","speaker_assignment_ms":71.123,"speaker_prefetch_compute_ms":null,"speaker_prefetch_used":false,"sentence_queue_ms":71.39},"at":29393.90000152588},{"type":"reply","text":"Wir könnten eine kleine Zeitreise unternehmen und uns überlegen, wie ein Roboter im alten Rom eine Toga tragen würde!","reply_id":3,"performance":{"sentences":[{"emotion":"interest","secondary":"affection","blend":0.25,"intensity":0.72,"direction":"interest, warm","text":"Wir könnten eine kleine Zeitreise unternehmen und uns überlegen, wie ein Roboter im alten Rom eine Toga tragen würde!","start_s":0,"end_s":6.46}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Wir könnten eine kleine Zeitreise unternehmen und uns überlegen, wie ein Roboter im alten Rom eine Toga tragen würde!","sentence_role":"answer","speaker_id":"speaker-2","speaker_name":"Speaker 2","speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":359.62,"finish_reason":"stop","usage":{"prompt_tokens":1925,"completion_tokens":32,"total_tokens":1957,"cost":0.0019338957,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00195343,"upstream_inference_prompt_cost":0.00190575,"upstream_inference_completions_cost":4.768e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":437.14,"generation_calls":1,"llm_total_ms":437.16,"response_complete":true,"sentence_ready_ms":438.26,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":29833.40000152588},{"type":"latency_update","reply_id":3,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":359.62,"finish_reason":"stop","usage":{"prompt_tokens":1925,"completion_tokens":32,"total_tokens":1957,"cost":0.0019338957,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00195343,"upstream_inference_prompt_cost":0.00190575,"upstream_inference_completions_cost":4.768e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":437.14,"generation_calls":1,"llm_total_ms":437.16,"response_complete":true,"answer_first_audio_ms":1201.46,"tts_first_audio_ms":763.04,"first_audio_ms":1201.46},"at":30595.40000152588},{"type":"pipeline_metrics","reply_id":3,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":359.62,"finish_reason":"stop","usage":{"prompt_tokens":1925,"completion_tokens":32,"total_tokens":1957,"cost":0.0019338957,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00195343,"upstream_inference_prompt_cost":0.00190575,"upstream_inference_completions_cost":4.768e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":437.14,"generation_calls":1,"llm_total_ms":437.16,"response_complete":true,"answer_first_audio_ms":1201.46,"tts_first_audio_ms":763.04,"first_audio_ms":1201.46,"tts_server_stages":[{"generation_ms":2507,"alignment_ms":40,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":2660,"performance":{"device_seconds":2.0550278992652893,"raw_audio_seconds":7.12,"aggregate_rtf":0.2886275139417541,"audio_seconds_per_second":3.4646731572576375,"batches":[{"batch_size":1,"device_seconds":2.0550278992652893,"audio_seconds":[7.12],"model_wall_seconds":2.3933705165982246,"first_audio_seconds":0.4790118408203125,"graph_capture_seconds":0,"live_callback_seconds":0.3339344784617424}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":2675.97,"audio_s":7.07}],"total_ms":3114.4,"audio_s":7.07,"sentences":1,"generation_rtf":0.4405},"at":32524.10000228882},{"type":"turn_done","barged":false,"at":38112.70000076294}],"reply":"Wir könnten eine kleine Zeitreise unternehmen und uns überlegen, wie ein Roboter im alten Rom eine Toga tragen würde!","captions":[{"status":"available","caption":"A female voice speaks in German. The speaker is asking a question, and the tone is neutral and conversational.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.42,"tokens":24,"preprocess_ms":10.64,"encoder_gpu_ms":21.17,"decoder_gpu_ms":210.29,"inference_ms":231.72,"total_ms":242.36,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":255.19,"capture_ms":2220,"event_age_ms":2475.2,"kind":"voice","eligible":false,"suppression_reason":"speech_or_quiet","at":28940.29999923706}],"pcm_chunks":7,"page_errors":[],"passed":true,"speakers":["Speaker 2"]},{"case":"selma-intro","events":[{"type":"asr_preview","text":"","status":"sentence_stream","source":"local_sentence_asr","at":40157.29999923706},{"type":"asr_preview","text":"","status":"sentence_stream","source":"local_sentence_asr","at":40654.5},{"type":"asr_preview","text":"Hey Robby.","status":"sentence_stream","source":"local_sentence_asr","at":41155},{"type":"asr_finalizing","model":"parakeet_v3","label":"Transcribing the completed speech segment…","at":41533.79999923706},{"type":"asr_preview","text":"Hey Robbie.","status":"sentence_stream","source":"local_sentence_asr","at":41593.40000152588},{"type":"asr_preview","text":"Hey Robi.","status":"sentence_stream","source":"native_asr","at":41826.10000228882},{"type":"asr_preview","text":"Hey Robbie, ich bin Selma.","status":"sentence_stream","source":"local_sentence_asr","at":42555.10000228882},{"type":"asr_preview","text":"Hey Robbie, ich bin Selma.","status":"sentence_stream","source":"native_asr","at":42789},{"status":"available","caption":"A female voice, perceived as adult, speaks in a calm and clear tone. The speech is delivered at a moderate pace, with distinct pronunciation.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.74,"tokens":30,"preprocess_ms":8.63,"encoder_gpu_ms":8.27,"decoder_gpu_ms":241.58,"inference_ms":250.1,"total_ms":258.73,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":270.92,"capture_ms":2660,"event_age_ms":2930.92,"kind":"voice","eligible":false,"suppression_reason":"speech_or_quiet","at":42925.29999923706},{"type":"asr_finalizing","model":"parakeet_v3","label":"Transcribing the completed speech segment…","at":43055},{"type":"asr_preview","text":"Hey Robbie, ich bin Selma.","status":"sentence_stream","source":"local_sentence_asr","at":43129.60000228882},{"type":"asr_sentence","text":"Hey Robbie, ich bin Selma.","audio_s":40.32,"queued":false,"asr":{"text":"Hey Robbie, ich bin Selma.","model":"parakeet_v3","chunks":1,"transcribe_ms":60.02,"speech_pause_ms":700,"observed_quiet_ms":8},"reason":"neural speech pause","speech_pause_ms":700,"observed_quiet_ms":8,"at":43129.79999923706},{"type":"identity","enabled":false,"mode":"open-sentences","sentence_mode":true,"open_enrollment":true,"speakers":[{"id":"speaker-1","name":"Christoph","named":true,"reference_ready":true,"voice_variants":1,"admitted":true},{"id":"speaker-2","name":"Speaker 2","named":false,"reference_ready":true,"voice_variants":1,"admitted":true},{"id":"speaker-3","name":"Selma","named":true,"reference_ready":true,"voice_variants":1,"admitted":true}],"users":["Christoph","Speaker 2","Selma"],"threshold":0.38,"conversation_focus":{"enabled":false,"threshold":0.28,"timeout_s":180,"speaker_id":null,"name":null,"primary_id":"speaker-1","participants":[{"id":"speaker-1","name":"Christoph"},{"id":"speaker-2","name":"Speaker 2"},{"id":"speaker-3","name":"Selma"}]},"handover_pending":false,"storage":"Session RAM","prediction":{"name":"Selma","similarity":0.3152},"at":43198.29999923706},{"type":"voice_caption","status":"available","caption":"Possibly very soft-onset, very fading, slow, liberation","uncertainty":"Uncalibrated acoustic estimates; noise, other speakers and acting may be misleading.","voice_impression_only":false,"window_s":2.956,"inference_ms":24.95,"model":"laion/humaneness-ears-base-medium","model_revision":"818506970d93c809a3295a0a02c40dee8ff2bfc3","rules_revision":"7ebd4a9b7da96df989af8fc6ad0ce34f4e8aa3ed","device":"cuda:1","precision":"bfloat16","selected":{"voicenet":[{"dim":"ATCK","name":"Attack","value":1.558,"z":-2.48,"direction":"below","intensity":"Extremely","always_on":false,"phrase":"extremely soft and gentle in onset","tag":"very soft-onset","reliability":1,"rank_score":2.481},{"dim":"DARC","name":"Dynamic Arc","value":2.203,"z":-2.1,"direction":"below","intensity":"Extremely","always_on":false,"phrase":"extremely fading in loudness","tag":"very fading","reliability":1,"rank_score":2.098},{"dim":"TEMP","name":"Tempo","value":2.348,"z":-1.17,"direction":"below","intensity":"Notably","always_on":true,"phrase":"notably slow and deliberate in tempo","tag":"slow","reliability":1,"rank_score":1.172}],"emotions":[{"emotion":"Relief","value":0.907,"z":1.43,"direction":"above","intensity":"Notably","synonym":"liberation","phrase":"notably carrying liberation","tag":"liberation"}]},"normalization":"speech reference z-scores; not probabilities","scope":"this turn","request_ms":35.98,"added_wait_ms":0,"at":43200.5},{"type":"endpoint","transcript":"Hey Robbie, ich bin Selma.","speaker_id":"speaker-3","asr":{"text":"Hey Robbie, ich bin Selma.","model":"parakeet_v3","chunks":1,"transcribe_ms":60.02,"speech_pause_ms":700,"observed_quiet_ms":8},"speaker_name":"Selma","by":"asr_complete_sentence","timing":{"speaker_assignment_ms":71.697,"speaker_embedding_wait_ms":71.126,"speaker_lookup_ms":0.571,"speaker_prefetch_used":false,"speaker_prefetch_compute_ms":null,"sentence_queue_ms":71.95,"sentence_audio_s":40.32},"at":43200.60000228882},{"type":"latency_start","reply_id":4,"input":{"kind":"voice","asr":{"text":"Hey Robbie, ich bin Selma.","model":"parakeet_v3","chunks":1,"transcribe_ms":60.02,"speech_pause_ms":700,"observed_quiet_ms":8},"voice_caption_ms":35.98,"voice_caption_status":"available","speaker_assignment_ms":71.697,"speaker_prefetch_compute_ms":null,"speaker_prefetch_used":false,"sentence_queue_ms":71.95},"at":43201.20000076294},{"type":"asr_preview","text":"schön.","status":"sentence_stream","source":"local_sentence_asr","at":43575.40000152588},{"type":"asr_preview","text":"schön, dass du","status":"sentence_stream","source":"local_sentence_asr","at":44065.40000152588},{"type":"reply","text":"Hallo Selma, schön, dass du dich vorstellst!","reply_id":4,"performance":{"sentences":[{"emotion":"interest","secondary":"affection","blend":0.25,"intensity":0.72,"direction":"interest, warm","text":"Hallo Selma, schön, dass du dich vorstellst!","start_s":0,"end_s":2.38}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Hallo Selma, schön, dass du dich vorstellst!","sentence_role":"answer","speaker_id":"speaker-3","speaker_name":"Selma","speaker_update":{"speaker_id":"speaker-3","display_name":"Selma","is_self_introduction":true,"source":"validated_acoustic_transcript"},"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":994.38,"finish_reason":"stop","usage":{"prompt_tokens":1975,"completion_tokens":19,"total_tokens":1994,"cost":0.0019637244,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00198356,"upstream_inference_prompt_cost":0.00195525,"upstream_inference_completions_cost":2.831e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":1047.03,"generation_calls":1,"llm_total_ms":1047.06,"response_complete":true,"sentence_ready_ms":1048.65,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":44256.20000076294},{"type":"asr_preview","text":"schön, dass du da bist.","status":"sentence_stream","source":"local_sentence_asr","at":44573.70000076294},{"type":"asr_finalizing","model":"parakeet_v3","label":"Transcribing the completed speech segment…","at":44841.29999923706},{"type":"asr_preview","text":"Schön, dass du da bist.","status":"sentence_stream","source":"local_sentence_asr","at":44897.90000152588},{"type":"asr_sentence","text":"Schön, dass du da bist.","audio_s":42.24,"queued":true,"asr":{"text":"Schön, dass du da bist.","model":"parakeet_v3","chunks":1,"transcribe_ms":44.11,"speech_pause_ms":700,"observed_quiet_ms":470},"reason":"neural speech pause","speech_pause_ms":700,"observed_quiet_ms":470,"at":44898.20000076294},{"type":"identity","enabled":false,"mode":"open-sentences","sentence_mode":true,"open_enrollment":true,"speakers":[{"id":"speaker-1","name":"Christoph","named":true,"reference_ready":true,"voice_variants":1,"admitted":true},{"id":"speaker-2","name":"Speaker 2","named":false,"reference_ready":true,"voice_variants":1,"admitted":true},{"id":"speaker-3","name":"Selma","named":true,"reference_ready":true,"voice_variants":1,"admitted":true}],"users":["Christoph","Speaker 2","Selma"],"threshold":0.38,"conversation_focus":{"enabled":false,"threshold":0.28,"timeout_s":180,"speaker_id":null,"name":null,"primary_id":"speaker-1","participants":[{"id":"speaker-1","name":"Christoph"},{"id":"speaker-2","name":"Speaker 2"},{"id":"speaker-3","name":"Selma"}]},"handover_pending":false,"storage":"Session RAM","prediction":{"name":"Speaker 2","similarity":0.4813},"at":44898.40000152588},{"type":"latency_update","reply_id":4,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":994.38,"finish_reason":"stop","usage":{"prompt_tokens":1975,"completion_tokens":19,"total_tokens":1994,"cost":0.0019637244,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00198356,"upstream_inference_prompt_cost":0.00195525,"upstream_inference_completions_cost":2.831e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":1047.03,"generation_calls":1,"llm_total_ms":1047.06,"response_complete":true,"answer_first_audio_ms":1807.57,"tts_first_audio_ms":758.71,"first_audio_ms":1807.57},"at":45006.29999923706},{"type":"pipeline_metrics","reply_id":4,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":994.38,"finish_reason":"stop","usage":{"prompt_tokens":1975,"completion_tokens":19,"total_tokens":1994,"cost":0.0019637244,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00198356,"upstream_inference_prompt_cost":0.00195525,"upstream_inference_completions_cost":2.831e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":1047.03,"generation_calls":1,"llm_total_ms":1047.06,"response_complete":true,"answer_first_audio_ms":1807.57,"tts_first_audio_ms":758.71,"first_audio_ms":1807.57,"tts_server_stages":[{"generation_ms":1121,"alignment_ms":40,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1276,"performance":{"device_seconds":0.8949256176948547,"raw_audio_seconds":3.12,"aggregate_rtf":0.28683513387655596,"audio_seconds_per_second":3.486323263419905,"batches":[{"batch_size":1,"device_seconds":0.8949256176948547,"audio_seconds":[3.12],"model_wall_seconds":1.0057845711708069,"first_audio_seconds":0.4666430969238281,"graph_capture_seconds":0,"live_callback_seconds":0.10714691132307053}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1289.46,"audio_s":2.67}],"total_ms":2338.33,"audio_s":2.67,"sentences":1,"generation_rtf":0.8758},"at":45542.60000228882},{"type":"turn_done","barged":false,"at":48137.79999923706},{"type":"voice_caption","status":"available","caption":"Possibly very smooth, choppy, slightly slow, very liberation","uncertainty":"Uncalibrated acoustic estimates; noise, other speakers and acting may be misleading.","voice_impression_only":false,"window_s":1.32,"inference_ms":25.66,"model":"laion/humaneness-ears-base-medium","model_revision":"818506970d93c809a3295a0a02c40dee8ff2bfc3","rules_revision":"7ebd4a9b7da96df989af8fc6ad0ce34f4e8aa3ed","device":"cuda:1","precision":"bfloat16","selected":{"voicenet":[{"dim":"DFLU","name":"Disfluency","value":-0.756,"z":-2.07,"direction":"below","intensity":"Extremely","always_on":false,"phrase":"extremely smooth and flawless","tag":"very smooth","reliability":1,"rank_score":2.069},{"dim":"CHNK","name":"Chunking","value":2.111,"z":-1.19,"direction":"below","intensity":"Notably","always_on":false,"phrase":"notably choppy and fragmented","tag":"choppy","reliability":1,"rank_score":1.19},{"dim":"TEMP","name":"Tempo","value":2.46,"z":-0.96,"direction":"below","intensity":"Somewhat","always_on":true,"phrase":"somewhat slow and deliberate in tempo","tag":"slightly slow","reliability":1,"rank_score":0.959}],"emotions":[{"emotion":"Relief","value":2.235,"z":4.37,"direction":"above","intensity":"Extremely","synonym":"liberation","phrase":"extremely carrying liberation","tag":"very liberation"}]},"normalization":"speech reference z-scores; not probabilities","scope":"this turn","request_ms":39.3,"added_wait_ms":0,"at":48138.60000228882},{"type":"endpoint","transcript":"Schön, dass du da bist.","speaker_id":"speaker-2","asr":{"text":"Schön, dass du da bist.","model":"parakeet_v3","chunks":1,"transcribe_ms":44.11,"speech_pause_ms":700,"observed_quiet_ms":470},"speaker_name":"Speaker 2","by":"asr_complete_sentence","timing":{"speaker_assignment_ms":0.487,"speaker_embedding_wait_ms":0.002,"speaker_lookup_ms":0.485,"speaker_prefetch_used":true,"speaker_prefetch_compute_ms":51.56,"sentence_queue_ms":3231.88,"sentence_audio_s":42.24},"at":48138.70000076294},{"type":"latency_start","reply_id":5,"input":{"kind":"voice","asr":{"text":"Schön, dass du da bist.","model":"parakeet_v3","chunks":1,"transcribe_ms":44.11,"speech_pause_ms":700,"observed_quiet_ms":470},"voice_caption_ms":39.3,"voice_caption_status":"available","speaker_assignment_ms":0.487,"speaker_prefetch_compute_ms":51.56,"speaker_prefetch_used":true,"sentence_queue_ms":3231.88},"at":48139},{"type":"reply","text":"Danke!","reply_id":5,"performance":{"sentences":[{"emotion":"interest","secondary":"affection","blend":0.25,"intensity":0.72,"direction":"interest, warm","text":"Danke!","start_s":0,"end_s":0.4}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Danke!","sentence_role":"answer","speaker_id":"speaker-2","speaker_name":"Speaker 2","speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":310.14,"llm_first_sentence_ms":336.22,"sentence_ready_ms":337.38,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":48471.70000076294},{"type":"latency_update","reply_id":5,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":310.14,"llm_first_sentence_ms":336.22,"finish_reason":"stop","usage":{"prompt_tokens":2032,"completion_tokens":20,"total_tokens":2052,"cost":0.0020210652,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00204148,"upstream_inference_prompt_cost":0.00201168,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":373.23,"response_complete":true,"answer_first_audio_ms":1079.67,"tts_first_audio_ms":742.15,"first_audio_ms":1079.68},"at":49208.10000228882},{"type":"reply","text":"Danke! Es freut mich wirklich, hier mit euch allen zu sein.","reply_id":5,"performance":{"sentences":[{"emotion":"interest","secondary":"affection","blend":0.25,"intensity":0.72,"direction":"interest, warm","text":"Danke!","start_s":0,"end_s":2.43},{"emotion":"interest","secondary":"affection","blend":0.25,"intensity":0.72,"direction":"interest, warm","text":"Es freut mich wirklich, hier mit euch allen zu sein.","start_s":2.43,"end_s":5.83}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":1,"sentence_text":"Es freut mich wirklich, hier mit euch allen zu sein.","sentence_role":"answer","speaker_id":"speaker-2","speaker_name":"Speaker 2","speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":310.14,"llm_first_sentence_ms":336.22,"finish_reason":"stop","usage":{"prompt_tokens":2032,"completion_tokens":20,"total_tokens":2052,"cost":0.0020210652,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00204148,"upstream_inference_prompt_cost":0.00201168,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":373.23,"response_complete":true,"answer_first_audio_ms":1079.67,"tts_first_audio_ms":742.15,"first_audio_ms":1079.68,"tts_server_stages":[{"generation_ms":1278,"alignment_ms":40,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1435,"performance":{"device_seconds":1.0120126299858094,"raw_audio_seconds":3.28,"aggregate_rtf":0.3085404359712834,"audio_seconds_per_second":3.2410662701373525,"batches":[{"batch_size":1,"device_seconds":1.0120126299858094,"audio_seconds":[3.28],"model_wall_seconds":1.1621837243437767,"first_audio_seconds":0.46112664794921876,"graph_capture_seconds":0,"live_callback_seconds":0.14725860953330994}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1444.37,"audio_s":2.43}],"sentence_ready_ms":1782.89,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"interest.soft"},"sources":[],"lookups":[],"at":49937.5},{"type":"latency_update","reply_id":5,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":310.14,"llm_first_sentence_ms":336.22,"finish_reason":"stop","usage":{"prompt_tokens":2032,"completion_tokens":20,"total_tokens":2052,"cost":0.0020210652,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00204148,"upstream_inference_prompt_cost":0.00201168,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":373.23,"response_complete":true,"answer_first_audio_ms":1079.67,"tts_first_audio_ms":742.15,"first_audio_ms":1079.68,"tts_server_stages":[{"generation_ms":1278,"alignment_ms":40,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1435,"performance":{"device_seconds":1.0120126299858094,"raw_audio_seconds":3.28,"aggregate_rtf":0.3085404359712834,"audio_seconds_per_second":3.2410662701373525,"batches":[{"batch_size":1,"device_seconds":1.0120126299858094,"audio_seconds":[3.28],"model_wall_seconds":1.1621837243437767,"first_audio_seconds":0.46112664794921876,"graph_capture_seconds":0,"live_callback_seconds":0.14725860953330994}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1444.37,"audio_s":2.43}]},"at":50655.20000076294},{"type":"pipeline_metrics","reply_id":5,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":310.14,"llm_first_sentence_ms":336.22,"finish_reason":"stop","usage":{"prompt_tokens":2032,"completion_tokens":20,"total_tokens":2052,"cost":0.0020210652,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00204148,"upstream_inference_prompt_cost":0.00201168,"upstream_inference_completions_cost":2.98e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"generation_calls":1,"llm_total_ms":373.23,"response_complete":true,"answer_first_audio_ms":1079.67,"tts_first_audio_ms":742.15,"first_audio_ms":1079.68,"tts_server_stages":[{"generation_ms":1278,"alignment_ms":40,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1435,"performance":{"device_seconds":1.0120126299858094,"raw_audio_seconds":3.28,"aggregate_rtf":0.3085404359712834,"audio_seconds_per_second":3.2410662701373525,"batches":[{"batch_size":1,"device_seconds":1.0120126299858094,"audio_seconds":[3.28],"model_wall_seconds":1.1621837243437767,"first_audio_seconds":0.46112664794921876,"graph_capture_seconds":0,"live_callback_seconds":0.14725860953330994}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}},{"generation_ms":1419,"alignment_ms":39,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":1578,"performance":{"device_seconds":1.1624761910438537,"raw_audio_seconds":4,"aggregate_rtf":0.2906190477609634,"audio_seconds_per_second":3.440930688144393,"batches":[{"batch_size":1,"device_seconds":1.1624761910438537,"audio_seconds":[4],"model_wall_seconds":1.3069601580500603,"first_audio_seconds":0.4620411071777344,"graph_capture_seconds":0,"live_callback_seconds":0.14029020071029663}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":1444.37,"audio_s":2.43},{"sentence":1,"total_ms":1591.08,"audio_s":3.79}],"total_ms":3374.15,"audio_s":6.22,"sentences":2,"generation_rtf":0.5425},"at":51511.60000228882},{"type":"turn_done","barged":false,"at":55868.10000228882}],"reply":"Danke! Es freut mich wirklich, hier mit euch allen zu sein.","captions":[{"status":"available","caption":"A female voice, perceived as adult, speaks in a calm and clear tone. The speech is delivered at a moderate pace, with distinct pronunciation.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.74,"tokens":30,"preprocess_ms":8.63,"encoder_gpu_ms":8.27,"decoder_gpu_ms":241.58,"inference_ms":250.1,"total_ms":258.73,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":270.92,"capture_ms":2660,"event_age_ms":2930.92,"kind":"voice","eligible":false,"suppression_reason":"speech_or_quiet","at":42925.29999923706}],"pcm_chunks":14,"page_errors":[],"passed":true,"speakers":["Selma","Speaker 2"]},{"case":"selma-followup-clear","events":[{"type":"asr_preview","text":"","status":"sentence_stream","source":"local_sentence_asr","at":59036.29999923706},{"type":"asr_preview","text":"","status":"sentence_stream","source":"local_sentence_asr","at":59537.29999923706},{"type":"asr_preview","text":"Was können können","status":"sentence_stream","source":"native_asr","at":59823.70000076294},{"type":"asr_preview","text":"Was können wir heute zusammen","status":"sentence_stream","source":"local_sentence_asr","at":60547},{"type":"asr_preview","text":"Was können wir heute zusammen","status":"sentence_stream","source":"native_asr","at":60814.5},{"status":"available","caption":"A female voice speaks in German. The speaker is asking a question, and the tone is neutral and conversational.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.42,"tokens":24,"preprocess_ms":11.23,"encoder_gpu_ms":8.23,"decoder_gpu_ms":184.13,"inference_ms":192.59,"total_ms":203.82,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":227.05,"capture_ms":2220,"event_age_ms":2447.05,"kind":"voice","eligible":false,"suppression_reason":"speech_or_quiet","at":61355.10000228882},{"type":"asr_preview","text":"Was können wir heute zusammen machen?","status":"sentence_stream","source":"local_sentence_asr","at":61542.79999923706},{"type":"asr_finalizing","model":"parakeet_v3","label":"Transcribing the completed speech segment…","at":61606.60000228882},{"type":"asr_preview","text":"Was können wir heute zusammen machen?","status":"sentence_stream","source":"local_sentence_asr","at":61672},{"type":"asr_sentence","text":"Was können wir heute zusammen machen?","audio_s":58.56,"queued":false,"asr":{"text":"Was können wir heute zusammen machen?","model":"parakeet_v3","chunks":1,"transcribe_ms":51.73,"speech_pause_ms":700,"observed_quiet_ms":688},"reason":"neural speech pause","speech_pause_ms":700,"observed_quiet_ms":688,"at":61672.70000076294},{"type":"identity","enabled":false,"mode":"open-sentences","sentence_mode":true,"open_enrollment":true,"speakers":[{"id":"speaker-1","name":"Christoph","named":true,"reference_ready":true,"voice_variants":1,"admitted":true},{"id":"speaker-2","name":"Speaker 2","named":false,"reference_ready":true,"voice_variants":1,"admitted":true},{"id":"speaker-3","name":"Selma","named":true,"reference_ready":true,"voice_variants":1,"admitted":true}],"users":["Christoph","Speaker 2","Selma"],"threshold":0.38,"conversation_focus":{"enabled":false,"threshold":0.28,"timeout_s":180,"speaker_id":null,"name":null,"primary_id":"speaker-1","participants":[{"id":"speaker-1","name":"Christoph"},{"id":"speaker-2","name":"Speaker 2"},{"id":"speaker-3","name":"Selma"}]},"handover_pending":false,"storage":"Session RAM","prediction":{"name":"Speaker 2","similarity":0.9948},"at":61676.5},{"type":"voice_caption","status":"available","caption":"Possibly smooth, slightly steady, very hope","uncertainty":"Uncalibrated acoustic estimates; noise, other speakers and acting may be misleading.","voice_impression_only":false,"window_s":2.6,"inference_ms":22.85,"model":"laion/humaneness-ears-base-medium","model_revision":"818506970d93c809a3295a0a02c40dee8ff2bfc3","rules_revision":"7ebd4a9b7da96df989af8fc6ad0ce34f4e8aa3ed","device":"cuda:1","precision":"bfloat16","selected":{"voicenet":[{"dim":"DFLU","name":"Disfluency","value":0.24,"z":-1.06,"direction":"below","intensity":"Notably","always_on":false,"phrase":"notably smooth and flawless","tag":"smooth","reliability":1,"rank_score":1.057},{"dim":"VOLT","name":"Volatility","value":1.287,"z":-0.82,"direction":"below","intensity":"Somewhat","always_on":false,"phrase":"somewhat emotionally steady and static","tag":"slightly steady","reliability":1,"rank_score":0.817},{"dim":"TEMP","name":"Tempo","value":2.912,"z":-0.09,"direction":"below","intensity":"About-average","always_on":true,"phrase":"moderate in tempo","tag":null,"reliability":1,"rank_score":0.092}],"emotions":[{"emotion":"Hope/Optimism","value":2.27,"z":2.18,"direction":"above","intensity":"Extremely","synonym":"hope","phrase":"extremely carrying hope","tag":"very hope"}]},"normalization":"speech reference z-scores; not probabilities","scope":"this turn","request_ms":33.99,"added_wait_ms":0,"at":61679.60000228882},{"type":"endpoint","transcript":"Was können wir heute zusammen machen?","speaker_id":"speaker-2","asr":{"text":"Was können wir heute zusammen machen?","model":"parakeet_v3","chunks":1,"transcribe_ms":51.73,"speech_pause_ms":700,"observed_quiet_ms":688},"speaker_name":"Speaker 2","by":"asr_complete_sentence","timing":{"speaker_assignment_ms":0.545,"speaker_embedding_wait_ms":0.002,"speaker_lookup_ms":0.544,"speaker_prefetch_used":true,"speaker_prefetch_compute_ms":66.18,"sentence_queue_ms":0.78,"sentence_audio_s":58.56},"at":61679.70000076294},{"type":"latency_start","reply_id":6,"input":{"kind":"voice","asr":{"text":"Was können wir heute zusammen machen?","model":"parakeet_v3","chunks":1,"transcribe_ms":51.73,"speech_pause_ms":700,"observed_quiet_ms":688},"voice_caption_ms":33.99,"voice_caption_status":"available","speaker_assignment_ms":0.545,"speaker_prefetch_compute_ms":66.18,"speaker_prefetch_used":true,"sentence_queue_ms":0.77},"at":61680.40000152588},{"type":"reply","text":"Wir könnten eine kleine Zeitreise unternehmen und überlegen, wie ein Roboter im alten Rom eine Toga tragen würde!","reply_id":6,"performance":{"sentences":[{"emotion":"amusement","secondary":null,"blend":0,"intensity":0.72,"direction":"amusement, playful","text":"Wir könnten eine kleine Zeitreise unternehmen und überlegen, wie ein Roboter im alten Rom eine Toga tragen würde!","start_s":0,"end_s":6.12}],"source":"qwen/qwen3.8-27b sentence cues + fixed GENERAL catalogue","timing":"Sentence starts anchored to actual preceding PCM duration; within-sentence estimates"},"avatar_actions":[],"avatar_action_id":null,"avatar_event":null,"sentence_index":0,"sentence_text":"Wir könnten eine kleine Zeitreise unternehmen und überlegen, wie ein Roboter im alten Rom eine Toga tragen würde!","sentence_role":"answer","speaker_id":"speaker-2","speaker_name":"Speaker 2","speaker_update":null,"timing":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":729.47,"finish_reason":"stop","usage":{"prompt_tokens":2141,"completion_tokens":31,"total_tokens":2172,"cost":0.0021441222,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00216578,"upstream_inference_prompt_cost":0.00211959,"upstream_inference_completions_cost":4.619e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":844.49,"generation_calls":1,"llm_total_ms":844.51,"response_complete":true,"sentence_ready_ms":845.61,"director_mode":"bolt","director_model":"qwen/qwen3.8-27b","template_id":"amusement.soft"},"sources":[],"lookups":[],"at":62522.70000076294},{"type":"latency_update","reply_id":6,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":729.47,"finish_reason":"stop","usage":{"prompt_tokens":2141,"completion_tokens":31,"total_tokens":2172,"cost":0.0021441222,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00216578,"upstream_inference_prompt_cost":0.00211959,"upstream_inference_completions_cost":4.619e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":844.49,"generation_calls":1,"llm_total_ms":844.51,"response_complete":true,"answer_first_audio_ms":1604.8,"tts_first_audio_ms":759.05,"first_audio_ms":1604.81},"at":63278.90000152588},{"type":"pipeline_metrics","reply_id":6,"metrics":{"quick_reply_status":"disabled_or_ineligible","llm_model":"qwen/qwen3.8-27b","llm_provider":"cerebras/fp16","llm_backend":"OpenRouter","llm_actual_provider":"Cerebras","llm_ttft_ms":729.47,"finish_reason":"stop","usage":{"prompt_tokens":2141,"completion_tokens":31,"total_tokens":2172,"cost":0.0021441222,"is_byok":false,"prompt_tokens_details":{"cached_tokens":1024,"cache_write_tokens":0,"audio_tokens":0,"video_tokens":0},"cost_details":{"upstream_inference_cost":0.00216578,"upstream_inference_prompt_cost":0.00211959,"upstream_inference_completions_cost":4.619e-05},"completion_tokens_details":{"reasoning_tokens":0,"image_tokens":0,"audio_tokens":0}},"llm_first_sentence_ms":844.49,"generation_calls":1,"llm_total_ms":844.51,"response_complete":true,"answer_first_audio_ms":1604.8,"tts_first_audio_ms":759.05,"first_audio_ms":1604.81,"tts_server_stages":[{"generation_ms":2366,"alignment_ms":39,"scoring_ms":0,"speaker_preparation_ms":0,"total_ms":2520,"performance":{"device_seconds":1.934513952255249,"raw_audio_seconds":6.72,"aggregate_rtf":0.28787410003798347,"audio_seconds_per_second":3.4737407771941107,"batches":[{"batch_size":1,"device_seconds":1.934513952255249,"audio_seconds":[6.72],"model_wall_seconds":2.2482414841651917,"first_audio_seconds":0.4745248107910156,"graph_capture_seconds":0,"live_callback_seconds":0.31331446021795273}],"streaming":true,"chunk_frames":20,"backend":"fused CFG + local CUDA graphs + compiled global","includes":"TTS prefill, autoregressive decode and PCM codec; GPU event span including launch gaps","excludes":"reference encoding, prompt packing, Luna, alignment, scoring, voice conversion, CPU audio copy, MP3, network","graph_capture_seconds":0,"oom_batch_retries":[]}}],"sentence_tts":[{"sentence":0,"total_ms":2535.82,"audio_s":6.59}],"total_ms":3381.6,"audio_s":6.59,"sentences":1,"generation_rtf":0.5131},"at":65061.5},{"type":"turn_done","barged":false,"at":70304}],"reply":"Wir könnten eine kleine Zeitreise unternehmen und überlegen, wie ein Roboter im alten Rom eine Toga tragen würde!","captions":[{"status":"available","caption":"A female voice speaks in German. The speaker is asking a question, and the tone is neutral and conversational.","uncertainty":"Fallible sound hypothesis, not a transcript or confirmed event.","window_s":2.42,"tokens":24,"preprocess_ms":11.23,"encoder_gpu_ms":8.23,"decoder_gpu_ms":184.13,"inference_ms":192.59,"total_ms":203.82,"model":"laion/sound-effect-captioning-whisper","revision":"bb36f9c6534037a8d1d78cff97d119dd37da478a","device":"cuda:0","precision":"float16","type":"sound_caption","request_ms":227.05,"capture_ms":2220,"event_age_ms":2447.05,"kind":"voice","eligible":false,"suppression_reason":"speech_or_quiet","at":61355.10000228882}],"pcm_chunks":7,"page_errors":[],"passed":true,"speakers":["Speaker 2"]}],"sound_provenance":{"dog-1":{"filename":"1-100032-A-0.wav","fold":"1","target":"0","category":"dog","esc10":"True","src_file":"100032","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-100032-A-0.wav","sha256":"f40a849a2375c8c63312a73dd2dd6c74007301fcc21b4be2ece29a642831e3d8","credit":"- [1-100032-A.ogg]: clip derived from rose_bark.wav (http://www.freesound.org/people/nfrae/sounds/100032/) by nfrae [CC0]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"dog-2":{"filename":"1-110389-A-0.wav","fold":"1","target":"0","category":"dog","esc10":"True","src_file":"110389","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-110389-A-0.wav","sha256":"4a18f3f834eb184b9f02595ac74c7de37ef8048a3da92f58f27018fec580b36a","credit":"- [1-110389-A.ogg]: clip derived from animals_dog_bark_springer_spaniel_001.wav (http://www.freesound.org/people/soundscalpel.com/sounds/110389/) by soundscalpel.com [CC-BY]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"cat-1":{"filename":"1-34094-A-5.wav","fold":"1","target":"5","category":"cat","esc10":"False","src_file":"34094","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-34094-A-5.wav","sha256":"ceddf296bcdda3996d35d3f4c292c338050bcb4c5a05a96ca22a69d33e33e4de","credit":"- [1-34094-A.ogg]: clip derived from 070422-cats-sample.wav (http://www.freesound.org/people/sazman/sounds/34094/) by sazman [CC-BY]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"cat-2":{"filename":"1-34094-B-5.wav","fold":"1","target":"5","category":"cat","esc10":"False","src_file":"34094","take":"B","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-34094-B-5.wav","sha256":"43814c96e62a5c2fbf14b84828cc929e381e53c0add7c1e8211d212e1da97734","credit":"- [1-34094-B.ogg]: clip derived from 070422-cats-sample.wav (http://www.freesound.org/people/sazman/sounds/34094/) by sazman [CC-BY]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"glass_breaking-1":{"filename":"1-20133-A-39.wav","fold":"1","target":"39","category":"glass_breaking","esc10":"False","src_file":"20133","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-20133-A-39.wav","sha256":"db6946133c37fbd4513fe932dbb71e0b077bf7d240beaed339eb6c88e48c903b","credit":"- [1-20133-A.ogg]: clip derived from breaking_glass.wav (http://www.freesound.org/people/gezortenplotz/sounds/20133/) by gezortenplotz [CC-BY]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"glass_breaking-2":{"filename":"1-84536-A-39.wav","fold":"1","target":"39","category":"glass_breaking","esc10":"False","src_file":"84536","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-84536-A-39.wav","sha256":"79efba4c0c54acadc88783e83b428fc9c430db94515b25339a6b5598e2ade0c0","credit":"- [1-84536-A.ogg]: clip derived from glass.wav (http://www.freesound.org/people/cmusounddesign/sounds/84536/) by cmusounddesign [CC-BY]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"car_horn-1":{"filename":"1-17124-A-43.wav","fold":"1","target":"43","category":"car_horn","esc10":"False","src_file":"17124","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-17124-A-43.wav","sha256":"9d5e01e07a386ee36c2e73f40817d132a54d91c936eccc01414280abca5a4209","credit":"- [1-17124-A.ogg]: clip derived from car sharp horn.wav (http://www.freesound.org/people/cognito%20perceptu/sounds/17124/) by cognito perceptu [CC0]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"car_horn-2":{"filename":"1-19026-A-43.wav","fold":"1","target":"43","category":"car_horn","esc10":"False","src_file":"19026","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-19026-A-43.wav","sha256":"67bb2fe5cf138930eb6a6a84bedeb7284093076738cd47ef73cc8b3b8393ff40","credit":"- [1-19026-A.ogg]: clip derived from claxon.wav (http://www.freesound.org/people/han1/sounds/19026/) by han1 [CC-BY]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"engine-1":{"filename":"1-18527-A-44.wav","fold":"1","target":"44","category":"engine","esc10":"False","src_file":"18527","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-18527-A-44.wav","sha256":"83eb3e6065246fd87d1acbbe414aa5bfb7c2081a57c188713cdc017819ffc38f","credit":"- [1-18527-A.ogg]: clip derived from garbage_truck.wav (http://www.freesound.org/people/Corsica_S/sounds/18527/) by Corsica_S [CC-BY]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"engine-2":{"filename":"1-18527-B-44.wav","fold":"1","target":"44","category":"engine","esc10":"False","src_file":"18527","take":"B","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-18527-B-44.wav","sha256":"40f0ab8a2fa68cbdf814f10fc8b2c26916b3dc54c1b0e719d4697ba58c0bdb81","credit":"- [1-18527-B.ogg]: clip derived from garbage_truck.wav (http://www.freesound.org/people/Corsica_S/sounds/18527/) by Corsica_S [CC-BY]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"rain-1":{"filename":"1-17367-A-10.wav","fold":"1","target":"10","category":"rain","esc10":"True","src_file":"17367","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-17367-A-10.wav","sha256":"d732acd2e0c7c40405968010f49f250740c86998fc416eff2ed5330fc8c89dea","credit":"- [1-17367-A.ogg]: clip derived from unlocki door to mod rain.wav (http://www.freesound.org/people/cognito%20perceptu/sounds/17367/) by cognito perceptu [CC0]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"rain-2":{"filename":"1-21189-A-10.wav","fold":"1","target":"10","category":"rain","esc10":"True","src_file":"21189","take":"A","url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/audio/1-21189-A-10.wav","sha256":"5f15ab2d8f6ef9932909b1b75c8ac254cbea46c93b4b2f0f3542abdf09b9abc0","credit":"- [1-21189-A.ogg]: clip derived from Louisiana Rain 1.wav (http://www.freesound.org/people/uberhuberman/sounds/21189/) by uberhuberman [CC0]","source_license_url":"https://raw.githubusercontent.com/karolpiczak/ESC-50/33c8ce9eb2cf0b1c2f8bcf322eb349b6be34dbb6/LICENSE","audio_redistributed":false},"authored_music":{"license":"CC0 authored deterministic melody, environment_fixtures.py"},"authored_impacts":{"license":"CC0 procedural impacts"}},"final_identity":{"type":"identity","enabled":false,"mode":"open-sentences","sentence_mode":true,"open_enrollment":true,"speakers":[{"id":"speaker-1","name":"Christoph","named":true,"reference_ready":true,"voice_variants":1,"admitted":true},{"id":"speaker-2","name":"Speaker 2","named":false,"reference_ready":true,"voice_variants":1,"admitted":true},{"id":"speaker-3","name":"Selma","named":true,"reference_ready":true,"voice_variants":1,"admitted":true}],"users":["Christoph","Speaker 2","Selma"],"threshold":0.38,"conversation_focus":{"enabled":false,"threshold":0.28,"timeout_s":180,"speaker_id":null,"name":null,"primary_id":"speaker-1","participants":[{"id":"speaker-1","name":"Christoph"},{"id":"speaker-2","name":"Speaker 2"},{"id":"speaker-3","name":"Selma"}]},"handover_pending":false,"storage":"Session RAM","prediction":{"name":"Speaker 2","similarity":0.9948},"at":61676.5},"named_speakers_present":true,"error":"AssertionError: Oh, das klingt nach einem treibenden Beat!","source_unchanged":true}