Each embedded DramaBox clip is reinterpreted by MOSS-TTS-4.55B voice-acting-v2 (the base checkpoint for our LoRAs): the original audio is fed back as the reference together with its own GENERAL/SCRIPT prompt, and the model regenerates the same line as a batched group of 8 candidates. Candidates are ranked by emotional similarity to the original × inverse WER; the top 3 are shown next to the original. Regenerating and reward-ranking cleans up disfluencies and mistakes while staying faithful to the original emotion.
#56 · “wasted. I kept telling myself it was for the right reasons that it would matter. But what's the point of all the running”
A the audio delivers a, voice at peak biological vigor. Timbre: perfect; standard human baseline; standard. Delivery: casual, conversational, dramatic. Emotion: disappointment, sadness, distress, helplessness, contemplat
● ORIGINAL · 36.63s
the embedded DramaBox clip
✦ TOP 1
reward 0.239 · emo-sim 0.589 · WER 0.595 · 35.92s
“Wasted. I kept telling myself it was for the right reasons that it would matter. But what”
✦ TOP 2
reward 0.191 · emo-sim 0.565 · WER 0.662 · 36.96s
“Wasted. I kept telling myself it was for the right reasons that it would matter. But wha”
✦ TOP 3
reward 0.056 · emo-sim 0.690 · WER 0.919 · 39.04s
“... ...wasted. I kept telling myself it was for the right reasons that it would matter. ”
#58 · “wasted. I kept telling myself it was for the right reasons”
A the audio delivers a, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: storytelling, monologue, narrator. Emotion: disappointment, contemplation, relief, sadness, h
● ORIGINAL · 4.3s
the embedded DramaBox clip
✦ TOP 1
reward 0.456 · emo-sim 0.456 · WER 0.000 · 4.24s
“Wasted. I kept telling myself it was for the right reasons.”
✦ TOP 2
reward 0.000 · emo-sim 0.794 · WER 1.000 · 5.68s
“Wasted. I kept telling myself it was for the right reasons. Wasted. I kept telling myself”
✦ TOP 3
reward 0.000 · emo-sim 0.633 · WER 1.000 · 5.12s
“Wasted. I kept telling myself it was for the right reasons. Wasted. I kept telling myself”
#59 · “that it would matter. But what's the point of all the running if there's nothing left to save?”
A the audio delivers a, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: conversational, playful, casual. Emotion: sadness, disappointment, helplessness, contemplatio
● ORIGINAL · 5.87s
the embedded DramaBox clip
✦ TOP 1
reward 0.875 · emo-sim 0.875 · WER 0.000 · 9.52s
“That it would matter. But what's the point of all the running if there's nothing left to ”
✦ TOP 2
reward 0.000 · emo-sim 0.873 · WER 1.000 · 8.48s
“that it would matter. But what's the point of all the running if there's nothing left to ”
✦ TOP 3
reward 0.000 · emo-sim 0.762 · WER 1.000 · 7.76s
“That it would matter. But what's the point of all the running if there's nothing left to ”
#60 · “They said I was making a difference, a shield,”
A feminine voice, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: casual, conversational, dramatic. Emotion: disappointment, sadness, triumph, pride, relief. Genuine
● ORIGINAL · 4.14s
the embedded DramaBox clip
✦ TOP 1
reward 0.771 · emo-sim 0.771 · WER 0.000 · 4.08s
“They said I was making a difference, a shield.”
✦ TOP 2
reward 0.000 · emo-sim 0.641 · WER 1.000 · 4.64s
“They said I was making a difference a shield They said I was making a difference a shield”
✦ TOP 3
reward 0.000 · emo-sim 0.604 · WER 1.000 · 3.84s
“They said I was making a difference. A shield. They said I was making a difference. A s”
#61 · “a shield for what? For a world that just keeps breaking. I”
A feminine voice, the acoustic signature reflects a fully. Timbre: the high frequencies are distinctly; highly functional but distinctly neutral-cool; lmost completely obliterated by violent. Delivery: dramatic, casual,
● ORIGINAL · 5.57s
the embedded DramaBox clip
✦ TOP 1
reward 0.110 · emo-sim 0.658 · WER 0.833 · 5.76s
“Shield for what? For a world that just keeps breaking. I- Shield for what? For a world”
✦ TOP 2
reward 0.063 · emo-sim 0.756 · WER 0.917 · 7.76s
“A shield for what? For a world that just keeps breaking. A shield for what? For a world”
✦ TOP 3
reward 0.051 · emo-sim 0.608 · WER 0.917 · 7.12s
“A shield for what? For a world that just keeps breaking. A shield for what? For a world”
#62 · “I should have I should have known it wouldn't be enough.”
A feminine voice, voice at peak biological vigor. Timbre: perfect; standard human baseline; standard. Delivery: casual, conversational, monologue. Emotion: disappointment, sadness, contemplation, shame, embarrassment. Ge
● ORIGINAL · 2.91s
the embedded DramaBox clip
✦ TOP 1
reward 0.458 · emo-sim 0.915 · WER 0.500 · 2.24s
“I should have known it wouldn't be enough. I should have known it wouldn't be enough.”
✦ TOP 2
reward 0.454 · emo-sim 0.907 · WER 0.500 · 2.64s
“I should have known it wouldn't be enough. I should have known it wouldn't be enough.”
✦ TOP 3
reward 0.443 · emo-sim 0.887 · WER 0.500 · 2.24s
“I should have known it wouldn't be enough. I should have known it wouldn't be enough.”
#63 · “The echoes, they just keep coming back, like a broken record.”
A the audio delivers a, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: casual, conversational, playful. Emotion: interest, disappointment, amusement, teasing, distr
● ORIGINAL · 13.84s
the embedded DramaBox clip
✦ TOP 1
reward 0.329 · emo-sim 0.329 · WER 0.000 · 6.64s
“The echoes, they just keep coming back, like a broken record.”
✦ TOP 2
reward 0.162 · emo-sim 0.162 · WER 0.000 · 8.32s
“The echoes, they just keep coming back, like a broken record.”
✦ TOP 3
reward 0.047 · emo-sim 0.047 · WER 0.000 · 7.52s
“The echoes, they just keep coming back, like a broken record.”
#64 · “They think they can just dismantle everything I've built,”
A feminine voice, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: conversational, ranting/angry, narrator. Emotion: interest, disappointment, sadness, distress, bitt
● ORIGINAL · 11.17s
the embedded DramaBox clip
✦ TOP 1
reward 0.488 · emo-sim 0.488 · WER 0.000 · 4.48s
“They think they can just dismantle everything I've built.”
✦ TOP 2
reward 0.483 · emo-sim 0.483 · WER 0.000 · 3.76s
“They think they can just dismantle everything I've built.”
✦ TOP 3
reward 0.000 · emo-sim 0.540 · WER 1.200 · 5.12s
“They think they can just dismantle everything I've built. Fuuuugh. They think they can j”
#65 · “erase it? They're wrong. This isn't over, not by a long shot.”
A feminine voice, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: monologue, narrator, dramatic. Emotion: anger, disappointment, impatience and irritability, fear, d
● ORIGINAL · 4.65s
the embedded DramaBox clip
✦ TOP 1
reward 0.476 · emo-sim 0.476 · WER 0.000 · 7.36s
“Erase it? They're wrong. This isn't over, not by a long shot.”
✦ TOP 2
reward 0.000 · emo-sim 0.848 · WER 1.000 · 6.24s
“Erase it. They're wrong. This isn't over, not by a long shot. Erase it. They're wrong.”
✦ TOP 3
reward 0.000 · emo-sim 0.079 · WER 1.071 · 9.36s
“Erouse it, they're wrong. This isn't over, not by a long shot. Erouse it, they're wrong. ”
#66 · “They underestimated the lengths people will go to when they're pushed.”
A the audio delivers a, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: narrator, monologue, formal. Emotion: interest, bitterness, sadness, pride, contempt. Genuine
● ORIGINAL · 3.68s
the embedded DramaBox clip
✦ TOP 1
reward 0.915 · emo-sim 0.915 · WER 0.000 · 5.52s
“They underestimated the lengths people will go to when they're pushed.”
✦ TOP 2
reward 0.879 · emo-sim 0.879 · WER 0.000 · 4.32s
“They underestimated the lengths people will go to when they're pushed.”
✦ TOP 3
reward 0.855 · emo-sim 0.855 · WER 0.000 · 4.48s
“They underestimated the lengths people will go to when they're pushed.”
#67 · “They think they've won, they haven't seen anything yet.”
A feminine voice, the acoustic signature reflects a fully. Timbre: the high frequencies are distinctly; standard human baseline; standard. Delivery: monologue, dramatic, conversational. Emotion: disappointment, emotional
● ORIGINAL · 3.62s
the embedded DramaBox clip
✦ TOP 1
reward 0.619 · emo-sim 0.619 · WER 0.000 · 3.52s
“They think they've won. They haven't seen anything yet.”
✦ TOP 2
reward 0.482 · emo-sim 0.482 · WER 0.000 · 3.76s
“They think they've won. They haven't seen anything yet.”
✦ TOP 3
reward 0.321 · emo-sim 0.321 · WER 0.000 · 2.96s
“They think they've won, they haven't seen anything yet.”
#68 · “I'll make them regret this, every single one of them [yawn]. be helpless,”
A the audio delivers a, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: playful, conversational, casual. Emotion: interest, malevolence malice, teasing, hope enthusi
● ORIGINAL · 12.03s
the embedded DramaBox clip
✦ TOP 1
reward 0.112 · emo-sim 0.521 · WER 0.786 · 6.96s
“I'll make them regret this. Every single one of them. I'll make them regret this. Every”
✦ TOP 2
reward -0.000 · emo-sim -0.048 · WER 1.071 · 7.92s
“I'll make them regret this, every single one of them. Meow-plus. I'll make them regret t”
✦ TOP 3
reward -0.000 · emo-sim -0.082 · WER 1.214 · 12.32s
“I'll make them regret this, every single one of them. It'd be helpless. I'll make them r”
#69 · “this is retribution, and it will be thorough. They'll wish they'd left it alone.”
A the audio delivers a, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: monologue, narrator, storytelling. Emotion: interest, triumph, concentration, relief, emotion
● ORIGINAL · 5.33s
the embedded DramaBox clip
✦ TOP 1
reward 0.096 · emo-sim 0.514 · WER 0.812 · 11.04s
“This is Retribution. This is Retribution.”
✦ TOP 2
reward 0.024 · emo-sim 0.377 · WER 0.938 · 5.68s
“This is retribution, and it will be thorough. Though wish they'd left it alone. This is r”
✦ TOP 3
reward 0.000 · emo-sim 0.430 · WER 1.000 · 5.6s
“This is retribution. And it will be thorough. They'll wish they'd left it alone. This i”
#70 · “It's fine, just a bit warm. I've felt this before many times. This seemed it's a long one, but they'll notice. They alwa”
A the audio delivers a, the acoustic profile is heavily defined. Timbre: the treble frequencies are significantly; highly pleasant; the acoustic signal is heavily. Delivery: monologue, whisper-talk, asmr. Emotion: longin
● ORIGINAL · 10.97s
the embedded DramaBox clip
✦ TOP 1
reward 0.026 · emo-sim 0.721 · WER 0.964 · 10.96s
“It's fine. Just a bit warm. I felt this before many times. This seemed it's a long one, ”
✦ TOP 2
reward 0.023 · emo-sim 0.656 · WER 0.964 · 12.24s
“It's fine, just a bit warm. I felt this before many times. This seemed it's a long one, ”
✦ TOP 3
reward 0.022 · emo-sim 0.615 · WER 0.964 · 14.48s
“It's fine. Just a bit warm. I felt this before many times. This seemed it's a long one, bu”
#71 · “It's fine, just a bit warm. I've felt this before”
A the audio delivers a, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: monologue, whisper-talk, storytelling. Emotion: relief, emotional numbness, sadness, fatigue
● ORIGINAL · 3.91s
the embedded DramaBox clip
✦ TOP 1
reward 0.098 · emo-sim 0.589 · WER 0.833 · 4.4s
“It's fine, just a bit warm. Felt this before. It's fine, just a bit warm. Felt this bef”
✦ TOP 2
reward 0.000 · emo-sim 0.553 · WER 1.000 · 4.24s
“It's fine. Just a bit warm. I've felt this before. It's fine. Just a bit warm. I've felt t”
✦ TOP 3
reward 0.000 · emo-sim 0.701 · WER 1.000 · 3.92s
“It's fine, just a bit warm. I've felt this before. It's fine, just a bit warm. I've fel”
#72 · “many times. This seemed it's a long one, but they'll notice. They always do.”
A the audio delivers a, the acoustic profile is heavily defined. Timbre: the acoustic signal is completely; highly pleasant; the acoustic signal is heavily. Delivery: monologue, whisper-talk, asmr. Emotion: longing, cont
● ORIGINAL · 7.06s
the embedded DramaBox clip
✦ TOP 1
reward 0.000 · emo-sim 0.591 · WER 1.000 · 9.36s
“Many times. This seemed it's a long one, but they'll notice. They always do. Many times”
✦ TOP 2
reward 0.000 · emo-sim 0.293 · WER 1.000 · 9.2s
“Many times, this seemed it's a long one, but they'll notice. They always do. Many times,”
✦ TOP 3
reward 0.000 · emo-sim 0.566 · WER 1.000 · 7.36s
“Many times this seemed it's a long one, but they'll notice They always do many times this”
#73 · “always there and there I would say great the glass tile left a nabbed way. That cracked tile made the thing spiked I ain”
A feminine voice, voice at peak biological vigor. Timbre: perfect; standard human baseline; standard. Delivery: casual, conversational, monologue. Emotion: interest, amusement, teasing, infatuation, contentment. Genuinen
● ORIGINAL · 32.7s
the embedded DramaBox clip
✦ TOP 1
reward 0.128 · emo-sim 0.408 · WER 0.686 · 15.52s
“Always there, and there I would say, great, The glass tile left a nabbed way That cracked”
✦ TOP 2
reward 0.078 · emo-sim 0.361 · WER 0.784 · 30.96s
“Always there and there. High quality studio recording, no background noise. It cracked t”
✦ TOP 3
reward 0.024 · emo-sim 0.617 · WER 0.961 · 44.0s
“Always there and there I would say, great the glass tile left a nabbed way. That cracked ”
#74 · “always there and there I would say great the glass tile left a nabbed way. That cracked tile made the thing spiked I ain”
A the audio delivers a, the acoustic signature reflects a fully. Timbre: perfect; standard human baseline; standard. Delivery: storytelling, monologue, narrator. Emotion: interest, confusion, fatigue exhaustion, impatien
● ORIGINAL · 12.93s
the embedded DramaBox clip
✦ TOP 1
reward 0.008 · emo-sim 0.068 · WER 0.889 · 20.24s
“Always there, and there I would say, Great the glass tile left a nabbed way. That cracke”
✦ TOP 2
reward 0.001 · emo-sim 0.026 · WER 0.978 · 2.4s
“Who is there? Who is there?”
✦ TOP 3
reward -0.000 · emo-sim -0.039 · WER 1.000 · 18.4s
“Always there, and there I would say, great, the glass tile left a nabbed way. That cracke”
#75 · “time.”
A the audio delivers a, voice at peak biological vigor. Timbre: the treble frequencies are significantly; standard human baseline; standard. Delivery: monologue, formal, asmr. Emotion: thankfulness gratitude, emotional n
● ORIGINAL · 0.46s
the embedded DramaBox clip
✦ TOP 1
reward 0.000 · emo-sim 0.914 · WER 1.000 · 1.28s
“Time. Time.”
✦ TOP 2
reward 0.000 · emo-sim 0.417 · WER 7.000 · 2.8s
“Ha ha ha! Time! Ha ha ha! Time!”
✦ TOP 3
reward 0.000 · emo-sim 0.876 · WER 1.000 · 1.28s
“Time. Time.”
#76 · “Funny how you remember smells like like the Christmas market [laughter]. Take cinnamon.”
A feminine voice, the acoustic signature reflects a fully. Timbre: the acoustic signal is completely; highly pleasant; standard. Delivery: casual, monologue, conversational. Emotion: disgust, interest, sexual lust, amuse
● ORIGINAL · 5.9s
the embedded DramaBox clip
✦ TOP 1
reward 0.163 · emo-sim 0.706 · WER 0.769 · 5.12s
“Funny how you remember smells like, like the Christmas market. Funny how you remember sme”
✦ TOP 2
reward 0.083 · emo-sim 0.360 · WER 0.769 · 11.84s
“Funny how you remember smells like like the Christmas market Funny how you remember smell”
✦ TOP 3
reward 0.041 · emo-sim 0.177 · WER 0.769 · 11.84s
“Funny how you remember smells like the Christmas market. Funny how you remember smells li”
#77 · “You always made this for me [throatclearing]. I used to sit right here on this chair and he”
A feminine voice, the acoustic signature reflects a fully. Timbre: the high frequencies are distinctly; standard human baseline; standard. Delivery: conversational, casual, storytelling. Emotion: affection, thankfulness
● ORIGINAL · 4.66s
the embedded DramaBox clip
✦ TOP 1
reward 0.190 · emo-sim 0.488 · WER 0.611 · 10.32s
“You always made this for me. You always made this for me.”
✦ TOP 2
reward 0.151 · emo-sim 0.452 · WER 0.667 · 10.32s
“You always made this for me. I used to sit right here. You always made this for me. I u”
✦ TOP 3
reward 0.030 · emo-sim 0.548 · WER 0.944 · 7.76s
“You always made this for me. I used to sit right here on this chair and he, you always m”
#78 · “you'd come in and just watch me. It felt so safe then.”
A feminine voice, voice at peak biological vigor. Timbre: the high frequencies are distinctly; standard human baseline; standard. Delivery: casual, conversational, storytelling. Emotion: relief, affection, fear, interest
● ORIGINAL · 3.33s
the embedded DramaBox clip
✦ TOP 1
reward 0.837 · emo-sim 0.837 · WER 0.000 · 4.4s
“You'd come in and just watch me. It felt so safe then.”
✦ TOP 2
reward 0.783 · emo-sim 0.783 · WER 0.000 · 4.72s
“You'd come in and just watch me. It felt so safe then.”
✦ TOP 3
reward 0.000 · emo-sim 0.713 · WER 1.000 · 4.32s
“You'd come in and just watch me. It felt so safe then. You'd come in and just watch me. ”
#79 · “Now tomorrow it'll just be a room empty, just a room.”
A feminine voice, the acoustic signature reflects a fully. Timbre: perfect; highly pleasant; standard. Delivery: whisper-talk, narrator, asmr. Emotion: emotional numbness, sadness, disappointment, teasing, helplessness.
● ORIGINAL · 5.42s
the embedded DramaBox clip
✦ TOP 1
reward 0.300 · emo-sim 0.600 · WER 0.500 · 5.84s
“Now tomorrow, it'll just be a room empty. Now tomorrow, it'll just be a room empty.”
✦ TOP 2
reward 0.000 · emo-sim 0.741 · WER 1.000 · 6.56s
“Now tomorrow, it'll just be a room empty. Just a room. Now tomorrow, it'll just be a room”
✦ TOP 3
reward 0.000 · emo-sim 0.532 · WER 1.000 · 7.04s
“Now tomorrow, it'll just be a room empty. Just... A room. Now tomorrow, it'll just be a”
#80 · “in loyal, outproofed, exlobed, fell horse, and so”
A feminine voice, the acoustic signal is dominated by. Timbre: high-frequency energy is present but; highly pleasant; standard. Delivery: monologue, asmr, whisper-talk. Emotion: interest, concentration, triumph, pride, s
● ORIGINAL · 5.21s
the embedded DramaBox clip
✦ TOP 1
reward 0.611 · emo-sim 0.698 · WER 0.125 · 8.4s
“In loyal outproved, exlobed, fell horse and so.”
✦ TOP 2
reward 0.500 · emo-sim 0.666 · WER 0.250 · 5.44s
“In loyal, out-proofed, exlobed, fell horse and so.”
✦ TOP 3
reward 0.000 · emo-sim 0.701 · WER 1.375 · 6.88s
“In loyal, out-proofed, exlobed, fell horse. And so, in loyal, out-proofed, exlobed, fell ”