10-generation evolution of prompt + LoRA-strength + sampling to maximize Amusement. Fitness = mean over 8 candidates. Best evolved 0.533 vs prompt-without-LoRA 0.005 (LoRA lift +0.528) vs neutral -0.000; winning λ≈1.75.
Method & reproducibility — how the evolution ran
Base model:laion/moss-tts-local-transformer-4.55b-voice-acting-v2 · LoRA: the matching v3 emotion LoRA (TTS-AGI/moss-emotion-loras-v3, rank 32), applied via PEFT adapter scaling at strength λ. Algorithm: genetic search — population 16, 10 generations, elitism (top 6 carried over). Each generation: every genome is generated best-of-8 and its fitness = the mean target-emotion strength over those 8 candidates (EmoNet score for the emotion, minus 0.25·max(0, WER−0.6) as a light intelligibility guard). Offspring = 60% crossover of two elites + mutation, 40% mutation of one elite. Genome (what evolves): the GENERAL voice description, the SCRIPT delivery cue, the carrier sentence, the LoRA strength λ (seeded 0.75–1.5, mutated within 0.4–2.0), and the sampling params (audio temperature, top_p, top_k). Mutation operators: 35% regenerate the description (new intensity + manner adjectives), 20% swap the carrier sentence, 23% jitter λ by ±0.25/0.5, 12% jitter temperature ±0.1, 10% change top_k. LoRA on vs off: the two LoRA OFF cards are baselines generated with the adapter disabled (λ=0) — a plain neutral prompt, and the best steering prompt without the LoRA. Every LoRA ON hall-of-fame recipe below was generated with the LoRA active at the evolved λ shown on the card. Seed is derived from the emotion name (deterministic).
Baselines — LoRA disabled
NEUTRAL promptLoRA OFF (λ=0)
mean-emo -0.000 · best -0.00 · blend 0.21 · genu 0.05 · WER 0.47
STEER promptLoRA OFF (λ=0)
mean-emo 0.005 · best 0.04 · blend 0.66 · genu 0.18 · WER 0.12
GENERAL: A voice powerfully expressing amusement, a voice bubbling with amusement, on the edge of laughter, playful and delighted, impossible to hide.
SCRIPT: (chuckling, highly amused, barely holding back laughter) "I need to stop by the store and pick up a few things on the way home."
Hall of fame — top 8 evolved recipes (all generated WITH the emotion LoRA active)
#1LoRA ON · λ=1.5 (150%)
mean-emo 0.533 · best 0.59 · blend 0.26 · genu 0.37 · WER 0.68
GENERAL: A voice overwhelmingly expressing amusement, a voice bubbling with amusement, on the edge of laughter, playful and delighted, impossible to hide.
SCRIPT: (chuckling, highly amused, barely holding back laughter) "I can't believe this is happening."
mean-emo 0.484 · best 0.62 · blend 0.16 · genu 0.26 · WER 0.60
GENERAL: A voice viscerally expressing amusement, a voice bubbling with amusement, on the edge of laughter, playful and delighted, raw and unfiltered.
SCRIPT: (chuckling, highly amused, barely holding back laughter) "I need to stop by the store and pick up a few things on the way home."
mean-emo 0.484 · best 0.59 · blend 0.22 · genu 0.32 · WER 0.68
GENERAL: A voice utterly expressing amusement, a voice bubbling with amusement, on the edge of laughter, playful and delighted, building and building.
SCRIPT: (chuckling, highly amused, barely holding back laughter) "There are twelve chairs arranged around the conference table."
mean-emo 0.454 · best 0.59 · blend 0.25 · genu 0.27 · WER 0.65
GENERAL: A voice utterly expressing amusement, a voice bubbling with amusement, on the edge of laughter, playful and delighted, pouring out.
SCRIPT: (chuckling, highly amused, barely holding back laughter) "I need to stop by the store and pick up a few things on the way home."
mean-emo 0.424 · best 0.59 · blend 0.20 · genu 0.23 · WER 0.08
GENERAL: A voice overwhelmingly expressing amusement, a voice bubbling with amusement, on the edge of laughter, playful and delighted, impossible to hide.
SCRIPT: (chuckling, highly amused, barely holding back laughter) "There are twelve chairs arranged around the conference table."
mean-emo 0.514 · best 0.61 · blend 0.18 · genu 0.29 · WER 0.99
GENERAL: A voice intensely expressing amusement, a voice bubbling with amusement, on the edge of laughter, playful and delighted, impossible to hide.
SCRIPT: (chuckling, highly amused, barely holding back laughter) "Please remember to lock the door before you head out."
mean-emo 0.410 · best 0.58 · blend 0.24 · genu 0.30 · WER 0.40
GENERAL: A voice utterly expressing amusement, a voice bubbling with amusement, on the edge of laughter, playful and delighted, pouring out.
SCRIPT: (chuckling, highly amused, barely holding back laughter) "I need to stop by the store and pick up a few things on the way home."
mean-emo 0.442 · best 0.62 · blend 0.45 · genu 0.33 · WER 0.73
GENERAL: A voice powerfully expressing amusement, a voice bubbling with amusement, on the edge of laughter, playful and delighted, in every breath.
SCRIPT: (chuckling, highly amused, barely holding back laughter) "I can't believe this is happening."
