10-generation evolution of prompt + LoRA-strength + sampling to maximize Shame. Fitness = mean over 8 candidates. Best evolved 0.099 vs prompt-without-LoRA 0.001 (LoRA lift +0.098) vs neutral 0.002; winning λ≈0.5.
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 0.002 · best 0.00 · blend 0.21 · genu 0.05 · WER 0.47
STEER promptLoRA OFF (λ=0)
mean-emo 0.001 · best 0.00 · blend 0.33 · genu 0.09 · WER 0.06
GENERAL: A voice overwhelmingly expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, in every breath.
SCRIPT: (with cringing shame and self-reproach) "There are twelve chairs arranged around the conference table."
Hall of fame — top 8 evolved recipes (all generated WITH the emotion LoRA active)
#1LoRA ON · λ=1.25 (125%)
mean-emo 0.099 · best 0.25 · blend 0.89 · genu 0.32 · WER 0.56
GENERAL: A voice deeply expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, building and building.
SCRIPT: (with cringing shame and self-reproach) "Look at me. Look at what this has become."
mean-emo 0.104 · best 0.33 · blend 0.89 · genu 0.30 · WER 0.62
GENERAL: A voice unmistakably expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, in every breath.
SCRIPT: (with cringing shame and self-reproach) "Look at me. Look at what this has become."
mean-emo 0.079 · best 0.28 · blend 0.99 · genu 0.28 · WER 0.50
GENERAL: A voice viscerally expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, in every breath.
SCRIPT: (with cringing shame and self-reproach) "Look at me. Look at what this has become."
mean-emo 0.084 · best 0.24 · blend 0.83 · genu 0.25 · WER 0.62
GENERAL: A voice intensely expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, building and building.
SCRIPT: (with cringing shame and self-reproach) "Look at me. Look at what this has become."
mean-emo 0.075 · best 0.28 · blend 0.81 · genu 0.27 · WER 0.50
GENERAL: A voice powerfully expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, with every word.
SCRIPT: (with cringing shame and self-reproach) "Look at me. Look at what this has become."
mean-emo 0.112 · best 0.25 · blend 0.75 · genu 0.31 · WER 0.75
GENERAL: A voice unmistakably expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, in every breath.
SCRIPT: (with cringing shame and self-reproach) "Look at me. Look at what this has become."
mean-emo 0.081 · best 0.31 · blend 0.57 · genu 0.31 · WER 0.62
GENERAL: A voice viscerally expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, building and building.
SCRIPT: (with cringing shame and self-reproach) "I need to stop by the store and pick up a few things on the way home."
mean-emo 0.095 · best 0.20 · blend 0.86 · genu 0.30 · WER 0.68
GENERAL: A voice utterly expressing shame, a small, shame-filled voice, cringing and self-reproaching, unable to look up, impossible to hide.
SCRIPT: (with cringing shame and self-reproach) "Look at me. Look at what this has become."
