Profile-reward recipe: generate 640 (batched) → ASR-filter+trim → SIDON → keep the top 20% by a per-character quality+profile reward → warm-start FT 3 epochs. Below: the quality range of the selected training set, then example generations after each epoch.
sheep
Characteristic dimensions (auto, from its own voice profile): ▲ S_STRY ▲ S_NARR ▲ S_MONO ▲ R_NASL ▼ EXPL ▼ S_RANT ▼ S_NEWS ▼ S_TECH · emo: Concentration Emotional_Numbness Interest
Training set = top-20% by reward = z(genu)+z(blend)+z(quality)−z(WER)+0.7·z(profile-match). Kept 123/615.
Best 5 of the kept top-20%
R 8.82 · genu 0.63 · blend 0.47 · qual 0.46 · WER 0.00
R 5.63 · genu 0.51 · blend 0.12 · qual 0.46 · WER 0.00
R 5.27 · genu 0.44 · blend 0.21 · qual 0.41 · WER 0.00
R 4.90 · genu 0.22 · blend 0.77 · qual 0.53 · WER 0.59
R 4.59 · genu 0.39 · blend 0.12 · qual 0.50 · WER 0.04
Weakest 5 still inside the top-20% (the selection boundary)
R 1.29 · genu 0.17 · blend 0.14 · qual 0.38 · WER 0.00
R 1.28 · genu 0.14 · blend 0.06 · qual 0.57 · WER 0.00
R 1.27 · genu 0.04 · blend 0.04 · qual 0.61 · WER 0.00
R 1.25 · genu 0.09 · blend 0.06 · qual 0.53 · WER 0.00
R 1.25 · genu 0.09 · blend 0.13 · qual 0.48 · WER 0.00
Fantasy line — refined ep1 / ep2 / ep3
“The old road winds past the ruined tower where the last dragon knights once kept their watch.”
Sci-fi line — refined ep1 / ep2 / ep3
“The station drifted in low orbit while the reactor cooled and the scanners swept the empty dark.”
Neutral line — refined ep1 / ep2 / ep3
“The meeting is scheduled for three o'clock and the report is on the desk by the printer.”