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.
voice-28
Characteristic dimensions (auto, from its own voice profile): ▲ S_STRY ▲ S_MONO ▲ S_NARR ▲ ESTH ▼ S_RANT ▼ EXPL ▼ S_NEWS ▼ S_TECH · emo: Concentration Interest Awe
Training set = top-20% by reward = z(genu)+z(blend)+z(quality)−z(WER)+0.7·z(profile-match). Kept 127/639.
Best 5 of the kept top-20%
R 6.08 · genu 0.21 · blend 0.85 · qual 0.58 · WER 0.00
R 5.69 · genu 0.27 · blend 0.53 · qual 0.55 · WER 0.00
R 5.38 · genu 0.16 · blend 0.79 · qual 0.53 · WER 0.00
R 4.92 · genu 0.30 · blend 0.78 · qual 0.38 · WER 0.00
R 4.54 · genu 0.24 · blend 0.37 · qual 0.63 · WER 0.00
Weakest 5 still inside the top-20% (the selection boundary)
R 1.30 · genu 0.11 · blend 0.19 · qual 0.61 · WER 0.00
R 1.29 · genu 0.12 · blend 0.14 · qual 0.53 · WER 0.00
R 1.27 · genu 0.06 · blend 0.10 · qual 0.61 · WER 0.00
R 1.25 · genu 0.03 · blend 0.14 · qual 0.72 · WER 0.00
R 1.25 · genu 0.00 · blend 0.14 · qual 0.71 · 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.”