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.
titan
Characteristic dimensions (auto, from its own voice profile): ▲ S_MONO ▲ S_STRY ▲ S_NARR ▲ ESTH ▼ EXPL ▼ S_RANT ▼ DARC ▼ S_CONV · 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 121/609.
Best 5 of the kept top-20%
R 4.30 · genu 0.13 · blend 0.59 · qual 0.48 · WER 0.00
R 3.91 · genu 0.21 · blend 0.52 · qual 0.28 · WER 0.06
R 3.83 · genu 0.16 · blend 0.79 · qual 0.34 · WER 0.03
R 3.78 · genu 0.15 · blend 0.36 · qual 0.63 · WER 0.00
R 3.63 · genu 0.02 · blend 0.43 · qual 0.71 · WER 0.00
Weakest 5 still inside the top-20% (the selection boundary)
R 1.46 · genu 0.10 · blend 0.17 · qual 0.61 · WER 0.00
R 1.45 · genu 0.00 · blend 0.08 · qual 0.75 · WER 0.00
R 1.45 · genu 0.06 · blend 0.23 · qual 0.55 · WER 0.00
R 1.43 · genu 0.02 · blend 0.07 · qual 0.73 · WER 0.00
R 1.42 · genu 0.14 · blend 0.45 · qual 0.32 · WER 0.08
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.”