← all emotions
Disgust
Best adapter: rank 64 , checkpoint seen 5000 · EmoNet-z -0.22 (baseline -0.528) · blend 0.359 · genu 0.158 · speaker-sim 0.145.
Instruction prompt: GENERAL: A voice that is adult; extremely carrying abhorrence; genuine in delivery.
SCRIPT:
Five neutral sentences, each in English / Deutsch / Français, no reference vs with emotion reference (a same-emotion training clip supplies the voice; the prompt applies the emotion).
Each cell = 5 seeds. We show mean±std of the emotion score and play the highest and lowest seed (top = strongest emotion, bottom = weakest), since a fixed seed pins the sampling and single clips vary.
Sentence 1
no reference
z mean -0.53 ±0.00 · bl 0.2 · gn 0.1 ▲ high -0.53
▼ low -0.53
z mean +0.86 ±1.08 · bl 0.2 · gn 0.1 ▲ high +1.91
▼ low -0.52
z mean -0.41 ±0.15 · bl 0.4 · gn 0.1 ▲ high -0.13
▼ low -0.54
with emo reference
z mean -0.53 ±0.00 · bl 0.3 · gn 0.1 ▲ high -0.53
▼ low -0.53
z mean +0.31 ±0.55 · bl 0.2 · gn 0.1 ▲ high +1.04
▼ low -0.48
z mean -0.47 ±0.12 · bl 0.3 · gn 0.1 ▲ high -0.22
▼ low -0.54
Sentence 2
no reference
z mean -0.53 ±0.00 · bl 0.7 · gn 0.1 ▲ high -0.53
▼ low -0.53
z mean -0.21 ±0.30 · bl 0.3 · gn 0.1 ▲ high +0.23
▼ low -0.57
z mean -0.51 ±0.03 · bl 0.5 · gn 0.2 ▲ high -0.47
▼ low -0.53
with emo reference
z mean -0.52 ±0.02 · bl 0.7 · gn 0.1 ▲ high -0.48
▼ low -0.53
z mean +0.06 ±0.69 · bl 0.3 · gn 0.1 ▲ high +1.32
▼ low -0.52
z mean -0.52 ±0.03 · bl 0.3 · gn 0.2 ▲ high -0.47
▼ low -0.53
Sentence 3
no reference
z mean -0.53 ±0.00 · bl 0.2 · gn 0.2 ▲ high -0.52
▼ low -0.53
z mean -0.40 ±0.18 · bl 0.3 · gn 0.1 ▲ high -0.08
▼ low -0.53
z mean -0.30 ±0.30 · bl 0.5 · gn 0.2 ▲ high +0.21
▼ low -0.53
with emo reference
z mean -0.53 ±0.00 · bl 0.2 · gn 0.1 ▲ high -0.52
▼ low -0.53
z mean +0.07 ±0.75 · bl 0.2 · gn 0.1 ▲ high +1.19
▼ low -0.53
z mean -0.45 ±0.16 · bl 0.5 · gn 0.2 ▲ high -0.12
▼ low -0.55
Sentence 4
no reference
z mean -0.53 ±0.00 · bl 0.1 · gn 0.1 ▲ high -0.53
▼ low -0.53
z mean -0.29 ±0.42 · bl 0.3 · gn 0.1 ▲ high +0.54
▼ low -0.56
z mean +0.02 ±0.62 · bl 0.5 · gn 0.2 ▲ high +0.83
▼ low -0.54
with emo reference
z mean -0.53 ±0.00 · bl 0.2 · gn 0.1 ▲ high -0.53
▼ low -0.53
z mean -0.45 ±0.16 · bl 0.4 · gn 0.1 ▲ high -0.13
▼ low -0.54
z mean +0.37 ±0.99 · bl 0.3 · gn 0.2 ▲ high +1.83
▼ low -0.53
Sentence 5
no reference
z mean -0.32 ±0.26 · bl 0.1 · gn 0.1 ▲ high +0.14
▼ low -0.53
z mean +2.25 ±0.61 · bl 0.4 · gn 0.1 ▲ high +2.80
▼ low +1.31
z mean +1.47 ±1.07 · bl 0.4 · gn 0.1 ▲ high +2.57
▼ low -0.46
with emo reference
z mean -0.16 ±0.74 · bl 0.3 · gn 0.1 ▲ high +1.31
▼ low -0.53
z mean +1.32 ±1.51 · bl 0.4 · gn 0.1 ▲ high +3.31
▼ low -0.57
z mean +1.04 ±0.98 · bl 0.3 · gn 0.1 ▲ high +2.31
▼ low -0.53