← all emotions
Pleasure Ecstasy
Best adapter: rank 32 , checkpoint seen 3336 · EmoNet-z -0.59 (baseline -0.74) · blend 0.681 · genu 0.203 · speaker-sim 0.179.
Instruction prompt: GENERAL: A voice that is adult, with extremely carrying bliss and 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 -0.48 ±0.21 · bl 0.2 · gn 0.1 ▲ high -0.13
▼ low -0.74
z mean -0.74 ±0.01 · bl 0.3 · gn 0.1 ▲ high -0.71
▼ low -0.74
z mean -0.49 ±0.43 · bl 0.5 · gn 0.2 ▲ high +0.38
▼ low -0.74
with emo reference
z mean +0.10 ±0.78 · bl 0.5 · gn 0.2 ▲ high +1.54
▼ low -0.68
z mean -0.51 ±0.30 · bl 0.4 · gn 0.2 ▲ high +0.01
▼ low -0.73
z mean -0.08 ±0.74 · bl 0.4 · gn 0.2 ▲ high +1.29
▼ low -0.74
Sentence 2
no reference
z mean +0.10 ±0.44 · bl 0.7 · gn 0.2 ▲ high +0.51
▼ low -0.73
z mean +0.04 ±0.49 · bl 0.4 · gn 0.1 ▲ high +0.68
▼ low -0.61
z mean -0.51 ±0.15 · bl 0.6 · gn 0.2 ▲ high -0.32
▼ low -0.72
with emo reference
z mean +0.53 ±0.64 · bl 0.5 · gn 0.2 ▲ high +0.96
▼ low -0.73
z mean +0.51 ±0.44 · bl 0.4 · gn 0.2 ▲ high +1.26
▼ low +0.05
z mean -0.39 ±0.22 · bl 0.5 · gn 0.2 ▲ high +0.04
▼ low -0.54
Sentence 3
no reference
z mean -0.74 ±0.00 · bl 0.2 · gn 0.1 ▲ high -0.74
▼ low -0.74
z mean -0.60 ±0.18 · bl 0.4 · gn 0.1 ▲ high -0.27
▼ low -0.74
z mean -0.72 ±0.04 · bl 0.6 · gn 0.2 ▲ high -0.63
▼ low -0.74
with emo reference
z mean -0.74 ±0.00 · bl 0.2 · gn 0.1 ▲ high -0.73
▼ low -0.74
z mean -0.72 ±0.04 · bl 0.4 · gn 0.1 ▲ high -0.64
▼ low -0.74
z mean -0.69 ±0.09 · bl 0.4 · gn 0.2 ▲ high -0.52
▼ low -0.74
Sentence 4
no reference
z mean +0.41 ±0.43 · bl 0.1 · gn 0.1 ▲ high +0.75
▼ low -0.38
z mean -0.32 ±0.17 · bl 0.5 · gn 0.1 ▲ high -0.11
▼ low -0.62
z mean +1.01 ±0.17 · bl 0.3 · gn 0.2 ▲ high +1.29
▼ low +0.78
with emo reference
z mean +0.83 ±0.35 · bl 0.2 · gn 0.2 ▲ high +1.07
▼ low +0.13
z mean +0.01 ±0.67 · bl 0.4 · gn 0.2 ▲ high +1.19
▼ low -0.64
z mean +1.01 ±0.15 · bl 0.5 · gn 0.3 ▲ high +1.17
▼ low +0.82
Sentence 5
no reference
z mean +0.11 ±0.59 · bl 0.2 · gn 0.2 ▲ high +1.10
▼ low -0.62
z mean -0.58 ±0.28 · bl 0.5 · gn 0.1 ▲ high -0.02
▼ low -0.74
z mean -0.15 ±0.34 · bl 0.5 · gn 0.2 ▲ high +0.30
▼ low -0.65
with emo reference
z mean +0.23 ±0.46 · bl 0.2 · gn 0.1 ▲ high +1.10
▼ low -0.27
z mean -0.16 ±0.27 · bl 0.5 · gn 0.2 ▲ high +0.22
▼ low -0.59
z mean -0.27 ±0.60 · bl 0.5 · gn 0.3 ▲ high +0.87
▼ low -0.74