← all emotions
Teasing
Best adapter: rank 16 , checkpoint seen 5000 · EmoNet-z +0.55 (baseline -0.605) · blend 0.417 · genu 0.259 · speaker-sim 0.332.
Instruction prompt: GENERAL: A voice that is adult, with extremely carrying teasing 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 -0.60 ±0.00 · bl 0.1 · gn 0.1 ▲ high -0.60
▼ low -0.60
z mean +1.36 ±1.54 · bl 0.2 · gn 0.1 ▲ high +4.00
▼ low -0.28
z mean +0.45 ±0.89 · bl 0.4 · gn 0.1 ▲ high +1.87
▼ low -0.60
with emo reference
z mean -0.60 ±0.00 · bl 0.2 · gn 0.1 ▲ high -0.60
▼ low -0.60
z mean +1.69 ±1.60 · bl 0.1 · gn 0.2 ▲ high +4.00
▼ low -0.60
z mean +0.12 ±0.89 · bl 0.1 · gn 0.1 ▲ high +1.22
▼ low -0.60
Sentence 2
no reference
z mean +0.28 ±0.77 · bl 0.6 · gn 0.1 ▲ high +1.28
▼ low -0.60
z mean +1.74 ±1.17 · bl 0.3 · gn 0.1 ▲ high +4.00
▼ low +0.65
z mean +1.31 ±1.45 · bl 0.5 · gn 0.1 ▲ high +4.00
▼ low -0.14
with emo reference
z mean +1.22 ±0.91 · bl 0.5 · gn 0.2 ▲ high +2.15
▼ low -0.53
z mean +2.41 ±0.32 · bl 0.1 · gn 0.1 ▲ high +2.75
▼ low +1.96
z mean +0.43 ±0.98 · bl 0.3 · gn 0.2 ▲ high +1.85
▼ low -0.60
Sentence 3
no reference
z mean -0.60 ±0.00 · bl 0.2 · gn 0.1 ▲ high -0.60
▼ low -0.60
z mean +0.38 ±0.89 · bl 0.3 · gn 0.1 ▲ high +1.66
▼ low -0.60
z mean -0.60 ±0.00 · bl 0.5 · gn 0.1 ▲ high -0.60
▼ low -0.60
with emo reference
z mean -0.40 ±0.40 · bl 0.1 · gn 0.3 ▲ high +0.40
▼ low -0.60
z mean +2.25 ±0.99 · bl 0.1 · gn 0.1 ▲ high +3.30
▼ low +0.91
z mean +0.07 ±0.97 · bl 0.3 · gn 0.3 ▲ high +1.89
▼ low -0.60
Sentence 4
no reference
z mean +0.58 ±1.68 · bl 0.1 · gn 0.1 ▲ high +3.69
▼ low -0.60
z mean -0.42 ±0.37 · bl 0.3 · gn 0.1 ▲ high +0.32
▼ low -0.60
z mean +2.73 ±0.62 · bl 0.4 · gn 0.2 ▲ high +3.56
▼ low +1.72
with emo reference
z mean +0.61 ±0.62 · bl 0.1 · gn 0.2 ▲ high +1.08
▼ low -0.60
z mean +0.93 ±1.84 · bl 0.1 · gn 0.1 ▲ high +3.79
▼ low -0.60
z mean +1.30 ±1.33 · bl 0.1 · gn 0.3 ▲ high +2.94
▼ low -0.31
Sentence 5
no reference
z mean -0.30 ±0.49 · bl 0.1 · gn 0.1 ▲ high +0.67
▼ low -0.60
z mean +0.37 ±0.66 · bl 0.3 · gn 0.1 ▲ high +1.27
▼ low -0.60
z mean -0.13 ±0.95 · bl 0.4 · gn 0.2 ▲ high +1.76
▼ low -0.60
with emo reference
z mean +0.04 ±0.65 · bl 0.2 · gn 0.1 ▲ high +1.10
▼ low -0.60
z mean +0.37 ±0.86 · bl 0.0 · gn 0.1 ▲ high +1.71
▼ low -0.60
z mean -0.23 ±0.47 · bl 0.1 · gn 0.2 ▲ high +0.50
▼ low -0.60