WoodstitchHuman evals for voice AI

MiniMax

4 evaluated voices · 2 use cases

VoicePhone supportNonfiction narration
Expressive narrator40 / 42 · 22–4116 / 42 · 1–38
CaptivatingStoryteller—37 / 42 · 7–42
PatientMan42 / 42 · 34–4240 / 42 · 15–42
Debator41 / 42 · 28–42—

Rank among evaluated voices and 95% rank range per use case; a cell links to that ranking. Selection: up to three stock voices per vendor per use case under the published rule. Nominate a voice →

Share: MiniMax

Share image
Suggested post text
Independent blind listening evaluation. Paid, vetted US listeners.
Methodology · Evaluated audio · Judgment data · Contact