The human benchmark for speech AI.

Woodstitch runs blinded listening studies on text-to-speech models with paid human raters. Every study states its method, its numbers and what it does not show.

For AI teams

Testing a release candidate? We run blinded human evaluations on your models, voices and scripts. Private by default — paid work never changes a public result.

Evaluate your model

For listeners

Get paid to compare short audio clips and pick the one that sounds better. Work whenever you want.

Become a rater

Research report · 11 November 2026

Which ElevenLabs voice is best for audiobooks?

1,122 blinded judgments · 125 paid listeners · 32 Voice Library voices · one 30-second nonfiction passage

66%Marie, first
95% CI 55–76

46%professional vs instant clones
studio tier no better

54%louder clip wins
15 dB spread between voices

The library's own signals do not predict preference: the professional badge and popularity told listeners nothing, voices tagged "old" were disfavoured, gender made no difference, and level did. Listen to all 32. Read the report →

How every study is run — listeners, payment, blinding, statistics, publication rules — is in the platform methodology.