Turing TTS benchmark: best voice AI fools listeners only ~10% of the time
omarsar0 · x · 2026-10-31
Turing introduces Turing TTS, an evaluation measuring how often human listeners mistake AI-generated speech for real human speech.
- Testing leading TTS models against real human speech, even the best model fooled listeners only 10% of the time
- Key insight: the best AI voices are now hard to tell apart from each other — not from humans
- The authors argue voice AI's next breakthrough may be better measurement, not another #1 model; Voice Arena is pushing the same conversation
More from Research
- New preprint: closed-loop ephys successfully trains in vitro neuron cultures to change behavior — DoctorJosh · 2026-10-31
- Forethought: first movers could lock rivals out of space for $49-73B premium — willmacaskill · 2026-10-31
- Correction: newer model gets worse on the extreme-hard tail, pass@5 crossover predicted — teortaxesTex · 2026-10-31
- Meta demos titanRL, TorchTitan's RL library for LLMs, teaching Qwen Connect 4 — ezyang · 2026-10-31
- Biohub's Remo-ID spots sepsis from blood cell images, 43/47 correct in early study — alex_peys · 2026-10-31
- 240-Year-Old Voting Paradox Gets a Shape: It's a Klein Bottle — stevenstrogatz · 2026-10-31