{"name":"Speko's Arena — TTS humanness leaderboard","description":"Blind A/B humanness ranking of text-to-speech models vs a real human baseline (100 = statistically indistinguishable). Votes are live blind rounds; the fit also carries a disclosed prior encoding Vapi's public Humanness Index.","license":"CC BY 4.0 — reuse freely with attribution to arena.speko.ai","source":"https://arena.speko.ai/leaderboard","methodology":"https://arena.speko.ai/methodology","as_of":"2026-08-31T22:58:58.065Z","rows":[{"rank":null,"provider":"Human","model":"reference recording","is_human":true,"humanness":100,"votes_live":120,"provisional":false,"latency_ms":null,"price_usd_per_min":null},{"rank":1,"provider":"MiniMax Speech","model":"speech-2.6-hd","is_human":false,"humanness":95,"votes_live":82,"provisional":false,"latency_ms":364,"price_usd_per_min":0.012},{"rank":2,"provider":"xAI Grok Text-to-Speech","model":"tts","is_human":false,"humanness":92,"votes_live":65,"provisional":false,"latency_ms":363,"price_usd_per_min":null},{"rank":3,"provider":"ElevenLabs","model":"eleven_v3","is_human":false,"humanness":89,"votes_live":75,"provisional":false,"latency_ms":526,"price_usd_per_min":0.1},{"rank":4,"provider":"Hume Octave","model":"octave-2","is_human":false,"humanness":88,"votes_live":31,"provisional":false,"latency_ms":613,"price_usd_per_min":0.15},{"rank":5,"provider":"OpenAI Speech","model":"gpt-4o-mini-tts","is_human":false,"humanness":86,"votes_live":71,"provisional":false,"latency_ms":695,"price_usd_per_min":0.015},{"rank":6,"provider":"Chatterbox","model":"0.5B open weights","is_human":false,"humanness":77,"votes_live":10,"provisional":true,"latency_ms":null,"price_usd_per_min":0},{"rank":7,"provider":"Alibaba Qwen-TTS","model":"qwen3-tts-instruct-flash","is_human":false,"humanness":76,"votes_live":65,"provisional":false,"latency_ms":1908,"price_usd_per_min":0.012},{"rank":8,"provider":"Rime","model":"arcanav3","is_human":false,"humanness":73,"votes_live":75,"provisional":false,"latency_ms":2213,"price_usd_per_min":0.02},{"rank":9,"provider":"Inworld TTS","model":"inworld-tts-2","is_human":false,"humanness":71,"votes_live":56,"provisional":false,"latency_ms":129,"price_usd_per_min":0.02},{"rank":10,"provider":"Cartesia","model":"sonic-3.5","is_human":false,"humanness":70,"votes_live":67,"provisional":false,"latency_ms":126,"price_usd_per_min":0.025},{"rank":11,"provider":"Gradium Text-to-Speech","model":"Gradium Text-to-Speech","is_human":false,"humanness":64,"votes_live":23,"provisional":false,"latency_ms":450,"price_usd_per_min":0.028}]}