whatmodel.sh / tts / Octave TTS (Octave 1)

Octave TTS (Octave 1)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Octave TTS (Octave 1) is a closed-weights text-to-speech model from Hume AI, released February 2025. Licence: proprietary.

57.9±3.9#46 of 118 text-to-speech modelsfrom 4 benchmark results across 3 sources

Score within the Vapi Humanness · Bradley–Terry rating, 2026-09-25 snapshot comparison set only.

$ how it compares

The text-to-speech models ranked either side of Octave TTS (Octave 1), out of 148 with a score.

#modelscorewhereRAMreleased
54MGM-Omni-TTS 2B (0927) MGM-Omni (CUHK)59±4comparison only2025-09
55CosyVoice v3 Plus Alibaba Cloud58±4api $26.00/M chars2025-09
56Octave TTS (Octave 1) Hume AI58±4api $50.00/M chars2025-02
57Sonic 3 Cartesia58±2api $65.00/M chars2025-10
58GPT-4o mini TTS OpenAI58±2comparison only2025-03

see all text-to-speech models · find the ones that fit your machine

$ what it's good at

other tests4 not in a task · 94% of its score
testresultvs the others in its rundifficultycountswhy
EmergentTTS · 6-category preference46.8% ✎▅▅▃▃█▅▅▃·▃·▅ beats 13/16hard · avg 30%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 50%at its level
InstructTTSEval EN · mean instruction-following score (APS, DSD, RP)71.1% ✎▄····█··▄··▄ beats 2/4easy · avg 66%▇▇▇▇▇▇▇▇ 21%at its level
EmergentTTS · v1 paper automatic MOS (WV-MOS)4.18 / 5 ✎▄·█·█▄····█▄ beats 6/8easy · avg 3.93 / 5▇▇▇▇▇ 12%blunt
EmergentTTS · 6-category WER12.8% ✎▂▄··▂▄▄··█▄▆ beats 10/16avg 16%▇▇▇▇ 11%blunt

$ where to run it

Through an API

api · $50.00/M chars · www.hume.ai ↗ · price sourceUSD per 1M characters: Hume lists Octave 1 additional-character usage at $0.05 per 1,000 characters on the Business plan ($500/month; lowest published self-serve rate); Creator $0.15, Pro $0.12, Scale $0.10 per 1,000 on lower plans (x1000). Product: Octave text-to-speech API. API model: Octave 1. Settings: Business-plan usage-based overage. Price basis: published list rate, per-1K to per-1M scaling only. Exclusions: monthly plan fee, included characters, EVI. Currency: USD. Hume official pricing page, observed 2026-09-25.

$ how the score was worked out · EmergentTTS-Eval six-category… 61%, InstructTTSEval: instruction-… 21%, EmergentTTS-Eval v1 paper (ar… 12%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
EmergentTTS-Eval six-category leaderboard2▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 61%Apache-2.0
InstructTTSEval: instruction-following TTS benchmark (arXiv:2506.16381v1)1▇▇▇▇▇▇ 21%CC BY 4.0
EmergentTTS-Eval v1 paper (arXiv:2505.23009v1) · Table 1 automatic MOS1▇▇▇ 12%CC BY 4.0
prior · pull toward the middle▇▇ 6%
weighted average= 57.9

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json