whatmodel.sh / stt / Fish Audio Transcribe 1

Fish Audio Transcribe 1

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

Fish Audio Transcribe 1 is a closed-weights speech-to-text model from Fish Audio. Licence: proprietary.

Catalogue identity notes

Fish Audio's speech-to-text model transcribe-1 on POST https://api.fish.audio/v1/asr. Fish docs: 'transcribe-1 | General transcription of short recordings. Used when the model header is missing or not an exact match'; the newer transcribe-1-pro (changelog 'Transcribe 1 Pro', September 18, 2026) is a separate model. verli ran it in July 2026 without a model header, i.e. transcribe-1. First-party release month not found, so released is left out. Closed weights.

62.2±4.9#19 of 162 speech-to-text modelsfrom 3 benchmark results across 1 source

Score within the Open ASR English short-form WER (March 2026) comparison set only.

$ how it compares

The speech-to-text models ranked either side of Fish Audio Transcribe 1, out of 176 with a score.

#modelscorewhereRAMreleased
20Granite Speech 5.0 470M TurboCTC IBM62±4local~1.6GB2026-08
21MAI-Transcribe-1.5 Microsoft AI62±8api $0.0060/audio min2026-06
22Fish Audio Transcribe 1 Fish Audio62±5api $0.0060/audio min
23Inworld STT-1 Inworld62±6api $0.0025/audio min
24Qwen-Audio-3.0-ASR Alibaba62±4comparison only2026-09

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

$ what it's good at

other tests3 not in a task · 91% of its score
testresultvs the others in its rundifficultycountswhy
verli STT benchmark · LibriSpeech test-clean WER1.7% ✎▃·····▃▅█▃▅▅ beats 8/11avg 3.4%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 48%typical
verli STT benchmark · SPGISpeech 1,200-clip sample WER2.6% ✎▂·····▄▂█▃▂▄ beats 8/11avg 4.6%▇▇▇▇▇▇▇▇▇ 22%typical
verli STT benchmark · Earnings-22 WER (2,741 utt.)10.3% ✎▃·▃▅··▃▃▅▅█▃ beats 12/13avg 16%▇▇▇▇▇▇▇▇ 21%blunt

$ where to run it

Through an API

api · $0.0060/audio min · docs.fish.audio ↗ · price sourceUSD per input audio minute: Fish Audio's Pricing & Rate Limits page, 'Automatic Speech Recognition (ASR) Models' table, lists `transcribe-1` at $0.36 / audio hour (÷60); `transcribe-1-pro` costs the same. Product: Fish Audio Speech to Text API (POST /v1/asr). API model: transcribe-1 (served when the model header is missing). Price basis: published pay-as-you-go list rate. Currency: USD. Fish Audio official docs, observed 2026-10-09.

$ how the score was worked out · verliapp/stt-benchmark · inde… 91%

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
verliapp/stt-benchmark · independent STT accuracy benchmark3▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 91%MIT
prior · pull toward the middle▇▇▇ 9%
weighted average= 62.2

$ 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