whatmodel.sh / stt / data2vec Audio Base 960h

data2vec Audio Base 960h

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

data2vec Audio Base 960h is an open-weights speech-to-text model from Meta, released February 2022. Its published size is 0.093B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact checkpoint facebook/data2vec-audio-base-960h (CTC fine-tuned on LibriSpeech 960h). Params: fp32 pytorch_model.bin 372,850,161 bytes / 4; released: arXiv 2202.03555v1 month.

4.4±3.5#159 of 162 speech-to-text modelsfrom 3 benchmark results across 2 sources

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 data2vec Audio Base 960h, out of 176 with a score, then more from Meta.

#modelscorewhereRAMreleased
171data2vec Audio Large 960h Meta13±4local~2GB2022-04
172wav2vec 2.0 Large 960h Meta7±3local~2GB2020-06
173data2vec Audio Base 960h Meta4±4local~1GB2022-02
174wav2vec 2.0 Base 960h Meta2±3local~1GB2020-06
175MMS-1B-FL102 Meta-9±5local~5.1GB2023-05
more from Meta
1Muse Voice Transcribe 1.0 Meta71±3api $0.0030/audio min2026-09
58Omnilingual ASR LLM 7B Meta58±2local~19GB2025-11
59Omnilingual ASR LLM 7B v2 Meta58±3local~19GB2025-12

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

$ what it's good at

other tests3 not in a task · 95% of its score
testresultvs the others in its rundifficultycountswhy
Open ASR English short-form WER (March 2026)28.3% ✎▁▁▃▃··▁▁▂██▆ beats 2/86avg 8.2%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 43%blunt
LibriSpeech test-other WER lab7.1% ✎▂·▃·▂······█ beats 1/6avg 0.2%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 38%typical
LibriSpeech test-clean WER lab2.8% ✎▂··▂▃·····▂█ beats 2/8avg 0.4%▇▇▇▇▇ 13%blunt

$ where to run it

Other ways to run it

  1. Run it as published · needs ~0.95 GB · from huggingface.co/facebook
    uv run --with transformers --with torch python -c "from transformers import pipeline; print(pipeline('automatic-speech-recognition', model='facebook/data2vec-audio-base-960h', chunk_length_s=30)('audio.wav')['text'])"

Weights and documentation: huggingface.co/facebook ↗

$ how the score was worked out · Meta (facebook org) Hugging F… 52%, Open ASR Leaderboard 43%

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
Meta (facebook org) Hugging Face ASR model cards: published LibriSpeech WER for its own checkpoints · lab-reported2▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 52%Apache-2.0
Open ASR Leaderboard1▇▇▇▇▇▇▇▇▇▇▇▇▇ 43%Apache-2.0
prior · pull toward the middle▇ 5%
weighted average= 4.4

$ 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