whatmodel.sh / stt / SpeechBrain wav2vec2 CTC (LibriSpeech)

SpeechBrain wav2vec2 CTC (LibriSpeech)

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

SpeechBrain wav2vec2 CTC (LibriSpeech) is an open-weights speech-to-text model from SpeechBrain, released June 2022. Its published size is 0.32B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact checkpoint speechbrain/asr-wav2vec2-librispeech. Params derived from fp32 weight files (wav2vec2.ckpt 1,261,920,693 + asr.ckpt 8,526,821 bytes / 4 = 0.318B). released: HF repo creation month.

34.9±4.4#143 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 SpeechBrain wav2vec2 CTC (LibriSpeech), out of 176 with a score, then more from SpeechBrain.

#modelscorewhereRAMreleased
155Moonshine Base Useful Sensors (Moonshine AI)38±3local~0.8GB2024-10
156Whisper base.en OpenAI35±3local~0.7GB2022-09
157SpeechBrain wav2vec2 CTC (LibriSpeech) SpeechBrain35±4local~2GB2022-06
158Vosk small en-us 0.15 Alpha Cephei (Vosk)31±6local~0.3GB2020-12
159Moonshine Tiny Useful Sensors (Moonshine AI)29±3local~0.6GB2024-10
more from SpeechBrain
126SpeechBrain Conformer (LargeScaleASR / Loquacious) SpeechBrain47±5local~2.8GB2025-02
150SpeechBrain Transformer + TransformerLM (LibriSpeech) SpeechBrain40±7comparison only2022-03

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

$ what it's good at

other tests3 not in a task · 93% of its score
testresultvs the others in its rundifficultycountswhy
Open ASR English short-form WER (March 2026)14.3% ✎▁▁▁▃··▃▁▂██▆ beats 12/86avg 8.2%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 66%blunt
LibriSpeech test-other WER lab4.0% ✎▁▁▁▂▅▃▅▄█▂▂▁ beats 39/54avg 2.8%▇▇▇▇▇▇▇▇ 20%small run
LibriSpeech test-clean WER lab1.9% ✎▁·▁·▁▁▁▁▃▇█▆ beats 62/100avg 1.6%▇▇▇ 6.7%blunt · small run

$ where to run it

Other ways to run it

  1. Run it as published · needs ~2.04 GB · from huggingface.co/speechbrain (official)
    uv run --with speechbrain --with transformers python -c "from speechbrain.inference.ASR import EncoderASR as A; print(A.from_hparams(source='speechbrain/asr-wav2vec2-librispeech', savedir='pretrained_models/asr-wav2vec2-librispeech').transcribe_file('audio.wav'))"

Official weights and documentation: huggingface.co/speechbrain ↗

$ how the score was worked out · Open ASR Leaderboard 66%, SpeechBrain Hugging Face ASR… 27%

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
Open ASR Leaderboard1▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 66%Apache-2.0
SpeechBrain Hugging Face ASR model cards: published LibriSpeech/VoxPopuli WER for its own checkpoints · lab-reported2▇▇▇▇▇▇▇▇ 27%Apache-2.0
prior · pull toward the middle▇▇ 7%
weighted average= 34.9

$ 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