whatmodel.sh / stt / Omnilingual ASR LLM 300M

Omnilingual ASR LLM 300M

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

Omnilingual ASR LLM 300M is an open-weights speech-to-text model from Meta, released November 2025. Its published size is 1.63B parameters. Licence: apache-2.0.

Catalogue identity notes

Exact checkpoint facebook/omniASR-LLM-300M (fairseq2 card 'omniASR_LLM_300M', first Omnilingual ASR release, Nov 2025); distinct from its '_v2' successor. Params: card table. Needs libsndfile (card).

52.0±2.4#88 of 162 speech-to-text modelsfrom 5 benchmark results across 4 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 Omnilingual ASR LLM 300M, out of 176 with a score, then more from Meta.

#modelscorewhereRAMreleased
97Parakeet primeline (German) primeLine53±4local~0.7GB2025-12
98Omnilingual ASR CTC 7B v2 Meta52±3local~16GB2025-12
99Omnilingual ASR LLM 300M Meta52±2local~4.4GB2025-11
100Rev AI Machine (Reverb ASR) Rev52±3api $0.0033/audio min
101LiteWhisper large-v3-fast Efficient-Speech (LiteASR)52±3local~3.6GB2025-02
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

multilingual transcription▂··▃▄█▆▅▄▂·▂ #34 of 36 * · 65% of its score · 1 test

Trails Omnilingual ASR CTC 1B v2, Omnilingual ASR LLM 300M v2, Omnilingual ASR CTC 1B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Open ASR multilingual WER9.3% ✎▂··▃▃▂█▃█▂▁▁ beats 2/29avg 8.9%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 65%typical
other tests4 not in a task · 33% of its score
testresultvs the others in its rundifficultycountswhy
Omnilingual ASR paper · FLEURS-81 test mean CER (Whisper-covered languages) lab7.8% ✎▅···▃···▃▃██ beats 5/10avg 13%▇▇▇▇▇▇ 14%sharp · noisy
Omnilingual ASR paper · Common Voice 22 (76 languages) test mean CER lab7.1% ✎▂▂▂····▂▆··█ beats 7/10avg 21%▇▇▇▇ 10%sharp · noisy
Open Universal Arabic ASR Leaderboard · 6-test-set average WER33.0% ✎▃·▂▂▅█▅▃▃▅▂▂ beats 17/22avg 40%▇▇ 5.6%blunt
Quranic ASR Leaderboard · held-out recitation (600 clips) overall WER22.1% ✎▃·▅·▅▃▃·▃▃▃█ beats 5/12avg 24%▇ 3.4%blunt

$ where to run it

Other ways to run it

  1. Run it as published · needs ~4.41 GB · from huggingface.co/facebook
    uv run --python 3.12 --with omnilingual-asr python -c "from omnilingual_asr.models.inference.pipeline import ASRInferencePipeline as P; print(P(model_card='omniASR_LLM_300M').transcribe(['audio.wav'], batch_size=1))"

Weights and documentation: huggingface.co/facebook ↗

$ how the score was worked out · Open ASR Leaderboard 65%, Omnilingual ASR: Open-Source… 24%, Open Universal Arabic ASR Lea… 6% and 1 more

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▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 65%Apache-2.0
Omnilingual ASR: Open-Source Multilingual Speech Recognition for 1600+ Languages (arXiv:2511.09690v1) · lab-reported2▇▇▇▇▇▇▇ 24%CC BY-SA 4.0
Open Universal Arabic ASR Leaderboard (Elm Research Center)1▇▇ 6%Apache-2.0
Quranic ASR Leaderboard (Quran-Lab, held-out recitation set)1▇ 3%Apache-2.0
prior · pull toward the middle▇ 2%
weighted average= 52.0

$ 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