Owsm CTC v4 1B
ESPnet · released Jan 16, 2025 · espnet/owsm_ctc_v4_1B
Speech to textTranscribes a recording into words
- Type
- Open weightsCreative Commons Attribution 4.0
- Languages
- 75
- Size
- 1B
Context measured in tokens
Our take
Written Sep 11, 2026Owsm CTC v4 1B is a one-billion-parameter speech-to-text model from ESPnet that turns audio into written words across 75 languages. It is extremely fast on benchmark hardware but sits below the middle of the pack on accuracy for most English audio conditions.
Pick this for batch archive processing where throughput matters more than precision, or for self-hosted deployments that need a permissive licence and broad language coverage. Use it when a roughly six percent word error rate is acceptable, such as indexing or keyword search. Skip it if you need top-tier accuracy on podcasts, accented speech or meetings, or if you want a managed host rather than running it yourself.
The case for it
- Extremely fast: about 777 times real time on the benchmark harness, turning an hour of audio into roughly five seconds of processing.
- 75 languages listed, unusually broad coverage for an open transcription model.
- Creative Commons Attribution 4.0 licence allows commercial use with attribution.
The case against it
- Below-median accuracy on most tested conditions: podcasts and video, accented speech, and recorded meetings all trail the field middle.
- No commercial hosting options available; you must self-host.
- Accuracy gap widens sharply on harder audio: meetings are more than five times the error rate of clean read speech.
How good is it?
An open speech-to-text model for turning recordings into written text, though it trails most models on accuracy.
- turning spoken English into written textOpen ASR WER · 59th of 76
- transcribing speakers with a range of accentsAccented speech · 60th of 76
- transcribing podcasts and video audioPodcasts and video · 74th of 92
- transcribing clear recordings of people reading aloudClean read speech · 77th of 92
TranscriptionTurning speech into text2.5 of 5Open ASR WER · 59th of 76
93.7%
Misses roughly one word in 16, averaged over nine English test sets.
777×38th of 74
an hour of audio in 5 seconds, on the board's own hardware. Your machine will differ.
75
Stated by the leaderboard; we do not hold the list itself.
Percentage of words wrong on each set, lower better. Bars are scaled to this model's own worst case; the placing beneath each rate is against every model measured on that set.
The figures above come from the Open ASR Leaderboard, an independent public test that runs every model on the same recordings. It is the only measurement of transcription quality we know of, so there are no other scores to show.
Each of these is the same transcription job on a different kind of recording, so together they say where it holds up and where it slips — not how closely it follows an instruction.
Every published score for this model9 scoresEvery figure we hold, from 9 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
Comfortable fit
GeForce RTX 4090 · 24 GB
Room to spare. 20.9 GB spare means a 10% error in the size would not change the answer.
Apple M1 Pro (16-core GPU) · 32 GB
Room to spare. 22.1 GB spare means a 10% error in the size would not change the answer.
Apple M2 (8-core GPU, 8GB unified) · 8 GB
Room to spare. 4.1 GB spare means a 10% error in the size would not change the answer.
These cards answer whether Owsm CTC v4 1B loads, not how fast it transcribes. Throughput figures for a transcription model come from its text decoder, so treat this as a fit answer rather than a speed one.
Memory use by level
Against a 24 GB card.
What is quantisation? →This model on every device we track71 devicesThe Q4 build most people download, on each device: what the weights come to, how much context the memory leaves, and whether it runs. Smallest device that runs it first. This is a fit answer: whether it loads, not how fast it transcribes.
Models people weigh against Owsm CTC v4 1B
When we formed this view
Recent changes
What moved
first indexed by our pipelineEach date is the day we first saw the change, or the day the maker announced it.
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
Licence and identifiers
What the licence allowsCreative Commons Attribution 4.0, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Creative Commons Attribution 4.0
Permissive content license: any use with attribution. Common for datasets and some model weights.
Identifiers
- Hugging Face
- espnet/owsm_ctc_v4_1B
- Takes in, gives back
- Audio in, text out
- Catalogue slug
- espnet-owsm-ctc-v4-1b