Moonshine Streaming Medium
Useful Sensors · usefulsensors/moonshine-streaming-medium
Speech to textTranscribes a recording into words
- Type
- Open weightsMIT License
- Languages
- 1
- Size
- 0.2B
Context measured in tokens
Our take
Written Aug 2, 2026Moonshine Streaming Medium is a tiny downloadable speech-to-text model from Useful Sensors that turns audio into written words. It is built for speed and edge deployment, with a permissive MIT licence and an error rate of roughly one word in seventeen on English audio overall.
Pick this when raw speed matters most — it processes an hour of audio in about a second on benchmark hardware. Use it for edge or offline deployment where a tiny footprint and permissive licence are essential, or for clean read-aloud English where it performs best. Skip it if you need languages other than English, if your audio is accented or from meetings and podcasts, or if you want a managed hosting option rather than self-hosting.
The case for it
- Extremely fast: 2,681 times real time on benchmark hardware, an hour of audio in one second.
- Permissive MIT licence allows commercial use, modification and redistribution.
- Strong on clean read-aloud English at 1.7% word error rate, roughly 3.4 times better than its own average.
- Tiny at 0.2B parameters, enabling deployment on very small devices.
The case against it
- Accuracy collapses on challenging audio: 11.4% word error rate on accented speech, nearly seven times worse than its clean-speech score.
- No commercial hosting options available in our catalogue; you must self-host.
- Only English is supported.
How good is it?
TranscriptionTurning speech into text3 of 5Open ASR WER · 38th of 74
94.2%
Misses roughly one word in 17, averaged over nine English test sets.
2,681×16th of 62
an hour of audio in 1 seconds, on the board's own hardware. Your machine will differ.
1
Listed on the model card. The accuracy above is English only.
Percentage of words wrong on each set, lower better. Bars are scaled to this model's own worst case, not to the board.
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.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
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?
- Fits in memory
- weights load entirely on the card
- Spills to system RAM
- some weights offload; much slower
- Too large
- will not load even with offload
- est
- size is calculated; the verdict could change by 10%
Comfortable fit
GeForce RTX 4090 · 24 GB
Room to spare. 21.4 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. 22.6 GB spare means a 10% error in the size would not change the answer.
Apple M1 (8-core GPU, 8GB unified) · 8 GB
Room to spare. 4.6 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.
Check against your own machine → · All 71 devices, with every size →
When we formed this view
Dates behind this page
Prices last checked 6h ago
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 allowsMIT License, 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
MIT License
Fully permissive: do anything with attribution. No patent grant, unlike Apache-2.0.
Identifiers
- Hugging Face
- usefulsensors/moonshine-streaming-medium
- Modality record
- audio->text
- Catalogue slug
- usefulsensors-moonshine-streaming-medium