Niagara 38m Batch.en
Applied Brain Research · released Feb 19, 2026 · abr-ai/niagara-38m-batch.en
Speech to textTranscribes a recording into words
- Type
- Open weightsCustom licence
- Languages
- 1
- Size
- 38M
Context measured in tokens
Our take
Written Aug 3, 2026Niagara is a 38-million-parameter English speech-to-text model built for extreme batch speed. It processes an hour of audio in under a second on benchmark hardware, though its accuracy drops sharply in meetings and accented speech.
Pick this for batch transcription of clean, scripted English audio where throughput matters more than perfection. It is also a strong fit for financial call transcription, where it achieves its lowest measured error rate. Skip it if you need languages other than English, if your audio is unscripted or heavily accented, or if you want a standard permissive licence.
The case for it
- Processes an hour of audio in under a second on benchmark hardware, with a real-time factor above four thousand.
- Strong on clean, structured English audio: about one word in twenty-seven wrong on read speech, and its best measured result on financial calls.
- Tiny hardware footprint at 38 million parameters total.
The case against it
- Accuracy collapses in natural conditions: more than three and a half times its clean-read error rate in recorded meetings, and more than double on accented speech versus European-accented speech specifically.
- English only; no verified data for other languages.
- No commercial hosting available, and the custom licence is not a standard permissive one.
How good is it?
TranscriptionTurning speech into text1 of 5Open ASR WER · 67th of 74
91.7%
Misses roughly one word in 12, averaged over nine English test sets.
4,049×10th of 62
an hour of audio in under a second, 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. 21.6 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. 22.8 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. 4.8 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 →
Models people weigh against Niagara 38m Batch.en
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 allowsCustom licence, 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
Custom licence
This model ships custom license terms that don't map to a known template. We haven't parsed them, so commercial use, redistribution and derivatives are unverified — review the original terms before shipping.
Identifiers
- Hugging Face
- abr-ai/niagara-38m-batch.en
- Architecture
- Dense
- Modality record
- audio->text
- Catalogue slug
- abr-ai-niagara-38m-batch-en