Niagara 19m Batch.en
Applied Brain Research · released Nov 13, 2025 · abr-ai/niagara-19m-batch.en
Speech to textTranscribes a recording into words
- Type
- Open weightsCustom licence
- Languages
- 1
- Size
- 20M
Context measured in tokens · download allowed, licence restricts use
Our take
Written Sep 11, 2026Niagara 19m Batch.en is a 20-million-parameter English speech-to-text model built for extreme speed. It processes an hour of audio in under a second, but accuracy lags behind almost everything else we track on every measured condition.
Pick this when raw throughput matters more than precision and the audio is clean read speech or financial earnings calls. Use it for local deployment on extremely constrained hardware where nothing larger will fit. Skip it if you need accurate transcription of meetings, podcasts, accented speech, or any language other than English.
The case for it
- Processes an hour of audio in under a second — 3,717 times real time on the benchmark hardware.
- At 20 million parameters, orders of magnitude smaller than typical speech models.
- Best relative performance on financial earnings calls, with a word error rate well below its own average.
The case against it
- Among the least accurate on clean read speech — the easiest test condition — at 4.51% word error rate.
- Worse than most models on everyday internet audio, meeting-room audio, and accented speech, with error rates roughly 1.6–2.1 times the median.
- No commercial hosting available and a non-standard licence; zero current offers and terms that are not Apache, MIT, or BSD.
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 · 71st of 76
- transcribing recordings of meetings in a roomRecorded meetings · 82nd of 92
- transcribing speakers with a range of accentsAccented speech · 70th of 76
- transcribing podcasts and video audioPodcasts and video · 85th of 92
- transcribing clear recordings of people reading aloudClean read speech · 89th of 92
TranscriptionTurning speech into text1 of 5Open ASR WER · 71st of 76
89.6%
Misses roughly one word in 10, averaged over nine English test sets.
4,324×17th of 74
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; 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. 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.
These cards answer whether Niagara 19m Batch.en 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 Niagara 19m Batch.en
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 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-19m-batch.en
- Architecture
- Dense
- Takes in, gives back
- Audio in, text out
- Catalogue slug
- abr-ai-niagara-19m-batch-en