Parakeet TDT CTC 110m
NVIDIA · released Sep 17, 2024 · nvidia/parakeet-tdt_ctc-110m
Speech to textTranscribes a recording into words
- Type
- Open weightsCreative Commons Attribution 4.0
- Languages
- 1
- Size
- 0.1B
Context measured in tokens
Our take
Written Sep 5, 2026Parakeet TDT CTC is a tiny English-only speech-to-text model from NVIDIA built for raw speed over accuracy. At 110 million parameters it processes an hour of audio in under a second, though it makes more errors than most models on every condition we measure.
Pick this when transcription speed is the overriding need — it is among the fastest models we track. Use it for offline batch processing of clean, read-aloud English audio where some accuracy loss is acceptable, or for research under a permissive attribution licence. Skip it if you need accuracy on accented speech or meetings, or if you want a hosted provider rather than self-hosting.
The case for it
- Processes audio at 6,119 times real time — an hour of audio in under a second on the benchmark rig.
- Creative Commons Attribution 4.0 licence allows commercial use with attribution.
- Small enough to run almost anywhere at 110 million parameters.
The case against it
- Below-median accuracy on every condition measured: clean read-aloud, podcasts and video, accented speech, and recorded meetings.
- Overall word error rate of 6.6% lags the 5.8% field median — roughly 14% more errors than a typical model.
- No hosted providers in our data; you must self-host.
How good is it?
An open transcription model for turning speech into text, though it struggles more than most with clear read-aloud recordings.
- transcribing clear recordings of people reading aloudClean read speech · 75th of 92
TranscriptionTurning speech into text2.5 of 5Open ASR WER · 56th of 76
93.9%
Misses roughly one word in 16, averaged over nine English test sets.
6,084×8th 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. 21.5 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. 22.7 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. 4.7 GB spare means a 10% error in the size would not change the answer.
These cards answer whether Parakeet TDT CTC 110m 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.
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
- nvidia/parakeet-tdt_ctc-110m
- Takes in, gives back
- Audio in, text out
- Catalogue slug
- nvidia-parakeet-tdt-ctc-110m