Speech to text · Hugging Face
Distil-Whisper large-v3.5
Distilled Whisper: faster, nearly the same English accuracy.
Datasheet
Maker | Hugging Face |
|---|---|
Parameters | 756 MVerifiedsafetensors.total from the HF API Source |
Licence | MITVerified |
Access | Public download |
Task (HF pipeline) | automatic-speech-recognition |
First published on HF | 2024-12-05 |
Last updated on HF | 2026-04-13 |
Weights in memory | 1.5 GB at 16-bit · 0.5 GB at ~4.85 bitsEstimateParameters × bits ÷ 8. Excludes context cache and runtime overhead; not every model has a 4-bit build. |
Benchmark
7.02%
Open ASR Leaderboard, average WER (English short-form) · #4 of 7 scored in this category
avg 7.0229, RTFx 898.21. Source
The description summarises the model card at https://huggingface.co/distil-whisper/distil-large-v3.5. Parameters and licence are read from the Hugging Face API; the weights-in-memory line is our arithmetic, not a measurement.