distil-whisper
/

distil-large-v3-ct2

Automatic Speech Recognition

Model card Files Files and versions Community

sanchit-gandhi HF staff commited on Mar 21

Commit

404c3b9

•

1 Parent(s): d08db9b

Update README.md

Files changed (1) hide show

README.md +5 -0

README.md CHANGED Viewed

@@ -14,6 +14,11 @@ This repository contains the model weights for [distil-large-v3](https://hugging
 converted to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format. CTranslate2 is a fast inference engine for
 Transformer models and is the supported backend for the [Faster-Whisper](https://github.com/systran/faster-whisper) package.
 ## Usage
 To use the model in Faster-Whisper, first install the PyPi package according to the [official instructions](https://github.com/SYSTRAN/faster-whisper#installation).

 converted to [CTranslate2](https://github.com/OpenNMT/CTranslate2) format. CTranslate2 is a fast inference engine for
 Transformer models and is the supported backend for the [Faster-Whisper](https://github.com/systran/faster-whisper) package.
+Compared to previous Distil-Whisper releases, distil-large-v3 is specifically designed to give one-to-one equivalence
+with the OpenAI Whisper long-form transcription algorithm. In our benchmark over 4 out-of-distribution datasets, distil-large-v3
+outperformed distil-large-v2 by 5% WER average. Thus, you can expect significant performance gains by switching to this
+latest checkpoint.
 ## Usage
 To use the model in Faster-Whisper, first install the PyPi package according to the [official instructions](https://github.com/SYSTRAN/faster-whisper#installation).