openai
/

whisper-large-v2

Automatic Speech Recognition

hf-asr-leaderboard

Model card Files Files and versions

Add forced decoder ids

#2

by sanchit-gandhi - opened Dec 7, 2022

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

•

edited Dec 7, 2022 by

Force <|startoftranscript|><|en|><|transcribe|><|notimestamps|> as the 4 first generated tokens.

Add forced decoder ids463445a5

I think we can remove <|startoftranscript|>as <|startoftranscript|> is set in decoder_start_token_id

ArthurZ changed pull request status to merged Dec 7, 2022

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment