openai/whisper-large · Discussions

add link for whisper large v3 to the readme

#49 opened about 2 months ago by

iitsg

Correct long-form generation config parameters 'max_initial_timestamp_index' and 'prev_sot_token_id'.

#48 opened 3 months ago by

Correct long-form generation config parameters 'max_initial_timestamp_index' and 'prev_sot_token_id'.

#47 opened 3 months ago by

Correct long-form generation config parameters 'max_initial_timestamp_index' and 'prev_sot_token_id'.

#46 opened 3 months ago by

How to Generate a .mlmodel File for Apple's CoreML Framework

#45 opened 3 months ago by

Garry1234

Correct forced decoder ids

#44 opened 3 months ago by

Upload tokenizer

#43 opened 4 months ago by

OpenAI Whisper offline use for production and roadmap

#42 opened 6 months ago by

bahadyr

How can whisper return the language type?

#41 opened 6 months ago by

polaris16

Correct added token ids

#40 opened 7 months ago by

add special tokens for fast

#39 opened 7 months ago by

add timestamp tokens

#38 opened 7 months ago by

add timestamp tokens

#37 opened 7 months ago by

Adding `safetensors` variant of this model

#36 opened 8 months ago by

Narsil

Update generation config with word-level alignment heads

#35 opened 9 months ago by

Fine-tunining Whisper models for shorter audio segments

#34 opened 11 months ago by

Malishevsky

About finetuning whisper

#33 opened 12 months ago by

lypspeech

Sagemaker endpoint deployment (image_uri)?

#32 opened 12 months ago by

MLLife

Update generation_config.json to suppress task tokens

#31 opened about 1 year ago by

guillaumekln

Update config.json to suppress task tokens

#30 opened about 1 year ago by

guillaumekln

Update the pad token

#29 opened about 1 year ago by

Narsil

Upload tokenizer.json

#28 opened about 1 year ago by

Upload tokenizer

#27 opened about 1 year ago by

Unknown error

#26 opened about 1 year ago by

Nevertree

Link of model download

3

#25 opened about 1 year ago by

eashanchawla

Add suggestion to use v2 model

#24 opened about 1 year ago by

Update README.md

#23 opened about 1 year ago by

Should large still exist? Or should it link to large-v2?

#22 opened about 1 year ago by

altryne

change hebrew code from iw (deprecated) to he

#21 opened about 1 year ago by

altryne

Changes for fast tokenizer

#20 opened about 1 year ago by

jonatanklosko

Update config for automatic language detection

#19 opened about 1 year ago by

prerequisites for fine-tuning whisper model

#18 opened over 1 year ago by

Achitha

ONNX implementation

#17 opened over 1 year ago by

kirankumaram

Audio file is not transcribed after 30 second mark.

#16 opened over 1 year ago by

kirankumaram

Source of audio used to train Whisper

#15 opened over 1 year ago by

mahelona

Add "<|startoftranscript|>" to forced decoder ids

3

#14 opened over 1 year ago by

Not transcribing the audio into text (for some audios)

7

#13 opened over 1 year ago by

uriii3

Allow single quotes "'" and hyphens "-"

#12 opened over 1 year ago by

Transcription

6

#11 opened over 1 year ago by

Spotex93

forced_decoder_ids not applied properly when generation

#10 opened over 1 year ago by

minseong-ringle

Add pad token to tokenizer

#9 opened over 1 year ago by

Only the logits for the decoder_input_ids are returned, not for the actual input_features

#8 opened over 1 year ago by

joeyontour

Minor Update to Readme.md

#7 opened over 1 year ago by

Aaryan

Decoding of 'mp3' failed

3

#6 opened over 1 year ago by

tyatabe

Add TF weights

#5 opened over 1 year ago by

amyeroberts

model whisper not recognized?

#4 opened over 1 year ago by

manuelsh

Input error

#3 opened over 1 year ago by

mrJezy

Unable to run it on my local machine