huggingface distil-whisper issues

huggingface / distil-whisper

Distilled variant of Whisper for speech recognition. 6x faster, 50% smaller, within 1% word error rate.

MIT License

3.32k stars 236 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

[eval] fix check for language arg

#139 sanchit-gandhi opened 1 week ago
0
[pseudo-labelling] fix concatenate datasets

#138 eustlb closed 2 weeks ago
0
ZeroDivisionError: division by zero

#137 wailokkwok opened 3 weeks ago
0
Repository Not Found for url

#136 wailokkwok closed 3 weeks ago
0
Discrepancy on WER benchmark result in Tedlium dataset

#135 MLMonkATGY opened 3 weeks ago
1
feat: add early stopping

#134 thomaschhh opened 3 weeks ago
0
Unable to set concatenate_audio parameter to False in run_pseudo_labelling.py

#133 lq0104 closed 2 weeks ago
0
[Eval] Use English normalizer when target lang is English

#132 sanchit-gandhi closed 1 month ago
0
Unable to reproduce results from the paper

#131 MLMonkATGY closed 1 month ago
6
How to set the target language for examples in README?

#130 clstaudt opened 1 month ago
7
Problems in concatenate_dataset

#129 George0828Zhang closed 2 weeks ago
0
recommendations and guidelines section

#128 eustlb closed 1 month ago
0
question about when to apply WER threshold filtering strategy with concatenated audio

#127 lq0104 opened 1 month ago
0
[training] freeze decoder

#126 eustlb closed 1 month ago
0
Freeze decoder

#125 eustlb closed 1 month ago
0
Freeze decoder

#124 eustlb closed 1 month ago
0
[eval] benchmark generation speed

#123 eustlb opened 1 month ago
0
[init] choose student model initialization layers

#122 eustlb closed 1 month ago
0
Optimize data loading, preprocessing, and metric computation

#120 arjun-234 opened 2 months ago
0
RuntimeError: User specified an unsupported autocast device_type 'mps'

#119 afsara-ben opened 2 months ago
0
Fix model saving at the end of training

#117 eustlb closed 2 months ago
0
Fix pseudo labelling map context manager

#116 eustlb closed 2 months ago
0
perceptually faster inference through pre-completion inference of audio

#114 sujitvasanth opened 2 months ago
0
Quantize distil-whisper?

#113 sujitvasanth opened 2 months ago
0
[training] update flax scripts for distil-large-v3

#112 sanchit-gandhi opened 2 months ago
0
[training] enable ddp during distillation

#111 sanchit-gandhi closed 2 months ago
0
add linter

#110 sanchit-gandhi closed 2 months ago
1
[training] use basic normaliser for multilingual eval

#109 sanchit-gandhi closed 2 months ago
1
[training] update pseudo-labelling for ddp

#108 sanchit-gandhi closed 2 months ago
1
[Question] Can we distill for multiple langauges for distil-small-whisper

#107 Killshot667 opened 2 months ago
3
[Issue] latest run_pseudo_labelling.py

#106 ckcraig01 opened 2 months ago
0
Resuming training fails

#105 hidoba opened 2 months ago
3
Finetuning on which model?

#104 RohitMidha23 opened 2 months ago
1
[training] fix readme config name

#103 sanchit-gandhi closed 3 months ago
0
[training] fix dataset table in readme

#102 sanchit-gandhi closed 3 months ago
0
[training] Updates for distil-large-v3

#101 sanchit-gandhi closed 3 months ago
1
large-v2 for english lost voice to text

#100 machenme opened 3 months ago
1
any executable script for running on custom data/given dataset

#99 afsara-ben closed 3 months ago
1
Voxpopuli text column "raw_text" HF dataset card shows empty string.

#98 guynich closed 3 months ago
1
Training README datasets table: text column and id column

#97 guynich closed 2 months ago
4
Pseudo-labelling librispeech_asr (train.360): KeyError `train-360` when not streaming.

#96 guynich opened 3 months ago
1
How to use distil-whisper-large-v3-de-kd model from HF?

#95 Arche151 opened 3 months ago
10
Cached English Common Voice dataset size.

#94 guynich closed 3 months ago
1
BetterTransformer optimization / flash_attn{_2}

#93 thomaschhh closed 3 months ago
6
Question: should the pseudo-labelling model and teacher model be the same?

#92 guynich closed 2 months ago
2
transcription results are inconsistent and timestamps are None type. Issue appears in the latest version of the transformers==4.38.1.

#91 kranipa opened 4 months ago
0
distil-small.en AttributeError

#90 andrewjones0198 closed 3 months ago
3
Best way to implement streaming application?

#89 9throok opened 4 months ago
0
Short form evaluation WER % for Librispeech clean test

#88 guynich closed 4 months ago
3
Errors running `run_long_form_eval.py`

#87 guynich closed 3 months ago
1