jumon whisper-finetuning issues

jumon / whisper-finetuning

[WIP] Scripts for fine-tuning Whisper

MIT License

205 stars 28 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

The best parameters to use during the training

#29 TalapMukhamejan opened 2 weeks ago
0
Cut the segment at the wrong place in create_data.py

#28 dmks opened 2 months ago
0
RuntimeError: Given groups=1, weight of size [1280, 128, 3], expected input[16, 80, 3000] to have 128 channels, but got 80 channels instead

#27 Binizox opened 2 months ago
0
this program support muti GPUs,How to use

#26 yjyz1011 opened 4 months ago
0
how to generate audio dataset in SRT format ?

#25 StephennFernandes opened 5 months ago
0
Great idea - thank you

#24 marceltud opened 9 months ago
0
model format for using in faster-whisper or similar transcription tools

#23 alirezashojaei opened 10 months ago
0
Sudden increase in error (validation loss sudden increase )

#22 alirezashojaei opened 10 months ago
0
Any other whisper fine tuning projects?

#21 jaggzh opened 11 months ago
1
ESC

#20 ShadowVicky opened 1 year ago
0
RuntimeError: No audi I/O backend is available

#19 japorto100 opened 1 year ago
0
Training Large model

#18 emcodem opened 1 year ago
0
Training with very short phrases for a patient.

#17 jaggzh opened 1 year ago
2
Crashing on run_finetuning.py

#16 Muertoe opened 1 year ago
2
ResultWriter.__call__() missing 1 required positional argument: 'options'

#15 jaggzh opened 1 year ago
2
Crash on run_finetuning.py

#14 Muertoe closed 1 year ago
4
the loss is nan

#13 ernestzhang opened 1 year ago
4
fine tuning questions using tiny model

#12 cchen028 opened 1 year ago
1
Fix missing tokenizer attribute

#11 dbalabka closed 1 year ago
0
this is more of a question than an issue

#10 silvacarl2 opened 1 year ago
5
finetuned model returns garbage

#9 jprobichaud opened 1 year ago
2
Fix merge

#8 marcuskbrandt closed 1 year ago
0
How to get some SRT or VTT data，How do I get the timestamp for each token ？

#7 daihuangyu closed 1 year ago
3
How to run finetuning on multiple GPUs?

#6 Johnson-NLP closed 1 year ago
2
Add option to use Adam 8bit to reduce VRAM usage

#5 flesnuk closed 1 year ago
1
change write_srt function to get_writer, since it changed in Whisper

#4 flesnuk closed 1 year ago
1
On Windows, when using run_finetuning it fails with cannot pickle 'builtins.CoreBPE' object

#3 flesnuk opened 1 year ago
4
Fix access to special_tokens function since it was changed

#2 flesnuk closed 1 year ago
1
Specify encoding when writing the json data in create_data

#1 flesnuk closed 1 year ago
3