search

Vaibhavs10 / insanely-fast-whisper

Apache License 2.0
7.76k stars 547 forks source link

CLI Feature requests: 1) Output .srt files, 2) Sequentially process all audio files in directory #174

Open ch826 opened 10 months ago

ch826 commented 10 months ago

@Vaibhavs10 Great tool! The speed is fantastic, roughly 50x realtime speed for me. Unfortunately, the resulting .json file is unusable for me.

My feature requests for Insanely Fast Whisper CLI:

1) Output transcript as .srt file. It would be great if Insanely Fast Whisper produced an .srt with standard timestamp format like is produced by generic Whisper AI.

2) Batch process an entire directory of audio files. Ideally, the script would be able to sequentially process an entire directory of audio files, one at a time, producing an .srt file for each, saving each .srt file to the same directory as the source audio file.

Thank you for your work and for releasing this tool!

Orion-Zheng commented 9 months ago

I also have the need to get .rst file so I wrote a simple script to implement feature 1. It works for me.😃You can try it by python convert_srt.py output.json -o my_caption.srt The PR is here: https://github.com/Vaibhavs10/insanely-fast-whisper/pull/176