Tools for content creation.
Extracts audio from a video and transcribes it with Whisper.
python3 transcribe/transcribe.py video.mp4
python3 transcribe/transcribe.py video.mp4 -m small -f srt
python3 transcribe/transcribe.py video.mp4 -f all -o ~/output| Model | VRAM | Speed | Accuracy |
|---|---|---|---|
| tiny | ~1 GB | Very fast | Acceptable |
| base | ~1 GB | Fast | Good |
| small | ~2 GB | Medium | Very good |
| medium | ~5 GB | Slow | Excellent |
| large | ~10 GB | Very slow | SOTA |
- Python 3.9+
- ffmpeg
- openai-whisper:
pip3 install openai-whisper
MIT