Transcribe MP4 with OpenAI Whisper | Generated by AI
No transcription command is wired into ww main.py yet. The existing ww/audio/speech_to_text.py uses Google Cloud Speech-to-Text with GCS uploads — heavyweight for a one-off local file.
For a 2-hour MP4, the easiest path is OpenAI Whisper (local, free, no cloud setup):
# Install once
pip install openai-whisper
# Run (large model = best accuracy, ~10-20 min on CPU)
whisper ~/Downloads/your-video.mp4 --model large --output_dir ~/Downloads
Or if you’d like a ww transcribe command built into the toolkit (using Whisper or the OpenAI Whisper API), I can add one. Which would you prefer?