Lecture Library

A semi quick vibe coded project because I needed it for school

A private, local Flask app for recording/uploading lectures, creating course libraries, transcribing audio with Whisper, and generating study summaries through Ollama.

Setup

1. Install Python 3.10+ and FFmpeg, ensuring ffmpeg is on your PATH.
2. Create and activate a virtual environment, then run pip install -r requirements.txt.
3. Install Ollama, then run ollama pull llama3.2 and start the Ollama service.
4. Copy .env.example to .env; adjust the model or storage settings if required.
5. Run python app.py, then open http://127.0.0.1:5000.

Whisper downloads its selected model automatically the first time a lecture is processed. Use WHISPER_MODEL=tiny, base, small, or another Whisper model according to your hardware.

Notes

• Audio and SQLite data are local in instance/ by default. Back up this directory to preserve your library.
• Browser recording needs microphone permission and is saved as WebM before entering the standard upload pipeline.
• FFmpeg is required by Whisper to read many audio formats. Ollama must be running at OLLAMA_URL and have OLLAMA_MODEL available.
• Processing activity and full error tracebacks are written to instance/lecture-library.log. Set LOG_LEVEL=DEBUG in .env for more detail.