Multi-Model Transcriber

by Jeff

Transcribes audio and video locally with multiple faster-whisper models, merges them into one majority-vote transcript, and writes a short summary. Audio never leaves the machine it runs on, and no API key is needed.