What problem does it solve? Converting audio and video recordings into written text manually is slow and error-prone, especially when meetings or interviews involve multiple speakers who need to be identified and labeled. ## Core Features & Use Cases - Fast Text Transcription: Transcribe audio files (mp3, mp4, m4a, wav, webm, and more) using gpt-4o-mini-transcribe with plain text, JSON, or diarized JSON output. - Speaker Diarization: Label speakers in meetings and interviews with gpt-4o-transcribe-diarize, including support for up to 4 known-speaker reference clips to improve attribution. - Deterministic CLI Workflow: A bundled Python CLI handles validation, chunking for long audio, output file management, and dry-run payload inspection. - Use Case: Transcribe a recorded team meeting in m4a format, automatically label Alice and Bob using short reference clips, and save a diarized JSON transcript for meeting notes. ## Quick Start Ask the assistant to transcribe your audio file, for example: transcribe meeting.m4a with speaker labels and save the transcript to the output folder.
