audio-transcription

What problem does it solve?

This skill enables local speech-to-text transcription using whisper.cpp, producing accurate transcripts with timestamps and optional speaker labeling, ideal for secure, offline processing of audio intake in healthcare contexts.

Core Features & Use Cases

• Local transcription with whisper.cpp, delivering timestamped transcripts and optional diarization.
• Supports common audio formats and live recording for intake pipelines.
• Post-processing options including punctuation restoration, number/date normalization, and artifact persistence.
• Deliverables include transcript.json (segments + timestamps), model_selection.md, and error_handling.md to simplify integration.

Quick Start

Prepare a local audio file (for example, recording.wav) and run the transcription workflow to generate transcript.json with segments and timestamps.