What problem does it solve? Converting audio recordings into text manually is slow and error-prone. This Skill automates speech-to-text transcription of audio files (.mp3, .m4a, .wav, etc.) using mlx-whisper, with automatic model selection for English and multilingual content. ## Core Features & Use Cases - Audio-to-Text Transcription: Converts audio files to plain text transcripts using mlx-whisper with Metal GPU acceleration on Apple Silicon Macs. - Translation and Multilingual Support: Auto-selects a multilingual model when translating non-English audio to English or transcribing in the original language via --task and --language flags. - Container Bridge Support: When running inside the multiplai container, bridges to the macOS host over SSH so transcription still uses the host's Metal GPU, with workspace path confinement for safety. - Use Case: You recorded a one-hour interview as an .m4a file and need a searchable text transcript; run the script and get a .txt file written next to the audio. ## Quick Start Ask the assistant to transcribe your audio file, for example: transcribe the file /path/to/interview.m4a into text.
