Openai Whisper

This runs OpenAI's Whisper speech recognition model entirely on your machine, no API keys or internet required once you've installed it via Homebrew. You point it at an audio file and get back transcriptions in whatever format you need: plain text, SRT subtitles, or others. Models cache locally in ~/.cache/whisper after first download, so subsequent runs are fast. The main tradeoff is the usual one: smaller models are quicker but less accurate, larger ones are slow but precise. Good for when you need to transcribe recordings without sending them to a third party service, or when you're offline and still need to turn speech into text.

Install to Claude Code

npx -y skills add steipete/clawdis --skill openai-whisper --agent claude-code