cyberneticlibrary

Transcribe speech locally with Whisper

openai-whisperskillsetup L1377,536
openclaw/openclaw
What it does

Transcribe audio files to text locally

Best for

Converting speech to text without API costs or latency when privacy matters

Inputs
  • · audio file path (mp3, m4a, etc)
Outputs
  • · transcribed text (txt, srt, or json)
Requires
  • · whisper CLI (brew install openai-whisper)
Preconditions

whisper binary installed; audio file accessible

Failure modes
  • · Model download fails if offline
  • · Unsupported audio format
Trust signals
  • · OpenAI official implementation
  • · Supports multiple languages and tasks (transcribe/translate)