openai-whisper

To transcribe audio locally without an API, use the openai-whisper Skill to process audio files directly on your machine via the Whisper CLI. It provides secure, private transcription without requiring external API calls or keys.

Yes, you need the whisper binary installed and accessible in your system's PATH to perform speech-to-text transcription. This local setup ensures your audio processing remains completely private and secure on your own machine.

Yes, you can generate SRT subtitle files from audio recordings using the Whisper CLI. The tool supports multiple output formats including plain text and subtitles, allowing you to choose the format that fits your needs.

The best way to balance speed and accuracy for local speech-to-text is by selecting different model sizes in the Whisper CLI. You can choose models like medium or turbo depending on whether you prioritize transcription speed or accuracy.

You should use local transcription instead of cloud services when you need secure and private audio processing. Running the Whisper CLI locally ensures sensitive meeting recordings or voice notes never leave your system, avoiding API costs.