openai-whisper-api

To transcribe audio files to text, run the provided curl-based script against OpenAI's /v1/audio/transcriptions endpoint. You need a valid OPENAI_API_KEY to process recordings like meeting notes or podcasts into searchable text transcripts.

Yes, you can get JSON output from the speech-to-text transcription. The script defaults to plain text but supports parameters to request JSON output, allowing you to generate metadata-friendly files for downstream programmatic processing.

Yes, you need curl installed to use this OpenAI audio transcription script. The Skill relies exclusively on a curl-based script to call the Whisper API and transcribe your audio files into text or JSON formats.

The best way to guide wording during audio transcription is to use the prompt parameter. Supplying a prompt helps the Whisper model preserve specific context like speaker names, ensuring accurate transcripts for your meeting recordings.

Yes, this OpenAI Whisper transcription script supports setting both the language and model. You can choose the transcription model, optionally set the language, and provide a prompt to ensure accurate speech-to-text conversion.

You can batch transcribe meeting recordings in formats like .m4a or .ogg. The script processes supported audio files through the OpenAI Whisper API, generating automatic searchable transcripts for each recording in your directory.