openai-whisper-api

You can transcribe audio to text using OpenAI Whisper by sending a request to the /v1/audio/transcriptions endpoint with your audio file. This Skill provides a curl-based script to automate the process and output searchable transcripts.

Yes, you can use curl to convert meeting recordings and podcasts into transcripts. The Skill calls the Whisper API directly from your terminal, making it suitable for transcribing lectures, podcasts, and voice memos for archival workflows.

Yes, you need an OPENAI_API_KEY to use the Whisper API for speech-to-text. You also need curl installed to run the transcription script and send the audio file to the API endpoint for processing.

The Whisper API returns audio transcription output in text or json formats. This allows you to choose between plain text for simple reading and json for structured data integration in your content creation or research workflows.

The best way to get started with local audio transcription testing is to use the provided transcribe.sh script. You run it with a local file path and your OPENAI_API_KEY to quickly produce a transcript without complex tooling setup.