openai-whisper-api

Transcribe audio files by running a shell script that uploads recordings to the OpenAI Audio Transcriptions API endpoint. The script uses curl to send the audio file and returns the converted text, automating the speech-to-text process.

Yes, you need curl installed locally and a valid OPENAI_API_KEY configured in your environment. These two prerequisites allow the shell script to authenticate and upload audio files to the OpenAI Audio Transcriptions endpoint.

Yes, the transcription script supports language hints and prompt hints to improve accuracy for specific contexts. You can also select the output format, choosing between plain text and JSON for the returned transcript.

Using a curl-based CLI script is the best way to transcribe short podcasts or meetings in shell workflows. It automates uploading audio files to the Whisper API and outputs searchable text files for documentation and analysis.

Yes, the API supports both text and JSON output formats for interview transcripts. You can configure the script to return JSON, providing structured transcription data suitable for further automated processing.

The Whisper API is ideal for short recordings, interviews, meetings, and podcasts, but may have limitations with very long audio files or poor audio quality. It requires a valid OpenAI API key and curl installed locally to function.