Transcribe Audio stored in Attachments converts audio attachments in Airtable into text using OpenAI’s Whisper model. The transcription is saved in a selected field on the same record for review, search, or later processing.
What you can do
- Select an attachment field containing audio files.
- Use your own OpenAI API key.
- Combine transcription results from qualifying files into one text output per record.
Use cases and examples
A research team can store interview recordings as attachments and generate a text field for review. Another workflow can use that field for summaries or categorization after someone has checked transcription quality.
- Meeting notes: Transcribe recordings of team meetings so colleagues can review and search the spoken discussion.
- Podcasts and interviews: Prepare editable transcripts of podcast episodes or interviews for review and content production.
- Video caption preparation: Use transcripts from supported audio recordings as source material for a separate subtitle or caption workflow. Review the text and add timing as needed; plain transcription text is not a ready-made timed caption file.
- Educational recordings: Create searchable text from recordings of lectures, seminars, and workshops.
How to get started
Add and select your OpenAI API key, choose the audio and transcription fields, and test a representative recording. Confirm provider access and choose whether later runs should preserve or replace existing transcripts.
Good to know
OpenAI usage charges and supported audio formats apply. The output is plain text rather than timestamped subtitles or speaker labels.
