Transcribe audio
Transcribes audio and video recordings to text, with optional speaker labels and hints for speakers you already know.
When to use it
- You have an interview or meeting recording to turn into text.
- Speakers need labelling in the transcript.
- You want text extracted from a video’s audio.
A practical example
You ask
“Transcribe this interview and label the two speakers.”
Your agent does
It transcribes the recording, separates the speakers and saves a clean transcript.
What gets installed
The whole folder is copied, so scripts, references and any license notices stay together.
- agents/openai.yaml414 bytes
- assets/transcribe-small.svg750 bytes
- assets/transcribe.png1,288 bytes
- LICENSE.txt10,776 bytes
- references/api.md457 bytes
- scripts/transcribe_diarize.py8,684 bytes
- SKILL.md2,899 bytes
Before you start: Uses Python and needs an OPENAI_API_KEY.
Credit and license
Adapted from OpenAI’s published skills. It’s shared under the Apache License 2.0, and the license file is installed with it however you choose to install.