Overview
Whisper is a speech-to-text model released by OpenAI under the MIT licence. It can transcribe interviews, dictation or meeting recordings in many languages, including French, on a computer you control, so audio does not have to leave your organization. It is a model and command-line tool rather than a finished app, so it needs some technical setup or a third-party program built on it. Transcripts can contain errors and should be reviewed before they are relied on.
Key facts
- Type
- Open-source speech recognition model
- Pricing
- Free and open source (MIT licence)
- Self-hosting
- Yes, runs on your own computer or server