Product

Whisper

by OpenAI · Speech Recognition

What it does

A general-purpose, multilingual speech recognition model that transcribes and translates audio across dozens of languages, available as an open-weight model and via the OpenAI Audio API.

Key features

  • Multilingual transcription and translation
  • Open-weight model available for self-hosting
  • Support for multiple audio formats
  • Language identification
  • Available via API (billed per minute of audio)

Integrations

OpenAI Audio APIChatGPT Voice Mode (underlying component)ZapierCustom self-hosted deployments