Whisper
Whisper is OpenAI's general-purpose speech recognition model for multilingual transcription, speech translation and language identification. It can run locally or be used through a paid API.
🌟
Free Plan
Free
The open-source model can run locally; compute costs are separate
🚀
Paid Plan
$0.006/minute for the whisper-1 API (USD)
whisper-1 API transcription: $0.006/minute (USD)
The whisper-1 API is deprecated and scheduled to shut down on February 26, 2027; this API shutdown does not affect the local open-source model
Information checked: 2026-10-23
Features & Use Cases
Key Features
Speech transcription
Multilingual recognition
Speech translation
Local open-source model
Use Cases
Audio ProcessingSpeech RecognitionSpeech to TextOpenAI