Whisper
Audio & Speech
OpenAI
FreeOpen Source
About
OpenAI's open-source speech recognition model for multi-language speech-to-text
Key Features
Open-source speech recognition
large-v3 model, 99 languages
Speech-to-text and translation
Robust to noise and accents
Local and API deployment
Free for commercial use
Pricing
free
Open-source, free to self-host
api
OpenAI API $0.006/minute
providers
Third-party hosted from ~$0.22/hr
Use Cases
Audio and video transcription
Subtitle generation
Meeting and interview notes
Multilingual translation
Voice interface backends
Accessibility captioning
Pros
Free and open-source
High accuracy across languages
Local deployment for privacy
Robust to noise
Cons
Requires GPU for speed
No official real-time UI
Technical setup needed
Latest Update
July 2026: Whisper large-v3 remains a leading open ASR model with 99-language support and wide provider hosting