← Model library
audio
Whisper large-v3
OpenAI · Family Whisper
Whisper large-v3 is the standard speech-recognition model for meetings, subtitles and search indexing. Strong in 90+ languages including Turkish; near real-time speed on an RTX 3090.
Key Features
- ✓ Transcription in 90+ languages
- ✓ Speech-to-English translation
- ✓ Timestamp support
Use Cases
- → Meeting and call records
- → Video subtitling
- → Content search indexing