Gemini 3.5 Transcribe and Transcribe Live
Gemini 3.5 Transcribe brings dedicated speech-to-text models with 85+ languages, diarization, word timestamps and a streaming Live variant.
Speech-to-text models built on Gemini's audio understanding, with custom-vocabulary biasing up to 1,000 terms and utterance-level language detection.
- Date
- Wednesday, 26 August 2026
- Lab
- Kind
- model
- Access
- closed API
Sources
- deepmind.google/blog/intelligent-transcription-with-gemini-3-5-transcribe/
- ai.google.dev/gemini-api/docs/changelog
This record was checked against its sources on 6 October 2026. How we check