Skip to main content
Audio modelActive

Gemini 3.5 Transcribe Live

Google

Released
-
Data date
September 22, 2026
Access

Profile

Specifications and access

Published information about this model. Existing estimates are explicitly labeled.

SpecificationValue and source
Model class
Live speech recognitionSource
Access
API model ID
gemini-3.5-transcribe-liveSource
Input
Audio streamSource
Output
TextSource
Task
Live transcriptionSource
Supported details
Low-latency speech recognitionSource
Languages
Over 85; code-switching in the streamSource
Session duration
Up to 10 minutesSource
Feature limits
No diarization or word timestamps in live modeSource