Intelligent Transcription With Gemini 3.5 Transcribe
Google, Wednesday, August 26th, 2026
Google releases Gemini 3.5 Transcribe, a speech-to-text model built for precise real-time transcription.
Google has released Gemini 3.5 Transcribe, its latest speech-to-text model designed for precise and intelligent real-time transcription.
The model comes from the Gemini Audio team and targets scenarios where transcription must keep pace with live speech rather than running as a batch job over recorded audio.
Google positions it as more intelligent than conventional speech recognition, implying handling of context, speaker intent and formatting rather than a literal word-by-word transcript. It is available for developers building transcription into their own products.