Google Launches Gemini 3.5 Transcribe With 70% Faster Transcription and Support for 85+ Languages
Summary
Google launches Gemini 3.5 Transcribe, its most advanced speech-to-text model yet, delivering a 70% faster transcription speed, a Word Error Rate as low as 2.6%, and support for 85+ languages — now available in public preview for developers and enterprises.
Key Points
- Google launches Gemini 3.5 Transcribe, its most precise speech-to-text model yet, achieving a Word Error Rate of 4.0% for streaming and 2.6% for non-streaming use cases, representing a 70% improvement in time to final transcription over its predecessor, Chirp 3.
- The model supports over 85 languages, multi-speaker identification, smart disfluency cleanup, custom vocabulary, and is available via two APIs — a real-time streaming Live API and a pre-recorded audio Interactions API — for developers building voice agents, captioning tools, and analytics pipelines.
- Gemini 3.5 Transcribe is already powering features across Google products including Rambler on Android, the Gemini app on macOS, Google Antigravity, and Google AI Studio, with Chrome support coming soon, and is available in public preview for developers and enterprises today.