Google's Gemini 3.5 Transcribe turns speech to text in 85 languages while auto-correcting your verbal stumbles

Google has launched Gemini 3.5 Transcribe, a speech-to-text model for real-time transcription. It recognizes over 85 languages automatically, strips filler words like "um," corrects slips of the tongue, and formats text on its own. Google reports a word error rate of 4.0 percent for streaming and 2.6 percent for recorded audio, with 70 percent lower latency than its predecessor, Chirp 3. Through "function calling," the model can hand off tasks like image generation or web searches to other Gemini models.

The model ships with two interfaces.The Live API handles real-time streaming with very low latency (gemini-3.5-transcribe-live), while the Interactions API processes recorded audio with speaker attribution and timestamps (gemini-3.5-transcribe). The model is live in Google AI Studio and on the Gemini Enterprise Agent Platform. It's already built into Gboard for Android (via "Rambler") and the Gemini app on macOS, with Chrome support coming soon.

AI News Without the Hype – Curated by Humans

Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.