NOISESaved copy
Coverage history

Assembled by NOISE.

Saved report copy

Recorded

Google introduces Gemini 3.5 Transcribe for faster, cleaner speech-to-text

Google has announced Gemini 3.5 Transcribe, a speech-to-text model that removes disfluencies, handles corrections and supports custom vocabulary. Ars Technica reports that Google says it is about 70% faster than Chirp 3, with a live-speech error rate of 5.5%, and supports 85 languages and up to three speakers in prerecorded audio. The model is available in Gboard’s Rambler on Pixel 11 phones and the Gemini macOS app, with Chrome support planned soon.