This week we introduced Gemini 3.5 Transcribe, our most precise speech-to-text model yet. Unlike conventional speech recognition models that struggle with background noise, complex jargon, and disfluency cleanup, Gemini 3.5 Transcribe turns audio into precise transcription in 85+ languages, removing filler words like "ums" and "ahs" while handling self-corrections and capturing your intent so you can get things done using just your voice. It's available now for developers in Google AI Studio, Gemini Enterprise Agent Platform, and Google Antigravity, and in everyday apps like Gboard for Android (via Rambler), the Gemini app on macOS, and coming soon to Chrome.

We recently introduced a new way to create, edit, and summarize in the Gemini app for macOS using just your voice, right where you’re already working. This intelligent dictation feature allows you to speak naturally into any window on your desktop, transcribing your spoken words into clean, polished text. It automatically removes the “ums” and “ahs,” catches mid-sentence corrections, and drops the formatted text instantly at your cursor.