Voice Editing in Gemini 3.5 Transcribe
Gemini 3.5 Transcribe can now edit, rewrite, and format text from spoken instructions, for developers using the Gemini API.
Details
- Works on freshly dictated speech or on text you pass in through the source_text field
- Supported in both gemini-3.5-transcribe and the live version, gemini-3.5-transcribe-live
- Covered in the Gemini API audio transcription and live transcription guides
Best for
Dictation tools where a user speaks a draft, then speaks changes to it.
When to use it
You are building a voice note or writing app and want cleanup and edits without a keyboard.
Who benefits
Developers building voice-first writing and note-taking features.
Join the newsletter to get ahead with AI.
I share strategies, new modules, free guides, and tools to get you better at using AI in your work.
Leading a team? See team programs →