Google has updated its Gemini Audio models with Gemini 3.5 capabilities, enhancing transcription accuracy across over 85 languages and specialized jargon. The new Gemini 3.5 Transcribe model automatically removes filler words like "um" and "uh," formats text, and can adapt to custom vocabularies. This update also includes Gemini 3.5 Live and Live Experimental for improved voice chat and real-time narration of complex tasks. These features are rolling out to macOS Gemini app users, Android devices, and developers via API, with Chrome support expected soon. AI
IMPACT Enhances usability of voice-controlled AI features by improving transcription accuracy and reducing manual editing needs.
RANK_REASON Product update for an existing AI model family, not a new frontier model release.
- Gemini 3.5
- AI Studio
- Chirp 3
- Gemini 3.5 Live
- Gemini 3.5 Live Experimental
- Gemini 3.5 Transcribe
- Gemini Audio
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →