Google has updated Gemini Audio with a new transcription model, Gemini 3.5 Transcribe. According to Google, the model can automatically detect specialized jargon and supports more than 85 languages. It also removes filler words like “ums” and “ahs” and lets users edit naturally using their voice.
Google describes 3.5 Transcribe as a significant step up from its previous transcription model, Chirp 3, particularly in multilingual performance and wording error rates. The release follows the launch of Gemini 3.5 Live Translate. The company has yet to release the Gemini 3.5 Pro model it said would roll out in June.
Why it matters
Automatic filler-word removal and multilingual support could make transcription output cleaner and more broadly usable across languages, reducing manual editing.
Who should care
Users who rely on transcription, including those working across multiple languages or with specialized terminology, may find the new capabilities relevant.