←── back to feed
/topics/gemini-3-5-transcribe-speech-to-text
Gemini 3.5 Transcribe speech-to-text
3 items●1 sources●updated 25d ago●trend 0
Google announced Gemini 3.5 Transcribe, a new speech-to-text model that powers intelligent transcription across Google products including Chrome and Gboard. The model automatically removes filler words like 'ums' and 'ahs', detects specialized jargon, and supports over 85 languages with improved multilingual performance compared to its predecessor Chirp 3.
- Supports more than 85 languages with enhanced multilingual performance
- Automatically removes filler words ('ums', 'ahs') from transcriptions
- Detects specialized jargon and technical terminology
- Rolling out to Chrome, Gboard, and other Google products
- Represents major advancement over previous Chirp 3 model with lower wording error rates