Google’s new AI transcription edits out your ‘ums’ and ‘ahs’

Google has announced a significant upgrade to its artificial intelligence transcription capabilities through Gemini 3.5 Transcribe, a new addition to the Gemini family designed to enhance audio processing and language recognition. The updated tool automatically removes common speech filler words such as "ums" and "ahs," while simultaneously supporting over 85 languages and detecting specialized jargon across various professional fields. This development represents Google's ongoing effort to refine AI-powered audio tools and follows the recent launch of Gemini 3.5 Live Translate.

Gemini 3.5 Transcribe introduces intelligent transcription technology that goes beyond basic speech-to-text conversion. The system automatically detects and removes disfluencies—those natural verbal hesitations that characterize spontaneous speech—creating cleaner, more professional transcripts. Additionally, the tool's multilingual support spanning more than 85 languages enables global accessibility, while its ability to recognize and preserve domain-specific terminology ensures accuracy in specialized fields such as medicine, law, and technology. These features position the tool as particularly valuable for professionals seeking high-quality transcripts of meetings, interviews, and presentations.

• Enhanced transcription accuracy reduces post-editing requirements for businesses relying on automated transcription services
• Multilingual support expands accessibility for international teams and global enterprises
• Automatic filler word removal produces more professional-grade transcripts suitable for publishing and formal documentation
• Specialized jargon detection maintains technical accuracy in regulated industries
• Integration within the broader Gemini ecosystem creates a comprehensive AI audio solution ecosystem
• Competitive pressure on existing transcription service providers to enhance their own AI capabilities

The launch of Gemini 3.5 Transcribe signals Google's commitment to advancing practical AI applications that address real-world communication challenges. As remote work and digital collaboration continue to dominate business operations, the demand for accurate, efficient transcription tools remains critical. By combining automatic disfluency removal with robust multilingual and specialized language recognition, Google is delivering a solution that improves both transcription quality and professional utility. This evolution in AI transcription technology reflects the broader industry trend toward more sophisticated, context-aware language processing tools that enhance rather than simply replicate human communication.