Google announces Gemini 3.5 Transcribe for AI-powered speech-to-text
Summary
Google has introduced Gemini 3.5 Transcribe, a new AI tool that converts speech to text more quickly and accurately than before. It removes filler words like “ums” and “uhs” and corrects errors during transcription. The tool is already used in some Google products and will expand to more devices and apps soon.Key Facts
- Gemini 3.5 Transcribe is an AI model for turning spoken words into written text.
- It is about 70% faster than Google’s previous tool, Chirp 3.
- The error rate for live speech transcription has improved from 7.32% to 5.5%.
- The model can remove filler words such as “ums” and “uhs” automatically.
- It can correct speech mistakes on the spot and understands specialized terms through custom vocabulary.
- Gemini 3.5 Transcribe works in 85 languages and can handle up to three speakers in recorded audio.
- Currently, it is used in the Pixel 11 phone’s Gboard Rambler feature and the Gemini app on macOS.
- Google plans to bring this AI transcription to more devices and to the Chrome browser soon.
Read the Full Article
This is a fact-based summary from The Actual News. Click below to read the complete story directly from the original source.
