Google Announces Gemini 3.5 Transcribe for AI-Powered Speech-to-Text
Google says the model cuts filler words, handles self-corrections and supports more than 85 languages for cleaner voice dictation.
- On Wednesday, Google unveiled Gemini 3.5 Transcribe, described as the company's most precise speech-to-text model yet, replacing the previous Chirp transcription system.
- Designed to handle unpolished speech, Transcribe automatically removes filler words like "um" and "ah," while catching self-corrections in real time so users can speak naturally without editing.
- Google reports a word error rate of 4% for real-time streaming and 2.6% for pre-recorded audio, with support for more than 85 languages and the ability to attribute speech to three speakers with timestamps.
- Transcribe currently powers Rambler on Android devices including Pixel phones, and the Gemini app, with developer access available through Studio and Antigravity platforms.
- Google plans to expand Transcribe into Search Live, Docs, Keep, and Gmail, while soon enabling speech-to-text functionality within any web field in Chrome.
17 Articles
17 Articles
Google announces Gemini 3.5 Transcribe for AI-powered speech-to-text
While we wait (possibly in vain) for Gemini 3.5 Pro to launch, Google is releasing a different model in the 3.5 branch. The company has announced Gemini 3.5 Transcribe, an AI model designed to streamline voice input by editing out "ums" and corrections, outputting polished AI text. This model already powers the Gboard "Rambler" feature on the Pixel 11, but it's about to appear throughout the Google ecosystem. According to Google, Gemini 3.5 Tran…
Gemini 3.5 Transcribe Speech-to-Text Model Released; Check Details | 📲 LatestLY
Google has introduced Gemini 3.5 Transcribe, an advanced speech-to-text model designed for noisy environments and complex voice interactions. It converts audio into polished text, supports real-time and pre-recorded processing, speaker attribution and over 85 languages, with access available through Google’s developer platforms. 📲 Gemini 3.5 Transcribe Speech-to-Text Model Released; Check Details.
Coverage Details
Bias Distribution
- 50% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
