Google Updates Gemini Audio With New Gemini 3.5 Transcribe Model
Google has announced a major update to Gemini Audio with the release of the Gemini 3.5 Transcribe model. The new transcription software introduces advanced automated features and is rolling out starting today for select users.
On August 26, 2026, Google officially updated its Gemini Audio platform by releasing the brand-new Gemini 3.5 Transcribe model. According to the company, this newly introduced software represents a major advancement from our previous transcription model, Chirp 3. The technology is designed to significantly improve how audio is processed and converted into written text across various digital environments.
The newly unveiled Gemini 3.5 Transcribe model comes equipped with a sophisticated set of automated capabilities tailored for modern users. The software can automatically detect specialized jargon, ensuring technical terms are accurately captured, while also providing support for over 85 languages. Furthermore, the model is built to automatically format text and intelligently remove common filler words, such as 'um' and 'uh', to deliver cleaner and more professional transcriptions.
As part of its initial deployment strategy, Google confirmed that Gemini 3.5 Transcribe is rolling out starting today in English for all macOS Gemini app users. Alongside the macOS rollout, the new technology is also coming to the Rambler dictation feature on Android devices, though this particular availability is currently limited to select countries and languages.
The release of Gemini 3.5 Transcribe continues Google's ongoing development of its audio processing capabilities within the Gemini ecosystem. By replacing older frameworks like Chirp 3, the updated model provides users with a more refined and capable tool for handling complex voice and dictation tasks across supported platforms and applications.
