Google launches Gemini 3.5 Transcribe with cleanup and speaker-detection features
Google unveiled Gemini 3.5 Transcribe, a new speech-to-text AI model that detects over 85 languages, strips filler words, and reformats casual speech into clean, structured text. It can also learn custom vocabulary, capture alphanumeric strings like order numbers, and assign dialogue to up to three speakers with timestamps from recorded audio.
GoKawiil's interpretation of the reporting above, not reported fact.
This model already underpins Android's Rambler feature and the Gemini app on macOS, and Google plans to extend it into Chrome, Search Live, Gemini Live, Docs, Keep, Gmail, and its Antigravity dev platform. By pushing voice-to-structured-text deeper into everyday tools, Google is betting that dictation—not typing—becomes a primary way people interact with software and AI agents.
- Gemini 3.5 Transcribe detects 85+ languages and cleans up filler words and self-corrections
- Supports speaker attribution for up to three voices with word-level timestamps
- Rolling out across Chrome, Docs, Gmail, Search Live, and available to developers via API
Blue Yeti USB Microphone — If you're going to lean on AI transcription for dictation, interviews or podcasts, clean input audio makes a huge difference — and the Blue Yeti is a classic plug-and-play USB condenser mic. Its multiple pickup patterns let you record yourself solo or capture a couple of people around a table, which pairs nicely with multi-speaker transcription.
See Blue Yeti USB Microphone on Amazon → Affiliate link — we may earn a commission on purchases, at no extra cost to you. Product picked by AI based on this article; it is not a tested recommendation.Source: engadget.com — Kris Holt, 2026-08-26
Published there as: “Google says its latest Gemini transcription model can turn your ramblings into structured text”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.