Model Releases
Google announces Gemini 3.5 Transcribe for AI-powered speech-to-text
Google has launched Gemini 3.5 Transcribe, a new voice‑to‑text AI model in the Gemini 3.5 branch that powers Gboard’s “Rambler” feature on Pixel 11 and is set to roll out across Google products such a
Google has launched Gemini 3.5 Transcribe, a new voice‑to‑text AI model in the Gemini 3.5 branch that powers Gboard’s “Rambler” feature on Pixel 11 and is set to roll out across Google products such as Chrome.
The model runs about 70 % faster than the previous Chirp 3 engine, reduces live‑speech error rate to 5.5 % (vs. 7.32 %), removes filler words like “ums”/“uhs”, allows on‑the‑fly corrections and custom vocabularies, and supports 85 languages with up to three speakers in pre‑recorded audio.
Related
- Today we’re introducing Gemini 3.5 Transcribe, our latest transcription model built for incredibly precise, smart dictation across your favo…
- Intelligent transcription with Gemini 3.5 Transcribe
Source: Ars Technica | 2026-08-26