Google Ships Gemini 3.5 Transcribe, Replacing Chirp 3
On August 26, 2026, Google launched Gemini 3.5 Transcribe—streaming and file speech-to-text with 4.0%/2.6% Artificial Analysis WER, 70% faster time-to-final than Chirp 3, 85+ languages, and live APIs already in Gboard Rambler on Pixel 11.
TLDR
Google on Wednesday, August 26, 2026 published Gemini 3.5 Transcribe, replacing Chirp 3. Artificial Analysis WER: 4.0% streaming / 2.6% non-streaming; FLEURS 5.50% / 5.04%. Google: 70% faster time-to-final vs Chirp 3; 85+ languages; up to 3 speakers on files. API IDs: gemini-3.5-transcribe and gemini-3.5-transcribe-live. Already in Gboard Rambler (Pixel 11), Gemini macOS, and Antigravity mic. @GoogleDeepMind posted the same day.
What shipped
| Item | blog.google Aug 26 |
|---|---|
| Family | Speech-to-text, not a chat Flash |
| Live + batch | Two API surfaces |
| Languages | 85+ |
| Speakers | Up to 3 on files |
Product-line de-dupe: not Pixel 11 hardware (Aug 12), not Gemini 3.7 Flash (Aug 13), not Lyria 3.5 (Jul 29). New speech model lane.
Why this story matters
Google is folding ASR into the Gemini name the way it folded image and video. Watch: whether Chirp disappears from Cloud Speech, and whether live WER holds on noisy call-center audio rather than FLEURS.