← Back to events
ActiveAI

Google's Gemini 3.5 Transcribe turns speech to text in 85 languages while auto-correcting your verbal stumbles

Photo: The Decoder

What happened

Google's new Gemini 3.5 Transcribe recognizes over 85 languages, strips filler words, and corrects slips of the tongue in real time. It hits a 4.0 percent word error rate in streaming mode, with 70 percent lower latency than its predecessor, Chirp 3. Through function calling, the model can hand off tasks to other Gemini models.

Summary assembled by rule from the sources below

Sources