Gemini 3.5 Transcribe supports real-time streaming and 85 languages

ammaar · x · 2026-08-27

Retweeting the introduction of Gemini 3.5 Transcribe. The speech-to-text model features smart transcription, function calling, lower WER, custom vocabulary, and multi-speaker identification. It supports over 85 languages and real-time streaming.

Related event: Google Launches Gemini 3.5 Transcribe with 85+ Languages and Smart Post-Processing(17 posts)→

Original post →

More from Multimodal

Multimodal channel →