AIMarkTechPostabout 2 hours ago
Google Launches Gemini 3.5 Transcribe Speech Model
4 min read
Google released Gemini 3.5 Transcribe, a speech-to-text model with two endpoints: gemini-3.5-transcribe for recorded audio and gemini-3.5-transcribe-live for streaming. It reports 2.6% non-streaming and 4.0% streaming word error rates, 70% lower latency than Chirp 3, and supports 85+ languages with code-switching.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyGoogle
Summary by ByteBrief
