kapynAI / Models

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

Gemini 3.5 Transcribe is Google's speech-to-text model for 85 languages with real-time self-correction. It achieves a 4.0 percent word error rate in streaming mode and cuts latency by 70 percent versus Chirp 3. Developers can use function calling to route downstream tasks to other Gemini models.

The Decoder·Aug 27, 2026

Opening Kapyn…