Description
Adding voice to an agent means integrating speech recognition, synthesis and real-time audio separately, rewritten for each provider. Sayna is a unified voice layer for AI agents wrapping STT and TTS behind one interface.
Written in Rust for low latency, it plugs into existing agent frameworks.
One interface:STT and TTS.
Providers:Switch anytime.
Low latency:Rust.
Frameworks:Existing agents.
Written in Rust for low latency, it plugs into existing agent frameworks.
Features
One interface:STT and TTS.
Providers:Switch anytime.
Low latency:Rust.
Frameworks:Existing agents.
