Sayna, an Open-Source Rust Server, Unifies Speech-to-Text and Text-to-Speech Across Deepgram, Google Cloud and Azure

A new open-source project called Sayna aims to collapse the tangle of voice-AI integrations into a single Rust server, offering unified speech-to-text and text-to-speech access across providers including Deepgram, ElevenLabs, Google Cloud and Microsoft Azure. The project, hosted at SaynaAI/sayna on GitHub, exposes one WebSocket and REST interface for real-time voice processing, letting developers switch between vendors without rewriting their applications.