Sinch Voice API: Choosing between v1 and v2
For all new projects, please use Voice API v2. This is especially important for projects involving:

- AI capabilities like Streams and Voice Relay
- Voice campaigns like Batch calling


You should only use the older Voice API v1 if one of the following is a strict requirement:

- The project must be in General Availability (GA).
- The project requires In-App voice or Conferencing (until these features are available in v2).


# Voice Relay

The Voice Relay WebSocket enables real-time **speech-to-text (STT)** and **text-to-speech (TTS)** interaction between your application and an active Sinch voice call.

Enables real-time voice-to-text and text-to-speech communication between the Sinch platform and customer endpoints. The Voice Relay service transcribes incoming audio to text, relays it to your AI chatbot or agent system for processing, and synthesizes text responses back into audio for playback on active voice channels.

It allows you to:

- Receive live transcriptions of caller audio
- Send text to be synthesized into audio
- Stream audio chunks
- Build conversational AI agents and IVRs


## Download AsyncAPI Spec

You can download the full AsyncAPI specification here:

👉 [Download voice-relay.yml](/docs/voice-2.0/api-reference/voice-relay)