Relay-1 model for speech-to-speech translation
Relay-1 is the model powering Pinch's real-time speech translation API, across 50+ languages, over a single WebSocket connection.
Relay-1 Model
Today we’re releasing Relay-1, the model powering Pinch’s real-time speech translation API. Relay-1 is a unified pipeline that takes streaming microphone audio in one language and returns translated speech and transcripts in another over a single WebSocket connection. It supports more than 50 input and output languages and runs in both speech-to-speech and speech-to-text modes. We built it for live multilingual conversation in meetings, calls, and in-person interactions, where waiting several seconds for a translation isn’t an option.
The model is available now through the Pinch API and our meeting products.
Share this post