Riva

NVIDIA Riva: GPU-accelerated microservices for building real-time, customizable speech AI applications.

Riva is NVIDIA’s core platform for high-performance conversational AI (Artificial Intelligence). It delivers a collection of GPU-accelerated microservices for production-grade speech applications, ensuring low-latency performance (often below 300 milliseconds). The platform includes Automatic Speech Recognition (ASR), Text-to-Speech (TTS), and Neural Machine Translation (NMT) capabilities. Developers leverage pre-trained models for languages like English, Spanish, and Mandarin, then fine-tune them for domain-specific accuracy. Deployment is flexible: use it across the cloud, data center, or at the edge (on devices like the NVIDIA Jetson AGX Orin) for maximum efficiency and control.

https://developer.nvidia.com/riva