NVIDIA Riva is a GPU-accelerated SDK for building Speech AI applications that are customized for your use case and deliver real-time performance.
Riva offers pretrained speech models in NVIDIA NGC™ that can be fine-tuned with the TAO Toolkit on a custom data set, accelerating the development of domain-specific models by 10x.
TAO models can be easily exported, optimized, and deployed as a speech service on premises or in the cloud with a single command using Helm charts.
Riva’s high-performance inference is powered by NVIDIA TensorRT™ optimizations and served using the NVIDIA Triton™ Inference Server, which are both part of the NVIDIA AI platform.
Riva services are available as gRPC-based microservices for low-latency streaming, as well as high-throughput offline use cases.
Riva is fully containerized and can easily scale to hundreds and thousands of parallel streams.
Some of the major benefits that Riva provides are:
Riva is part of the NVIDIA AI platform - built on a decade of AI innovations by NVIDIA across hardware, model architectures, training techniques, inference optimizations, and deployment solutions.
Flexibility at every step, from modifying model architectures to fine-tuning models on your data and customizing pipelines, as well as the ability to deploy on any platform.
Continued optimizations across the entire stack from models to software to hardware delivered 12X the gain versus the previous generation.
What Can You Do with Riva?#
With a few commands, you can access the high-performance services through API operations and try demos. Using Riva, you can easily fine-tune state-of-art models on your data to achieve a deeper understanding of their specific contexts. You can also optimize for inference to offer real-time services that run in 150 milliseconds (ms) compared to the 25 seconds required on CPU-only platforms.
You can use Riva to access highly optimized Automatic Speech Recognition (ASR) and speech synthesis services for use cases like real-time transcription and virtual assistants. The ASR skill is available in multiple languages. It is trained and evaluated on a wide variety of real-world, domain-specific datasets. With telecommunications, podcasting, and healthcare vocabulary, it delivers world-class production accuracy.
You can use Riva’s text-to-speech (TTS) or speech synthesis skills to generate human-like speech. Riva uses nonautoregressive models to deliver 12x higher performance on NVIDIA A100 GPUs compared to Tacotron 2 and WaveGlow models on NVIDIA V100 GPUs. Furthermore, with TTS, you can create a natural custom voice for every brand and virtual assistant with only 30 minutes of an actor’s voice data.
Some of the major tasks that you can perform using Riva are:
Customizing a model with your data Using the NVIDIA TAO Toolkit, you can use a custom-trained model in Riva. NVIDIA TAO Toolkit is a low-coding tool for fine-tuning models on the domain-specific dataset.
Deploying a model in Riva Riva is designed for speech AI at scale. To help you efficiently serve models across different servers robustly, NVIDIA provides push-button model deployment using Helm charts.