Nvidia Announces General Availability Of Jarvis Framework

Jarvis is a framework for multimodal conversational AI services that delivers real-time performance on GPUs.

Recently, at its GTC conference 2021, Nvidia announced the general availability of its Jarvis framework, which provides you pre-trained AI models and software tools to create interactive conversational experiences. 

Jarvis models, launched in May 2020 in preview, offer automatic speech recognition, as well as real-time language translations, and text-to-speech capabilities for conversational agents.

According to Nvidia Jarvis’ pipeline can be run in under 100 milliseconds and deploy in the cloud, in a datacenter, or at the edge. The Jarvis framework includes models trained on more than 1 billion pages of text and over 60,000 hours of speech. These models can be adjusted, and optimized with custom data, and tailored to different tasks, industries, and systems.

With Jarvis, you can automate the steps that go from pre-trained models to optimized services deployed in the cloud. Jarvis framework, under the hood, applies powerful NVIDIA TensorRT optimizations to models, configures the NVIDIA Triton Inference Server, and exposes the models as a service through a standard API.

In order to deploy, user can use a single command to download, set up, and run the entire Jarvis application or individual services via Helm charts on Kubernetes clusters. The Helm charts can be customized for your requirement and are freely available in NGC.