NGC | Catalog
CatalogCollectionsAccelerate Development of Intelligent Virtual Assistants

Accelerate Development of Intelligent Virtual Assistants

For contents of this collection and more information, please view on a desktop device.
Logo for Accelerate Development of Intelligent Virtual Assistants

Description

Quickly develop and provide accurate and engaging English intelligent virtual assistant for contact centers, at a large scale, around the clock, at lower operational costs.

Curator

NVIDIA

Modified

May 23, 2023
Containers
Sorry, your browser does not support inline SVG.
Helm Charts
Sorry, your browser does not support inline SVG.
Models
Sorry, your browser does not support inline SVG.
Resources
Sorry, your browser does not support inline SVG.

Overview

NVIDIA’s intelligent virtual assistant AI solution workflow integrates NVIDIA Riva English automatic speech recognition (ASR) and text-to-speech (TTS) with Haystack, a third-party, open-source natural language processing (NLP) service for information retrieval and question answering (IRQA), and Rasa, an open-source dialog manager.

image1

Example training and inferencing pipelines are built using Riva, Haystack, and Rasa, to demonstrate how to customize and build a virtual assistant solution for a specific use case, in this case financial services. Additional supporting components are also integrated as shown in the below diagram.

image1

The packaged components of this workflow are cloud native and include examples of enterprise-ready best practices for authentication, monitoring, reporting, and load balancing, while allowing a path to deviate.

Key benefits of the Riva intelligent virtual assistant for contact centers AI workflow:

  • The Best Possible Accuracy: Get an intelligent virtual assistant with customized Riva English ASR service for the best possible real-time accuracy for the financial service industry use case.

  • Expressive OOTB English Voices: Provide intelligent virtual assistants with professional English female and male voices with Riva pretrained TTS models.

  • Seamless Scaling: Instantly scale to hundreds and thousands of intelligent virtual assistants with microservices deployable on any cloud Kubernetes distribution. 

  • Fast & Flexible Deployment: Quickly get started with a deployable intelligent virtual assistant in the cloud, on-premises, or on edge.

Getting Started

To get started, review the documentation linked below for more information on what is involved and included in the workflow, and how to deploy and run the workflow.

Join the NVIDIA Developer Program to access a 90-day free trial of NVIDIA Riva and access the intelligent virtual assistant workflow through NGC.

Have an upcoming speech AI project? Try the intelligent virtual assistant AI workflow on NVIDIA LaunchPad.

Contact NVIDIA to find out how you can purchase NVIDIA Riva for your production deployment.

Documentation

Additional Resources

Use the NVIDIA Riva SDK to build your own speech- and translation-AI-based solutions.

Learn more about how to use NVIDIA Riva through our Deep Learning Institute platform. You can work with Riva interactively on hardware provided by NVIDIA hosted in the cloud.

You must have a developer account and be signed in to access the following Riva courses: