A conversational AI system engages in humanlike dialog, understands context, and provides intelligent responses. Such AI models are often huge and highly complex. With NetApp® ONTAP® AI, powered by NVIDIA DGX systems and NetApp cloud-connected storage systems, massive, state-of-the-art language models can be trained and optimized to run inference rapidly.
In conversational AI systems, developers can use NetApp Cloud Sync to archive conversation history from the cloud to data centers to enable offline training of natural language processing (NLP) models. By training models to recognize more intents, the conversational AI system is better equipped to manage more complex questions from end users.
Using NVIDIA Jarvis, an end-to-end framework for building conversational AI services, NetApp has built a virtual retail assistant that accepts speech or text input and answers questions about weather, points of interest, and inventory pricing by connecting to the WeatherStack API, the Yelp Fusion API, and the eBay Paython SDK.
For example, the conversational AI system is able to remember conversation flow and ask a follow-up question if the user doesn’t specify a location for weather or points of interest. The system also recognizes complex entities such as “Thai food” or “laptop memory.” It understands natural language questions like “will it rain next week in Boston?” and it can archive conversation history and annotate sentences with intents and slots for NVIDIA NeMo training so that NLP service improves as more users interact with the system. Click this link to watch a demonstration of the NetApp retail assistant (NARA).
In summary, high-quality conversational AI systems can be built on ONTAP AI to allow businesses across verticals to offer previously unattainable personalized services when engaging with customers. To learn more, read the white paper WP-7328: NetApp Conversational AI using NVIDIA Jarvis.