What's new in Paris release for Virtual Agent & NLU

[Charles Steiner]

Virtual Agent has been making it faster and easier for everyone to resolve routine IT, HR and customer service requests for a few releases now, by taking chat from conversation to resolution. Workers can use simple, everyday language to get exactly what they need, whenever they need it, thanks to the native natural language understanding.  

In Paris we’ve added functionality to deliver a seamless user experience across multiple bots​, connect live and virtual agents to any messaging app​, and deliver natural, intelligent conversations​ in new languages.

For more be sure to check out Paris release Virtual Agent Academy and the product documentation for Virtual Agent and Natural Language Understanding.

Conversational custom chat integrations
In addition to the out-of-the-box integrations with Slack, Microsoft Teams, Workplace by Facebook, and Facebook Messenger, you can now build your own integrations from live agent and Virtual Agent chats to any messaging app including WhatsApp, Signal, or Telegram.

Conversations can be linked to specific integrations and message content can also be converted upon transmission – giving you the flexibility to configure conversations that deliver the most value to your customers and employees. Custom chat integrations are available with a new API that lets you build adapters that interface with external messaging app APIs.

Virtual Agent API
In addition to these messaging apps, you can now connect both live agent chat and Virtual Agent upstream to primary chatbots like IBM Watson. This means that users can interact with the primary bot, which then passes requests along to one or more secondary bots for processing and response retrieval behind the scenes. This gives end users a seamless, unified experience with a single chatbot – while also letting them benefit from the Virtual Agent & live agent capabilities and connect directly to ServiceNow workflows and data.

The framework behind this functionality consists of scriptable APIs that are invoked via a REST endpoint asynchronously from a 3^rd party chat interface or bot. Key features of this framework include:

• Security through provider authentication and user authorization
• Keyword and NLU support for conversations
• Control transformation to render the conversational experience in the customer's chat interface
• Support for long-running conversations as well real-time messaging

Multi-Language NLU​
We strive to deliver a single conversational chatbot that intelligently learns and responds to human-expressed intent. Up until now English was the only supported language but in Paris, the NLU engine now includes support for German and French language. ​

With models in these languages, administrators can use a single Virtual Agent conversation topic for all languages by mapping the topic’s intents and entities between each language. ​ This expands our support for global organizations that need to deliver chatbot automation to users with different native languages.

Customizable chat menu
Building on the multiple strides made in Virtual Agent branding and customization with the Orlando release, we now support chat menu customization. The chat menu provides more options for contacting support. In Paris we are increasing your ability to brand and customize the chat window with allowing modification to this menu. Add, move, or delete options like links to web pages or FAQs and change the menu title and menu items including icons.

 

---

© 2020 ServiceNow, Inc. All rights reserved. ServiceNow, the ServiceNow logo, Now, and other ServiceNow marks are trademarks and/or registered trademarks of ServiceNow, Inc. in the United States and/or other countries. Other company names, product names, and logos may be trademarks of the respective companies with which they are associated.