A leader in Conversational Service Automation (CSA), Uniphore, announced new enhancements to its portfolio of products using artificial intelligence (AI). The company has now expanded its capabilities to deliver transformative experiences to organizations before, during, and after their initial contact.
Umesh Sachdev, CEO and co-founder of Uniphore, said, “From the beginning, Uniphore has led the industry by focusing on delivering AI + Automation solutions that make a tangible difference in conversations between customers and agents. These latest enhancements help our customers drive transformational experiences by delivering greater intelligence and recommendations through the application of deep learning AI models. I am extremely proud of the work our team has done to deliver these technology innovations for our customers.”
U-Assist, a Uniphore platform that provides for real-time agent assistance, has been enhanced with a deep learning AI model that enhances both agent performance as well as customer experience. Uniphore’s latest AI innovations are centered on adding enhanced intent detection with next best action, an enhanced agent promises model, automated supervisor alerts and automatic feedback loops designed to optimize AI models.
These new enhancements further bolster Uniphore’s industry-leading products and those recently announced, which included offerings such as integrated front-end customer interaction and backend fulfillment from engagements, as well as automated authentication and data protection for its agents.
The latest product enhancements from Uniphore include:
- Modeling AI for inference of intent: In order to identify the intent of a customer, agents and AI models must compare two primary factors: what the customer is saying and how it is said. Using Uniphore’s latest AI models, the company’s coaches can now provide recommendations in real time, based on enhanced sentiment analysis.
- AI Intelligence for Informed Supervisor Alerts: Using AI models, Uniphore combines all the relevant information (what is the state of the call, what is being offered as a resolution, is the agent following the coaching, etc.) to inform when and how supervisors should be proactively alerted to improve engagement. False alerts from supervisors can be costly, so a real intelligence model is critical to preventing them.
