Talkmap, the leading generative AI platform for contact center Conversational Intelligence used by some of the largest mobile service providers and financial services institutions, is thrilled to announce the release of version 8.3 of the Talkdiscovery platform. Talkdiscovery 8.3 revolutionizes the way businesses understand and engage with their customers, providing a comprehensive suite of tools designed to drive growth, enhance customer satisfaction, and streamline operations.
Based on two years of research and feedback from customers, Talkmap has updated its patented self-supervised generative AI platform to improve the quality of enriched conversations as a data source and to set the stage for a future release in 2024 that will allow contact center managers and data scientists to create or utilize a library of customizable generative AI tasks. In this 8.3 release, customers can immediately take advantage of the improved generative AI platform features including Conversation Summaries and Contact Intents.
Contact Intents is a new conversation level data attribute that gets generated for each conversation. While the concept of a conversation level goal has been part of our underlying conversation model from the beginning, the 8.3 release enables it to be a stand-alone data attribute that identifies the main theme (e.g. the true driver) of the interaction with the contact center. This innovation is a game-changer versus other approaches because for the first time ever, enterprises now have transformational, real-time visibility into their customers, including the ‘silent majority’ of core customers who don’t engage with VOC or other survey attempts. While the platform still identifies all turn-level customer intents (the ‘agent+customer interactions’), this new attribute provides better accuracy than the previous method, including a high-level view of how customer intents change over time.
While Conversation Summaries have been supported by the Talkdiscovery platform for several years, our customers wanted them at a quality level the marketplace had never seen, so we developed it. Our research team built an evaluation framework that solves the challenges of existing summary benchmarks, which do not address the specific needs of conversation summaries in a contact center context. We used this evaluation framework to select the right generative AI LLM, tune the prompts, and evaluate the outputs. Talkmap’s ongoing development efforts will continue to refine the framework and the Conversation Summary output as LLMs continue to improve in quality, size, and efficiency. Conversation Summaries will ultimately become part of a “task library” in a future release that will allow customers to further refine and customize summaries to their specific needs.
For more such updates, follow us on Google News Martech News