-->

Nuance Launches Nina Coach

Nuance Communications, today introduced Nina Coach to enable organizations to train and deploy virtual assistants faster by tapping into artificial intelligence and leveraging existing human knowledge.

Nina Coach equips virtual assistants with pre-existing knowledge, such as chat transcripts and industry-specific dialogue. Nina will leverage AI and human assistance from live engagements to constantly improve the VA's knowledge and natural language understanding (NLU).

Through Nina Coach, when the virtual assistant has low confidence or does not know an answer to a consumer's question, a live chat agent seamlessly enters the conversation, within the same engagement window, and has visibility into the transcript and history of the conversation. The human assistant can then either choose the best answer from a list of choices identified by the virtual assistant, or type in a separate response. As a next step, the action is recorded, analyzed, and folded back into Nina’s semantic brain. This learning loop makes the NLU technology smarter and more accurate over time, so the virtual assistant knows the answer on its own moving forward.

"There is simply no one-size-fits-all when it comes to offering an intelligent virtual assistant for customer service, and simple question/answer bots no longer make the cut. Instead, these customer service virtual assistants, or chatbots, need to be smarter and capable of answering whatever question is posed to them regarding a particular domain," said Robert Weideman, executive vice president and general manager of the Enterprise Division at Nuance, in a statement. "Adding cognitive, data-driven machine learning through Nina Coach will enable organizations to more quickly and accurately understand consumers' intents and tailor content from the start. These more accurate and intuitive conversations build brand loyalty, reduce churn, and ultimately help organizations to obtain a faster ROI."


SpeechTek Covers
Free
for qualified subscribers
Subscribe Now Current Issue Past Issues