Speakeasy AI Delivers Dynamic Audio Machine Learning That Creates Tailored Language Models Based on Voice of the Customer.
Speakeasy AI announced the release and general availability of its category-defining speech-to-intent machine learning solution, Nerv™. This latest release works seamlessly within the Speakeasy AI solution, including Amplify customer journey reporting. Speakeasy AI can now generate tailored language models to deliver customer intent insights across the breadth of voice customer journeys and engagements, all while enabling the quickest and most robust update cycle in voice automation.
Nerv™ empowers Speakeasy AI users to tailor audio intent recognition by creating a machine learning model that is tailored to a business’ specific customer population. This means how customers talk, what they discuss, and what they need will inform how voice AI is enabled, improved upon and created. The resulting solution provided by Speakeasy AI emulates the training of a human call center agent, but via automated self-learning. There is a training period on the floor when Nerv™ learns about customers and their needs, and all future learning and experiences inform that foundation once Nerv™ is working with the Speakeasy AI solution to solve customers’ inquiries via self-service.
Nerv™’s AI also creates custom transcription models to deliver entity extraction capabilities well beyond traditional templates. This means the proper nouns that represent a business and its customers, products, services, solutions and transactional data can now be custom created and improved upon for accuracy at scale. As a result, customer experience in voice self-service is more accurate and gets better over time. These transcription models are created via Scriber™, a part of Nerv™. Nerv™ offers highly accurate custom transcription models as a by product of properly identified contexts that improve over time. These two neural networks within Nerv™ work together to leverage context and memory to have conversations that get better over time.
“We set out to create the world’s most advanced audio neural network with as much art as science,” said CTO, Moshe Villaizan. “Nerv™, including Scriber™, enables a voice AI solution that truly becomes a subject matter expert for customers in their own voices, which provides a powerful foundation of trust between AI and the business it serves.”
“Voice AI is finally ready to live up to the promise of audio machine learning based on voice of the customer,” said Chief Executive Officer, Frank Schneider, “Moshe and our team have raised the bar with Nerv™ by creating self-learning AI that will drive the industry forward and most importantly, serve our customers and their customers’ needs.”
Nerv™ will be available to customers during Active Listening phase as well as with the full delivery of the Speakeasy AI solution.