Nuance Communications is releasing a new voice recognition engine that can identify a speaker in less than a second without the use of a passphrase or any other personal information. Dubbed the Lightning Engine, the AI platform is built with deep neural net technology and can be deployed in virtual assistants, smart speakers, and other tools that rely on vocal interaction.
The engine is being pitched to enterprise-level clients, who will then be able to pass on the utility to consumers. Those consumers will create a unique voice profile when registering a new account, at which point the Lightning Engine will be able to verify their identities based on any natural voice sample.
“With the engine, organizations don’t have to spend time checking to see if a customer is who they say they are on voice channels,” said Brett Beranek, the General Manager of Nuance’s Enterprise Security Business. “Instead, they can start personalizing the interaction as soon as that person speaks, making it an effective, enjoyable and natural engagement.”
The Lightning Engine is currently being offered to select clients, with a full launch expected in the fall. Nuance has previously provided secure voice biometric technology for several providers of financial services, including Allied Irish Banks and Santander bank in Spain. The company has argued that its security tech has saved its partners more than $1 billion, and recently sold its SRS business in India to place a greater focus on conversational AI.
July 10, 2019 – by Eric Weiss