A common scenario A customer calls her bank to report a lost credit card. In order to make sure that the caller is the legitimate cardholder, the call center agent requests that she provides the correct answer to a number of knowledge questions, for instance her address, her mother's maiden name, the name of her favorite pet, or any other information available in her file that an impostor would be unlikely to know. We have all been through such authentication dialogs numerous times and we accept this as necessary in order to protect us, as well as the organizations which with we deal, from fraudsters. In fact, it is likely to happen for any call that requires access to confidential information. For instance, to: change a credit limit; change an address; reset a forgotten password; report a lost or stolen cellular phone.

The costs of authentication

There are many situations where simple means of authentication – such as a PIN – cannot be used, either because the caller doesn’t have one or has forgotten it. For the great majority of these cases, caller authentication then needs to be performed by contact center agents, which results in very significant costs to the organization.

Automating identity verification can bring two key benefits to organizations:

• It reduces the amount of time an agent has to spend per call. For large contact centers, this alone can produce million-dollar annual savings.
• It enables the introduction — or increases the usage — of self-service applications requiring caller authentication.

Leveraging its unique expertise in speech application optimization, Nü Echo has developed Nü ID™, a high performance identity verification application that can be tailored for the great majority of organizations.

Towards high accuracy automated caller authentication

The key challenge of automating identity verification with speech recognition technology is to simultaneously provide high automation rate and high security. High automation rate means that the application must use a fast and efficient dialog that minimizes the probability of rejecting correct responses as well as the probability of abandon, both of which will reduce automation. High security means that the application must minimize the probability of accepting a caller that has incorrectly responded to some of the security questions.

Of course, these are conflicting requirements. In order to successfully address this challenge, Nü Echo has spent over three years developing fundamentally novel — and now patent pending — approaches and technologies that today enable Nü ID™ to deliver truly superior performance.

For example, two key performance metrics for an identity verification application are its False Reject (FR) rate —the percentage of correct responses rejected by the application — and its False Accept (FA) rate – the percentage of incorrect responses accepted by the application. Our technologies have allowed us to obtain breakthrough improvements in both the FR and the FA rates, compared with approaches traditionally used in the industry.

Key advantages of Nü ID

Unbeatable automation and security. Nü ID leverages breakthrough patent pending technology in order to provide truly superior performance. Each recognition grammar used by Nü ID has been thoroughly optimized for a specific type of security question. In addition, each recognition result is processed by a neural network classifier specially designed and optimized for the appropriate question type. The result is simply unbeatable authentication accuracy.

Shorter call duration. By avoiding unnecessary confirmations and by only using the number of questions necessary in order to achieve the required security level, Nü ID can complete the authentication process faster and more efficiently than other applications.

High configurability. Nü ID can be configured in order to adapt to a wide variety of situations and/or business rules.

Solidly supported. Since Nü ID is a solidly supported product, clients will be able to benefit from all future enhancements and performance improvements.