Automatic enrollment

Automatic enrollment Process in the IAFD product where an employee, customer, or target is registered and the system is then able to assist with identity verification and watch list detection when that speaker is on a call. uses Recorder Analytics Rules to enter many people in voice biometrics.

During automatic enrollment, the Recorder Analytics Rule selects interactions, combined with a unique identifier, to create a record in the database for the person. The interaction In Speech Analytics, an interaction represents a single part of the contact between one employee and the same customer. In Text Analytics, an interaction is the communication session between one or more employees and the same customer with a unifying contextual element. is used to create the biometric voiceprint File containing a mathematical summary of the vocal patterns of a person’s voice used in the IAFD product to assist with identity verification and detection. model for later verification Feature in voice biometrics where the speaker’s voice is compared to a collection of employee or customer voiceprints that should match the call. If a match occurs, the speaker is said to be verified. of the person.

It is important to understand the following about automatic enrollment:

  • Targets cannot be enrolled automatically. Automatic enrollment applies only to employees and customers.

  • Do not use a single Recorder Analytics Rule to enroll both employees and customers. To use automatic enrollment for both types of people, create a separate Recorder Analytics Rule for each type of enrollment.

  • Always use historical interactions (campaign-based) for automatic enrollment. Do not attempt to use automatic enrollment with real-time interaction selection.

  • For employee and customer interaction selection, use different campaigns. One campaign selects interactions for customers, a separate campaign selects interactions for employees.

Improving accuracy

To improve the quality of the voice biometric analysis and the quality of the voiceprint model, a predefined number of interactions can be required (based on the Recorder Analytics Rule configuration). Enrollment for a person is not complete until:

  • The Recorder Analytics Rule has found the configured number of interactions, the interactions have been approved for use

  • The system updates the interactions to indicate that they are being used by the voiceprint model (Currently In use state).

After the enrollment is complete, the voiceprint model is considered fully trained.

About the unique identifier

A unique identifier is required to reference the voiceprint model.

  • The employee identifier is the employee ID, referred to in some areas of the user interface as Employee Name.

  • A customer identifier could be a reference number used by a business to identify each customer uniquely.

    A customer identifier could be an account number. However, if a customer account is used by more than one person, the account does not fully identify a person. In such instances, use multiple fields for identification. For example, the account number combined with the full name of the person. The combined fields function as a unique identifier.

Call selection

Rules for Identity Authentication and Fraud Detection

Enrollment