Voice biometric events

When viewing an 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. in the full-screen mode of the Player Application that displays recorded interaction data in audio waveform format, which is a color-coded display of the voice energy levels of the employee and customer. The Player also provides metadata about the interaction., you can use the Tags panel to view voice biometric events.

These events operate independently from settings configured in Recorder Analytics Rules.

Voiceprint model match events

Voiceprint model match events identify every instance when the voice of a person matches an enrolled and trained person (employee, customer, target). Every match with a non-zero score, no matter how low, is registered. You can have multiple events for the same person in a single interaction.

Event

Description

Customer Match

The customer voice on the Interaction yielded a non-zero 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. score for an enrolled and trained customer 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.

Employee Match

The employee voice on the Interaction yielded a non-zero verification score for an enrolled and trained employee voiceprint model.

Target Match

A voice on the Interaction yielded a non-zero detection Function of the IAFD product that compares a caller’s voice to one or more voiceprints in a watch list to detect if a target is participating in a call. score when compared to target voiceprint models.

Condition events

Condition events identify every instance the Interaction would have been eligible for matching a condition setting available in a Recorder Analytics Rule. For example, a target detection or a customer verification could have been used to trigger some action. Use these events when evaluating how the system interprets the spectrum of voice biometric activity Core component of both schedules and time records in Workforce Management (WFM). When an employee performs any kind of work, activities specify the scheduled work and capture employee adherence to their schedule. and to optimize your Recorder Analytics Rules.

Event

Description

Detection

A target match occurred above the detection thresholds configured in a Recorder Analytics Rule and the potential detection threshold set for the Voice Biometrics Engine server role Entity that contains a logical, predefined set of components (system software or certified third-party software) deployed in the Data Center and Site Zones that provide specific functionality for the system..

Example: Configure a rule that notifies your case management team when a target is on a interaction.

Potential Detection

A target match occurred between the detection thresholds configured in a Recorder Analytics Rule and the potential detection threshold set for the Voice Biometrics Engine server role.

Example: Alter a detection threshold for a rule based on the potential detection.

Employee Verified

An employee verification occurred above a certain threshold.

Example: Configure a rule that stores the Interaction in a case when employee verification does occur.

Customer Verified

A customer verification occurred above a certain threshold.

Example: Configure a rule that stores the Interaction in a case when customer verification does occur.

Employee Not Verified

An employee verification did not occur.

If an employee voiceprint exists but verification returns a score of zero, the employee name associated with the voiceprint is provided. This state occurs when the voice on the interaction is not consistent with the voiceprint retrieved for the employee.

Example: Configure a rule that notifies your case management team that the employee voice does not match an existing voiceprint model.

Customer Not Verified

A customer verification did not occur.

If a customer voiceprint exists but verification returns a score of zero, the customer name associated with the voiceprint is provided. This state occurs when the voice on the interaction is not consistent with the voiceprint retrieved for the customer.

Example: Configure a rule that notifies your case management team that the customer voice does not match an existing voiceprint model.

Employee No Voiceprint

The contact The entire communication experience for a customer, from beginning to end. attributes used to identify an employee could not be used to find a voiceprint model.

Example: Configure a rule that notifies your administrator when 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. for the employee is required.

Customer No Voiceprint

The contact attributes used to identify a customer could not be used to find a voiceprint model.

Example: Configure a rule that notifies your administrator when enrollment for the customer is required.

Employee No Attributes

The contact attributes used to identify an employee had no value.

Example: Configure a rule that notifies your administrator when there is problem with your Identification Rules configuration.

Customer No Attributes

The contact attributes used to identify a customer had no value.

Example: Configure a rule that notifies your administrator when there is problem with your Identification Rules configuration.

Other events

Other events refer to general voice biometric activity.

Event

Description

Engine Completed (Employee Channel)

The engine finished processing audio for the employee channel on a stereo recording.

Engine Completed (Other Channel)

The engine finished processing audio for the other channel (customer channel or another employee on an inter-employee Interaction) on a stereo recording.

Processing Error

The voice biometrics system was unable to analyze the channel due to audio quality issues. To determine which channel experienced the error, examine the event details. Audio quality issues include:

  • High Saturation: Channel level for the voice of a person is too high, which results in signal clipping (distortion).

  • Low Net Speech: Insufficient speech on the channel, which can result when the person listens for long periods or speaks little, aside from short or monosyllabic responses.

  • Low S/N: Speech not sufficiently loud relative to background noise.

Replay an interaction