Person Details

Configuring person records is required for 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. and 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 employees or customers, and for enrollment and 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. of targets.

Details

Item	Description
Enrollment ID	If enrolled, the system-generated identifier that uniquely identifies the enrollment details.
Employee Name	Provided only for Employees, use to select the person from the organizational hierarchy Hierarchy set up by administrators (via the EM) that is structured according to the managerial and employee hierarchy in the company. as defined in User Management Module that an administrator uses to create a profile for each employee in their organization.. Selection of an Employee Name is required.
Enrollment Name	Provided only for Employees, name displayed in Risk Management Component of the Real-Time Analytics (RTA) Framework that provides interaction search and replay functionality, combined with integration of the Real-Time Analytics (RTA) Framework. to identify the person. Enter any sequence of alphanumeric characters, including space. Default value: Blank Minimum characters: 1, Maximum characters: 100
Name	Provided only for Customers or Targets, the name displayed in Risk Management to identify the person. Enter any sequence of alphanumeric characters, including space. Default value: Blank Minimum characters: 1, Maximum characters: 100
[Risk Management attribute fields]	Field values, derived from CTI attributes, used to match the employee or customer on 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. to a person in voice biometrics. Enter any sequence of alphanumeric and special characters, including space. Default value: Blank Minimum characters: 0, Maximum characters: 30 These fields are not used for watch list List that contains one of more voiceprints of people who are of particular interest to an enterprise. detection. Although you can assign values for targets, they are not used. You cannot assign values for employees.

Voiceprint Model

Use the Voiceprint Model area to add, approve/unapprove, and delete audio segments used by the 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. You can also use the opt-out and opt-in feature to determine which people are eligible for voice biometric analysis.

Approval and unapproval of audio segments requires the Approve/Unapprove Interactions Product that assists contact centers increase operational effectiveness and improves the customer experience through full-time recording, powerful quality monitoring, compliance/liability management, rich reports, and an intuitive, dashboard-style interface. for Enrollment privilege Permissions associated with each role that define the features of the application a user is able to view and the functionality in the application the user can access. assigned to your role in User Management.

To change the columns displayed, use the modify columns features. See the related topics section to learn more.

General details

Item	Description
Description	Provide any text to assist with identification of the audio segment. Enter any sequence of alphanumeric and special characters. Default value: Blank Minimum characters: 1, Maximum characters: 100 Characters not allowed: Backslash (\)
Audio	Select an audio file you want to import from your file system to the Recorder. Use imported audio to create or enhance a voiceprint model. Enter any sequence of alphanumeric characters or spaces. Default value: Blank Minimum characters: 1, Maximum characters: 425
Channel	For stereo audio, select the channel to use for the audio segment. Stereo Left / Mono: Use the stereo left / mono channel for what is typically the employee voice in stereo recording environments. Always use the left channel for mono recording environments. Stereo Right: Use the stereo right channel for what is typically the customer voice in stereo recording environments.
Training Status	Shows the current training status of the voiceprint model. See the table that follows to learn more.

Audio Segments

Item	Description
	Click to search for interactions. Use to locate interactions you want to use to create or enhance the voiceprint model.
	Approve audio segment. Use after you have reviewed the currently selected audio segment and want to approve its use for the voiceprint model. This option is available only to users with the role privilege of Approve/Unapprove Interactions for Enrollment.
	Unapprove audio segment. Use after you have reviewed the currently selected audio segment and want to exclude its use for the voiceprint model. This option is available only to users with the role privilege of Approve/Unapprove Interactions for Enrollment.
	Delete audio segment. Use to remove the currently selected audio segment from the voiceprint model.
Assignment	Indicates how the audio segment was assigned to the person. Assignment is one of the following: Automatic Manual
Channel	Identifies the channel used for the voiceprint model. Stereo Left / Mono: Use the stereo left / mono channel for what is typically the employee voice in stereo recording environments. Always use the left channel for mono recording environments. Stereo Right: Use the stereo right channel for what is typically the customer voice in stereo recording environments.
Currently In Use	Provides status of the audio segment in the voiceprint model. Yes: segment is approved and is part of the model. No: segment is not approved or approved but not used in the model. If approved but not in use, a reason for its omission displays. Example: No (Low Net Speech) The channel used lacks sufficient voice content for use by the model.
Description	Text used that uniquely identifies the audio segment.
Enrollment Audio	Total recorded length of the approved audio from the interaction usable for the voiceprint model. Hidden by default. Use the modify columns feature to configure which columns display for you. The interaction audio can consist of multiple segments. Voices of unrelated parties are not included in Enrollment Audio. Enrollment Audio is displayed as hh:mm:ss where: hh: Hours, shown only when the length exceeds 59 minutes and 59 seconds. mm: Minutes, 00-59; displays 0 if the enrollment audio is less than 60 seconds. ss: Seconds, 00-59.
INum	Unique identifier for the recording segment associated with the enrollment. Hidden by default. Use the modify columns feature to configure which columns display for you.
Net Speech	The total amount of the approved enrollment audio from the recording channel that is usable for the voiceprint model. One of the following icons displays for approved audio: Good: Over 30 seconds of net speech found for the person. Fair: 7–30 seconds of net speech found for the person. Bad: Less than 7 seconds of net speech found for the person. To view the exact amount of net speech found, point to the icon. The net speech is displayed as hh:mm:ss.sss where: hh: Hours, shown only when the length exceeds 59 minutes and 59 seconds. mm: Minutes, 00-59; displays 00 if the enrollment audio is less than 60 seconds. ss: Seconds, 00-59. sss: Partial seconds, 000-999
Saturation (%)	Amount of signal clipping (distortion) found in the approved enrollment audio. High saturation means that the audio is distorted. Hidden by default. Use the modify columns feature to configure which columns display for you. One of the following icons displays for approved audio: Good: Less than 40% saturation. Fair: 40% to 49% saturation. Bad: 50% or more saturation. To see the exact saturation value, point to the icon. The saturation is expressed as a percentage from 0.0% to 100.0%.
SNR (dB)	Signal-to-noise ratio (SNR) in decibels (dB) found in the approved enrollment audio. SNR identifies the background noise level relative to spoken words. A low ratio means that the speech is attenuated or that the noise level is high. Hidden by default. Use the modify columns feature to configure which columns display for you. One of the following icons displays for approved audio: Good: Over 15 dB SNR. Fair: 10.1–15-dB SNR. Bad: 10 dB or lower SNR. To see the exact SNR value, point to the icon. The SNR is expressed in dB, from 0.0 to 99.9.
State	Status of the audio segment for use in the voiceprint model. The state is one of the following: Approved Unapproved
Transaction ID	The Transaction ID is a system identifier that you can use to determine the recording segment used by the suite.

Training status for a voiceprint model

The voiceprint model training status displays on the Voiceprint Model tab of the Person Details page. The status shows the current state of the voiceprint model for a customer, employee, or target.

The following training statuses, listed alphabetically, are possible for a person.

Training status	Description
Cooling Off Period (interval Remaining)	Used for customers only, the voiceprint model is fully trained, but is not used for verification until the period shown elapses. For example: A customer voiceprint model continues to use the cooling-off period when anyone uses Risk Management to: Manually enroll the customer. Manually update the customer voiceprint model. To reset the cooling-off period manually using Risk Management, opt out the customer and then opt in the customer. The cooling-off period is configured on the Voice Enrollment 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. settings.
In Progress	Training of the voiceprint model started, but the system has not yet completed the training. This status is usually temporary.
Invalidated	The voiceprint model has been updated, which means retraining needs to occur. However, the model is awaiting retraining by the system. This status is usually temporary. For example, you add more audio segments to the voiceprint model, so the status changes to Invalidated. Wait a few minutes for the system to retrain the model.
Opted Out	The person has been excluded from all voice biometrics processing (enrollment and verification).
Pending Approval	Interactions are assigned for use in the voiceprint model of the person, but one or more interactions need to be manually approved.
Pending Auto Enrollment	Automatic enrollment found interactions to train the voiceprint model. However, one of the following is true: There are insufficient interactions to complete training. One or more of the interactions found lack sufficient quality to use for training.
Trained	Training is complete and successful. The voiceprint model is used for voice biometric analysis.
Training Failed	The system attempted to train the voiceprint model, but was unable to do so. This status results for any of the following reasons: Already Assigned: This recording has already been used to create a voiceprint for this person. Detected: The audio matches a target in a configured watch list. Inconsistent: The new audio does not match the existing voiceprint. Not Enough Time: The audio was recorded too close to audio already used for the voiceprint. Poor Quality: The audio quality was insufficient for use by enrollment.
Untrained	The voiceprint model for the person has not been trained. This status is the default state for a new person or anyone without audio segments assigned for use in the model.