Configuration parameters

Here is a list of configuration parameters. Valid values (if any) are shown at the end of each description.

Scope configuration parameters

Application Audit

EnableApplicationAuditLogging

Boolean
Enables application audit logging.

Audio

RedactionMethod

String
Specifies whether to enable redaction feature. The redaction method applies to audio sensitive intervals. One Of:

  • None
  • Mask
  • Redact

Audit

PurgingDays

Integer
The number of days that the system will retain non-enrollment audio and historical engagement data. The system will keep the data for no less than the retention period. At the most, the system will keep the data for one month after the retention period. Until purging is completed, historical records may contain partial data. Must be less than or equal to 550. Must be greater than or equal to 30.

Contact Center

AfterCallWorkModeDuration

Integer
If set to a value greater than 0, agents will be able to perform some operations in the WAC after the call ended for that number of seconds. Setting this to 0 disables the after-call work mode feature. Must be less than or equal to 86400. Must be greater than or equal to 0.

ContinueEnrollForRiskySession

Boolean
Determines if the system will continue enrollment and avoid displaying a suspected fraudster screen, even if the session was evaluated as risky.

ConvoprintLanguage

String
The language to use for transcription, in the form “en-US” or “eng-USA”. If not set, it will use the value set in “Transcription, Language” of the active configuration set (DefaultConfigSetName, VerifyConfigSetName, etc.)

ConvoprintLanguageKey

String
The session variable key that contains the language to use for the current call, in the form “en-US” or “eng-USA”. If not found, ConvoprintLanguage or the value of “Transcription, Language” of the active configuration set will be used.

ConvoprintLanguageMapping

If the language code is found in a session variable using ConvoprintLanguageKey, then use this mapping table to replace its value to a supported language code. You can find the list of language codes here: https://docs.mix.nuance.com/languages/#languages  .

sourceLanguageCode

If the code matches exactly this value, then replace it with targetLanguageCode. This is case-sensitive. String.

targetLanguageCode

The replacement for the language code. String.

DefaultConfigSetName

String
Name of the default Configset.

DetectFraudConfigSetName

String
The Configset that the system uses for fraud and anti-spoofing detection.

DetectFraudOnEnroll

Boolean
Enables fraudster detection prior to enrollment.

DetectFraudOnLocked

Boolean
Enables fraudster detection if a Person’s Voiceprint is locked.

DetectFraudOnMismatch

Boolean
Enables fraudster detection in case the verification result is mismatch.

DetectFraudOnOptedOut

Boolean
Enables fraudster detection at the beginning of a call when the Person has already opted out in a previous session.

DetectFraudOnOptingOut

Boolean
Enables fraudster detection when opt out is requested prior the completion of authentication or enrollment.

DeviceprintTagKey

String
The session variable that has the tag to use to identify the caller’s device print. This should uniquely identify the caller’s device, for example its ANI.

DeviceprintTagNotFound

String
The tag to use for the device print if DeviceprintTagKey is not set or cannot be found.

EnableCallerIdValidationKey

String
If this is set, enable Caller ID validation only if this session variable matches EnableCallerIdValidationValue at the beginning of a call.

EnableCallerIdValidationValue

String
If EnableCallerIdValidationKey is set, enable Caller ID validation only if this session variable matches this value at the beginning of a call.

EnableCallerIdWatchlistRiskAssessment

Boolean
Enables Caller ID watchlist-based risk assessment.

EnableCallerIdWatchlistRiskAssessmentKey

String
If this is set, enable Caller ID watchlist-based risk assessment only if this session variable matches EnableCallerIdValidationValue at the beginning of a call. Changing the value of the session variable later in the call is not supported.

EnableCallerIdWatchlistRiskAssessmentValue

String
If EnableCallerIdWatchlistRiskAssessmentKey is set, enable Caller ID watchlist-based risk assessment only if this session variable matches this value at the beginning of a call.

EnableConvoprint

Boolean
Enable the use of conversation prints.

EnableConvoprintKey

String
If this is set, enable the use of conversation prints only if this session variable matches EnableConvoprintValue at the beginning of a call.

EnableConvoprintValue

String
If EnableConvoprintKey is set, enable the use of conversation prints only if this session variable matches this value at the beginning of a call.

EnableConvoprintEnrollment

Boolean
Enable enrollment of conversation prints when a voice print is enrolled. This is ignored if EnableConvoprint is set to false.

EnableConvoprintDetectKnownFraudsters

Boolean
If set to true, known fraudsters detection will also make use of conversation prints. For this to work, the selected Watchlist must contain at least one person that has a conversation print with a tag that matches the current language. This is ignored if EnableConvoprint is set to false.

EnableDeviceprint

Boolean
Enable the use of device prints.

EnableDeviceprintKey

String
If this is set, enable the use of device prints only if this session variable matches EnableDeviceprintValue at the beginning of a call.

EnableDeviceprintValue

String
If EnableDeviceprintKey is set, enable the use of device prints only if this session variable matches this value at the beginning of a call.

EnableDeviceprintEnrollment

Boolean
Enable enrollment of device prints when a voice print is enrolled. This is ignored if EnableDeviceprint is set to false.

EnableDeviceprintDetectKnownFraudsters

Boolean
If set to true, known fraudsters detection will also make use of device prints. For this to work, the selected Watchlist must contain at least one person that has a device print. This is ignored if EnableConvoprint is set to false.

EnrollConfigSetName

String
The Configset that the system uses for enrollment.

ExternalPersonIdKey

String
The session variable key that the system uses as the external person ID.

ExternalPersonIdValidationRegex

String
If ExternalPersonIdKey is defined, then the person ID value will be used only if its entire string matches this regular expression.

FlowVersion

String
Name or version of the business flow currently in use.

IdentifyConfigSetName

String
The Configset that the system uses for voiceprint identification.

OptOutDurationOnFraudsterDetected

Integer
If a fraudster is detected in a session where the Agent opted in the speaker, then the speaker will be opted out and the Agent will be reminded to attempt to get the speaker consent’s again in this number of days. Setting this to 0 means the Agent will be reminded in the next call. Must be greater than or equal to 0.

OptOutReasonOnFraudsterDetected

String
If a fraudster is detected in a session where the Agent opted in the speaker, then the speaker will be opted out with this reason.

PerformAudioSpoofingDetection

Boolean
Enables audio spoofing detection prior to enrollment and on verification match. With FlowVersion “dkf-v1”, enables audio spoofing detection on every call.

PerformAudioSpoofingDetectionOnEnroll

Boolean
Enables audio spoofing detection prior to enrollment if this or PerformAudioSpoofingDetection is set to true.

PerformAudioSpoofingDetectionOnMatch

Boolean
Enables audio spoofing detection on verification match if this or PerformAudioSpoofingDetection is set to true.

PerformAudioSpoofingDetectionOnLocked

Boolean
Enables audio spoofing detection if a Person’s Voiceprint is locked.

PerformAudioSpoofingDetectionOnOptedOut

Boolean
Enables audio spoofing detection at the beginning of a call when the Person has already opted out in a previous session.

PerformAudioSpoofingDetectionOnOptingOut

Boolean
Enables audio spoofing detection when opt out is requested prior the completion of authentication or enrollment.

PerformAudioSpoofingUsingMediaSegment

Boolean
If set to true, audio spoofing detection will using the call’s media segment rather than the already processed audio.

PersonIdOnlyFromExternalPersonIdKey

Boolean
If true, ignore the person ID submitted by an agent or from any sources other than the key set in ExternalPersonIdKey.

ReAuthenticateAfterMatchDelay

Integer
After this number of seconds, if set above 0, perform authentication again with the same person when successfully matched. Must be less than or equal to 86400. Must be greater than or equal to 0.

VerificationCheckpointForMoreFactors

Integer
When performing voice print verification, wait until this verification checkpoint is reached (counting starting with 1) before performing verification using additional factors (for example, conversation print). If set to 0, verification will start verifying all factors right away. This value is ignored if no other verification factor was enabled (for example EnableConvoprint). Must be greater than or equal to 0.

VerificationMaxWaitTimeForMoreFactors

Integer
When more verification factors start (see VerificationCheckpointForMoreFactors), the maximum amount of time, in seconds, to wait for the additional factors to complete. Must be greater than or equal to 1.

VerifyConfigSetName

String
The Configset that the system uses for verification.

VoiceprintTag

String
The associated Voiceprint tag.

Watchlist

String
The Watchlist to be used for Detect Known Fraudster operations.

WatchlistVoiceprintTag

String
The Voiceprint tag that the system uses with the associated Watchlist.

Contact Center Integration

AuthenticationResultKVP

String
The CTI custom data key to use to store the authentication results. The possible values are: “Verification Match”, “Verification Mismatch” or “Verification Fraudster Detected”. The CTI custom data will not be set if there was some unexpected error during the authentication process.

AutoEnrollSessionKey

String
If set, the person will be automatically enrolled (opted in) when the enrollment process starts if this session variable is set to the expected value (see AutoEnrollValue). This will also disable the “Enroll” button, unless DisableEnrollButtonSessionKey is set.

AutoEnrollValue

String
If set, the value of AutoEnrollSessionKey must match this value to perform the enrollment. If not set, any non-empty value will match.

AutoOptOutDaysToOfferOptIn

Integer
If an automatic opt out is done through AutoOptOutSessionKey, set the number of days (integer) when the agent should offer to opt in the person again. Use the value of -1 to never offer opt in. Values below -1 will be ignored. Must be greater than or equal to -1.

AutoOptOutDaysToOfferOptInSessionKey

String
If an automatic opt out is done through AutoOptOutSessionKey, set the number of days (integer) when the agent should offer to opt in the person again to the value of this session variable, or the value of AutoOptOutDaysToOfferOptIn if the session variable value is not found, empty, not a decimal integer or below -1. Must be greater than or equal to -1.

AutoOptOutReason

String
If an automatic opt out is done through AutoOptOutSessionKey, set the opt out reason to this string.

AutoOptOutReasonSessionKey

String
If an automatic opt out is done through AutoOptOutSessionKey, set the opt out reason the value of this session variable, or the value of AutoOptOutReason if the session variable value is empty or not found. It is highly recommended to enable DetectFraudOnOptingOut when using this configuration.

AutoOptOutSessionKey

String
If set, the person will be automatically opted out if this session variable is set to the expected value (see AutoOptOutValue). This will also disable the “Opt out” button, unless DisableOptOutButtonSessionKey is set.

AutoOptOutValue

String
If set, the value of AutoOptOutSessionKey must match this value to opt out the person. If not set, any non-empty value will match.

BusinessLogicWebhookConfig

String
Custom Business Logic Webhook configuration in JSON form.

BusinessLogicWebhookType

String
Custom Business Logic Webhook Type. One Of:

  • none
  • gRPC
  • REST

CallerIdKeyForWatchlistRiskAssessment

String
The session variable to use as the caller ID for the caller ID watchlist.

DisableEnrollButtonSessionKey

String
If set, the “Enroll” button will be disabled if this session variable matches the expected value, and is enabled otherwise. Setting this to a non-existing session variable will effectively always enable the “Enroll” button.

DisableEnrollButtonValue

String
If set, the value of DisableEnrollButtonSessionKey must match this value to disable the “Enroll” button. If not set, any non-empty value will match.

DisableOptOutButtonSessionKey

String
If set, the “Opt out” button will be disabled if this session variable matches the expected value, and is enabled otherwise. Setting this to a non-existing session variable will effectively always enable the “Opt out” button.

DisableOptOutButtonValue

String
If set, the value of DisableOptOutButtonSessionKey must match this value to disable the “Opt out” button. If not set, any non-empty value will match.

DisableRestartAudioForCallTypes

Array
Disable the “Restart Audio” button in the WAC for the following call types. Any Of:

  • INBOUND
  • OUTBOUND
  • EXTERNAL
  • INTERNAL
  • CONSULT
  • TRANSFER
  • CONFERENCE

EnableAudioAuthenticationAndFraudKey

String
If this is set, enable voice biometrics (authentication, enrollment, detect known fraudster, etc.) only if this session variable matches EnableAudioAuthenticationAndFraudValue at the beginning of a call. Disabling voice biometrics also disables audio recording.

EnableAudioAuthenticationAndFraudValue

String
If EnableAudioAuthenticationAndFraudKey is set, enable voice biometrics only if the value of the session variable set in EnableAudioAuthenticationAndFraudKey matches this value at the beginning of a call.

EnableCallerIdValidation

Boolean
Enables Caller ID validation.

EngagementCorrelationKeys

String
The session variable keys found in CALL_START_EVENT that can be used to find the engagement correlation key. This is required to group multiple sessions in the same engagement and pass biometric results from one session to another. More than one key can be set by separating them with commas, in which case each key will be searched for in the CALL_START_EVENT, in order, until the first non-empty value is found.

EnrollmentResultKVP

String
The CTI custom data key to use to store the enrollment results. The possible values are: “Enrollment Started”, “Enrollment Success”, “Enrollment Failed” or “Enrollment Fraudster Detected”.

ExternalPersonIdCtiEventTypes

Array
If ExternalPersonIdKey starts with “cti.”, then accept the new external person ID value only if the call event is one of the following types. Any Of:

  • CALL_EVENT_UNSPECIFIED
  • CALL_STARTED
  • CALL_ENDED
  • CALL_HOLD
  • CALL_RESUMED
  • CALL_PARTY_ADDED
  • CALL_PARTY_DELETED
  • CALL_DATA_UPDATED
  • AUDIO_CHANNEL_CHANGED
  • UNRECOGNIZED

GKEngagementIdentifierKey

String
The session variable key found in CALL_START_EVENT that has the Gatekeeper engagement ID to use for this session.

IgnorePreviousSessionDecisionAtSessionStart

Boolean
If true, the current engagement’s session decision will be ignored when the call session starts, which forces the current call session to perform biometric operations again.

IgnoredCallTypes

Array
The call types to ignore. Ignored calls won’t be recorded automatically at call start. Any Of:

  • INBOUND
  • OUTBOUND
  • EXTERNAL
  • INTERNAL
  • CONSULT
  • TRANSFER
  • CONFERENCE

LightRulesConfig

String
Light Rules configuration in JSON form.

MaxRecordingDuration

Integer
Maximum recording duration, in seconds. A value equal or below 0 means unlimited recording duration.

PersonIDKVP

String
The CTI custom data key to use to store the Person ID.

RecordingIdentifierKey

String
The session variable that the system will use as recording identifier to start recording. Typically used with SIPREC as a correlation ID with CTI. If not set, this will default to using the recording_id provided by the Media Manager connector, if available.

RecordingSide

String
The channel to be recorded. One Of:

  • RECORDING_SIDE_1
  • RECORDING_SIDE_2
  • RECORDING_SIDE_ALL

RestartSessionEndsConference

Boolean
A restart session (“restart audio”) executed during a conference implies that the conference already ended.

SetKVPCallIdentifierKey

String
The session variable to use as the Call Identifier when writing CTI custom data (also called “KVP”). If not set, this defaults to the value set in RecordingIdentifierKey.

SpeakerChannel

Integer
The channel to use for biometric operations.

WebhookAbortFlowOnError

Boolean
If there was an error with the Custom Business Logic Webhook, report the error to the agent and stop the current call’s flow.

Contact Center Post Call

PostCallAudioSpoofingDetectionOnEnroll

Boolean
Controls if Gatekeeper will perform post-call anti-spoofing detection following an enrollment.

PostCallAudioSpoofingDetectionOnLockedOrOptedOut

Boolean
Controls if Gatekeeper will perform post-call anti-spoofing detection following the event that a claimed speaker is already either opted out or their voiceprint is locked.

PostCallAudioSpoofingDetectionOnMatch

Boolean
Controls if Gatekeeper will perform post-call anti-spoofing detection following a match in voiceprint verification.

PostCallAudioSpoofingDetectionOnMismatch

Boolean
Controls if Gatekeeper will perform post-call anti-spoofing detection following a mismatch result in speaker verification.

PostCallAudioSpoofingDetectionOnOptingOut

Boolean
Controls if Gatekeeper will perform post-call anti-spoofing detection following the operation of a speaker opting out.

PostCallAudioSpoofingDetectionOnRiskySession

Boolean
Controls if Gatekeeper will perform post-call anti-spoofing detection for a session evaluated as risky, for example fraud was detected.

PostCallAudioSpoofingDetectionOnSuspicious

Boolean
Controls if Gatekeeper will perform post-call anti-spoofing detection following a call marked as suspicious.

PostCallDetectFraudConfigSetName

String
The Configset that the system uses for fraud and anti-spoofing detection after the call ended.

PostCallDetectFraudOnEnrollment

Boolean
Controls if Gatekeeper will perform post-call fraud detection following a speaker enrollment operation (includes partial enrollments).

PostCallDetectFraudOnLockedOrOptedOut

Boolean
Controls if Gatekeeper will perform post-call fraud detection following the event that a claimed speaker is already either opted out or their voiceprint is locked.

PostCallDetectFraudOnMatch

Boolean
Controls if Gatekeeper will perform post-call fraud detection following a match in voiceprint verification.

PostCallDetectFraudOnMismatch

Boolean
Controls if Gatekeeper will perform post-call fraud detection following a mismatch result in speaker verification.

PostCallDetectFraudOnOptingOut

Boolean
Controls if Gatekeeper will perform post-call fraud detection following the operation of a speaker opting out.

PostCallDetectFraudOnRiskySession

Boolean
Controls if Gatekeeper will perform post-call fraud detection for a session evaluated as risky, for example fraud was detected.

PostCallDetectFraudOnSpoofingDetected

Boolean
Controls if Gatekeeper will perform post-call fraud detection if audio spoofing was detected.

PostCallDetectFraudOnSuspicious

Boolean
Controls if Gatekeeper will perform post-call fraud detection following a call marked as suspicious.

PostCallWatchlist

String
The Watchlist to be used for post-call Detect Known Fraudster operations.

PostCallWatchlistVoiceprintTag

String
The Voiceprint tag that the system uses with the associated Watchlist for post-call Detect Known Fraudster operations.

Data Export

EnableDataExport

Boolean
Enables data export.

ExportCsvHeaderPrefix

String
CSV header prefix for data export.

ExportCustomData

Array
Custom data configuration for data export.

SchemaName

Schema name. String. One Of:

  • ent-engagements
  • ent-fraudsters
  • ent-groups
  • ent-media-segments
  • ent-persons
  • ent-processed-audio
  • ent-sessions
  • ent-voiceprints
  • ent-watchlists

CustomKeys

Comma delimited custom data keys to export. String.

Engagement

MaxNumberOfSessionsPerEngagement

Integer
Maximum number of Sessions per Engagement. Must be less than or equal to 100. Must be greater than or equal to 1.

Post Call

NearTermAdaptationConfigSetId (adaptation manager)

String
The Configset that the system uses for near term adaptation. if empty, near term adaptation will not be performed.

NearTermAdaptationConfigSetId (post call activities manager)

String
The Configset that the system uses for near term adaptation. if empty, near term adaptation will not be performed.

PostCallDetectAudioSpoofingOnOrphanConfigSetId

String
The Configset that the system uses for audio spoofing detection after the call ended.

PostCallDetectFraudVoiceprintOnOrphanConfigSetId

String
The Configset that the system uses for fraud detection after the call ended.

RedactionConfigSetId (adaptation manager)

String
The Configset that the system uses for redaction. if empty, redaction will use the configset used in the session.

RedactionConfigSetId (post call activities manager)

String
The Configset that the system uses for redaction. if empty, redaction will use the configset used in the session.

Preload

CalibrationModelPreloadList (voiceprints processor)

String
Comma separated list of the calibration models to be loaded as soon as the voiceprints processor starts.

WatchlistPreloadList (voiceprints processor)

String
Comma separated list of the watchlists to be loaded as soon as the voiceprints processor starts.

Training

AutoTrainThreads (voiceprints processor)

Float
Threads used to perform parallel autotrain of the prints of a group when PrintUpgradeMode is enabled. 0 means that prints will be retrained sequentially. Must be less than or equal to 10. Must be greater than or equal to 0.

EnablePrintUpgradeMode (voiceprints processor)

Boolean
If set to true, it enables the print upgrade mode for the voiceprints processor, used when all the voiceprints of a scope must be upgraded to a new major engine version. Warning: enabling this feature increases the processing time of verifications/identifications.

VoiceprintForcedUpdates (voiceprints processor)

Float
Reserved for future use. Currently not used. Must be greater than or equal to 0.

Web Agent Console

Web Agent Console Grouping Mode

Web Agent Console Language Settings

Web Agent Console Platform Integration

Web Application

EngagementsPageSize

Float
Number of engagements to fetch in the Engagements Page (paging) Must be less than or equal to 100. Must be greater than or equal to 1.

SuggestedConvoprintLanguages

String
A comma separated list of Convoprint languages with an optional friendly text delimiter that the system will suggest to the Administrative web interface user for various operations Must be of pattern ^([a-zA-Z0-9_,-| ()]+$).

SuggestedConvoprintTags

String
A comma separated list of Convoprint tags that the system will suggest to the Administrative web interface user for various operations Must be of pattern ^([a-zA-Z0-9_,-]+$).

SuggestedDeviceprintTags

String
A comma separated list of Deviceprint tags that the system will suggest to the Administrative web interface user for various operations Must be of pattern ^([a-zA-Z0-9_,-]+$).

SuggestedVoiceprintTags

String
A comma separated list of Voiceprint tags that the system will suggest to the Administrative web interface user for various operations Must be of pattern ^([a-zA-Z0-9_,-]+$).

Operational configuration parameters

Adaptation

NearTermAdaptationInterval (adaptation manager)

String
Minimum time between the first train and a near term adaptation. Use s,m,h,d as suffix to specify the time unit.

NearTermAdaptationInterval (post call activities manager)

String
Minimum time between the first train and a near term adaptation. Use s,m,h,d as suffix to specify the time unit.

NearTermAdaptationProfileTag (adaptation manager)

String
Profile tag to be used for fraudster detection in near term adaptation.

NearTermAdaptationProfileTag (post call activities manager)

String
Profile tag to be used for fraudster detection in near term adaptation.

NearTermAdaptationWatchlist (adaptation manager)

String
The Watchlist that the system will use for fraudster detection in near term adaptation. if empty, fraudster detection will not be executed.

NearTermAdaptationWatchlist (post call activities manager)

String
The Watchlist that the system will use for fraudster detection in near term adaptation. if empty, fraudster detection will not be executed.

Antispoofing Audio

AudioTooLoudThreshold (antispoofing processor)

Float
Number of saturated samples per minute that defines an audio segment as too loud. Must be greater than or equal to 0.

AudioTooNoisyThreshold (antispoofing processor)

Float
Energy difference (active-silent) threshold that defines an audio segment as too noisy (SNR in dB). Must be greater than or equal to 0.

AudioTooSoftThreshold (antispoofing processor)

Float
Energy threshold defining audio segments as too quiet. Must be greater than or equal to 0.

MinNetAudioLengthInMilliseconds (antispoofing processor)

Integer
Minimum audio length (in milliseconds) required for any audio-based operation. Must be greater than or equal to 100.

StopOnLastCheckpoint (antispoofing coordinator)

Boolean
Specifies if the system should stop processing audio when it reaches the last checkpoint (when set to TRUE), or continue until audio end (when set to FALSE). Default value is TRUE

UseExactCheckpoints (antispoofing coordinator)

Boolean
Determines if the system should return intermediate results for the exact amount of net audio specified by the checkpoints (when set to TRUE), or for all available audio (when set to FALSE). Default value is TRUE

Antispoofing Checkpoints

CheckpointsForAntiSpoofing (antispoofing coordinator)

Array
Checkpoints that the system uses for spoofing detection.

netAudioMilliseconds

Milliseconds of net audio required to reach the checkpoint. Integer.

reliabilityThreshold

If reliability is below this value, the system cannot reach a decision. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

fraudThreshold

If risk is above this threshold, the decision will be Fraud. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

Audio

AudioModelCreationEnabled (deviceprints processor)

Boolean
Specifies whether to create and save audiomodels for future use.

AudioModelCreationEnabled (voiceprints processor)

Boolean
Specifies whether to create and save audiomodels for future use.

EnableSegmentation (voiceprints processor)

Boolean
Specifies whether to perform segmentation to detect number of speakers. It applies only to ProcessAudio method.

MaxProcessedAudioLengthSeconds (ncslengine)

Integer
Maximum length (gross audio) of an audio segment (in seconds) that can be processed by the system. Must be greater than or equal to 1.

MaxProcessedAudioLengthSeconds (nvslengine)

Integer
Maximum length (gross audio) of an audio segment (in seconds) that can be processed by the system. Must be greater than or equal to 1.

MaxProcessedAudioLengthSeconds (nvslengineclustering)

Integer
Maximum length (gross audio) of an audio segment (in seconds) that can be processed by the system. Must be greater than or equal to 1.

RejectMultiSpeakersAudio (voiceprints processor)

Boolean
Specifies whether to consider multiple speakers as an error condition. It applies only to ProcessAudio method.

BackwardSearch

BackwardSearchVoiceprintCalibrationModelId (backward search coordinator)

String
Voiceprint Calibration Model used for Backward Search.

EnableAutomaticPublish (backward search coordinator)

Boolean
If true, backward search result will be published automatically.

EngagementPagingSize (backward search coordinator)

Integer
Paging size used when retrieving engagements. Must be less than or equal to 500. Must be greater than or equal to 1.

ExcludeFraudsterEnrollmentAudio (backward search coordinator)

Boolean
If set to true, audio used for enrollment of the target fraudster is excluded from the backward search collection.

MaxNumberOfCompatibleVoiceprints (backward search coordinator)

Integer
Maximum number of compatible voiceprints considered for Backward Search. Must be greater than or equal to 1.

MaxNumberOfEngagements (backward search coordinator)

Integer
Maximum number of engagements considered for Backward Search. Must be greater than or equal to 1.

MaxNumberOfResults (backward search coordinator)

Integer
Maximum number of results returned by a Backward Search operation. Must be greater than or equal to 1.

MinRankingBiometricScoreThreshold (backward search coordinator)

Float
Used only when ranking is based on Biometric Score. Only results with a ranking biometric score above this threshold will be considered. Must be greater than or equal to 0.

MinRankingRiskThreshold (backward search coordinator)

Float
Used only when ranking is based on Risk. Only results with a ranking risk above this threshold will be considered. Must be greater than or equal to 0.

RankingSubGrouping (backward search coordinator)

Integer
Size of the group used for retrieving engagements during ranking Must be less than or equal to 500. Must be greater than or equal to 1.

SelectedVoiceprintTag (backward search coordinator)

String
Tag of the fraudster voiceprint selected for Backward Search.

Billing

Caller ID Risk Assessment

CallerIdRiskAssessmentFraudThreshold

Float
Risk assessments that score above this threshold indicate that results are fraudulent. Must be less than or equal to 1000. Must be greater than or equal to 0.

CallerIdRiskAssessmentReliabilityThreshold

Float
If reliability is below this threshold, the decision is uncertain. Must be less than or equal to 100. Must be greater than or equal to 0.

Caller ID Validation

CallerIdValidationCacheTimeMinutes

Float
Number of minutes that Gatekeeper will wait before clearing Caller Id Validation results from the cache. Must be greater than or equal to 1.

CallerIdValidationFraudThreshold

Float
Risk assessments that score above this threshold indicate that results are fraudulent. Must be less than or equal to 1000. Must be greater than or equal to 0.

CallerIdValidationReliabilityThreshold

Float
If reliability is below this threshold, the decision is uncertain. Must be less than or equal to 100. Must be greater than or equal to 0.

Clustering

ClusterPurityMaxClusterSize (clustering processor)

Integer
Max size of clusters considered for Purity confidence. Only clusters with size smaller than or equal to this value will be considered. Must be greater than or equal to 3.

ClusterPurityMaxClusterSize (clustering processor)

Integer
Max size of clusters considered for Purity confidence. Only clusters with size smaller than or equal to this value will be considered. Must be greater than or equal to 3.

ClusterPurityMinClusterSize (clustering processor)

Integer
Minimal size of clusters considered for Purity confidence. Only clusters with size greater than or equal to this value will be considered. Must be greater than or equal to 2.

ClusterPurityMinClusterSize (clustering processor)

Integer
Minimal size of clusters considered for Purity confidence. Only clusters with size greater than or equal to this value will be considered. Must be greater than or equal to 2.

ClusteringCalibrationModelId (clustering coordinator)

String
Calibration Model used for Clustering

ClusteringThreshold (clustering processor)

Float
Threshold for speaker identification that is performed for clustering. Must be less than or equal to 200. Must be greater than or equal to -200.

ClusteringThreshold (clustering processor)

Float
Threshold for speaker identification that is performed for clustering. Must be less than or equal to 200. Must be greater than or equal to -200.

EnableAutomaticThreshold (clustering processor)

Boolean
Calculate the speaker clustering threshold automatically. If set to false, the internal threshold is assigned the value of the SpeakerClusteringThreshold parameter.

EnableAutomaticThreshold (clustering processor)

Boolean
Calculate the speaker clustering threshold automatically. If set to false, the internal threshold is assigned the value of the SpeakerClusteringThreshold parameter.

EnableFastClustering (clustering processor)

Boolean
Enables the new fast clustering algorithm.

EnableFastClustering (clustering processor)

Boolean
Enables the new fast clustering algorithm.

EngagementPagingSize (clustering coordinator)

Integer
Paging size used when retrieving engagements. Must be less than or equal to 500. Must be greater than or equal to 1.

MaxNumberOfClusters (clustering coordinator)

Integer
Maximum number of results returned by a Clustering operation. Must be greater than or equal to 1.

MaxNumberOfCompatibleVoiceprints (clustering coordinator)

Integer
Maximum number of compatible voiceprints considered for Clustering. Must be greater than or equal to 1.

MaxNumberOfEngagements (clustering coordinator)

Integer
Maximum number of engagements considered for Clustering. Must be greater than or equal to 1.

MinClusterSize (clustering coordinator)

Integer
Minimal cluster size. Only clusters greater than this value will be considered valid. Must be greater than or equal to 1.

MinRankingScoreThreshold (clustering coordinator)

Float
Only clusters with a ranking score above this threshold will be considered. Must be greater than or equal to 0.

ProgressDecimationFactor (clustering processor)

Integer
Used to reduce the number of progress notification, only 1 progress notification out of this value will be sent. Example: if the value is 20, only 1 progress out of 20 will be sent. Must be greater than or equal to 1.

ProgressDecimationFactor (clustering processor)

Integer
Used to reduce the number of progress notification, only 1 progress notification out of this value will be sent. Example: if the value is 20, only 1 progress out of 20 will be sent. Must be greater than or equal to 1.

ProgressMinInterval (clustering processor)

Integer
Minimum interval in seconds between progress notifications. Must be greater than or equal to 0.

ProgressMinInterval (clustering processor)

Integer
Minimum interval in seconds between progress notifications. Must be greater than or equal to 0.

RankingWeights (clustering coordinator)

Array
Different weights to be used for the ranking criteria

differentPersons

Weight to be used to increase the rank if there are different Claimed Person in the cluster. Float. Must be less than or equal to 1000. Must be greater than or equal to 0.

fraudVerdict

Weight to be used to increase the rank if the verdict of one engagement of the cluster is Fraud. Float. Must be less than or equal to 1000. Must be greater than or equal to 0.

averageRiskNormalized

Weight to be used to increase the rank according to the normalized risk (average risk of the engagments/1000). Float. Must be less than or equal to 1000. Must be greater than or equal to 0.

fraudDecision

Weight to be used to increase the rank if the decision of one engagement of the cluster is Fraud. Float. Must be less than or equal to 1000. Must be greater than or equal to 0.

unknownANI

Weight to be used to increase the rank if the ANI is unknown in at least one engagement. Float. Must be less than or equal to 1000. Must be greater than or equal to 0.

differentANI

Weight to be used to increase the rank if there are different ANI in the cluster. Float. Must be less than or equal to 1000. Must be greater than or equal to 0.

clusterPurity

Weight to be used to increase the rank according to the cluster purity. Float. Must be less than or equal to 1000. Must be greater than or equal to 0.

Comparison

MaxResults (convoprints processor)

Float
Maximum number of results returned by a biometric comparison. Must be greater than or equal to 1.

MaxResults (deviceprints processor)

Float
Maximum number of results returned by a biometric comparison. Must be greater than or equal to 1.

MaxResults (voiceprints processor)

Float
Maximum number of results returned by a biometric comparison. Must be greater than or equal to 1.

Convoprints Checkpoints

CheckpointsForAuthentication (convoprints coordinator)

Array
Checkpoints that the system uses for authentication. Checkpoints can either be based on rawQuality or QualityCheck metrics.

quality

Raw quality of text required to reach the checkpoint. Float.

reliabilityThreshold

If reliability is below this value, the system cannot reach a decision. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

authenticThreshold

If risk is below this threshold, the decision will be Authentic. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

CheckpointsForFraudDetection (convoprints coordinator)

Array
Checkpoints that the system uses for fraud detection. Checkpoints can either be based on rawQuality or QualityCheck metrics.

quality

Raw quality of text required to reach the checkpoint. Float.

reliabilityThreshold

If reliability is below this value, the system cannot reach a decision. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

fraudThreshold

If risk is above this threshold, the decision will be Fraud. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

Convoprints Text

StopOnFirstMatch (convoprints coordinator)

Boolean
Specifies if the system should stop processing text when a Match decision is reached, or continue until text end.

StopOnLastCheckpoint (convoprints coordinator)

Boolean
Specifies if the system should stop processing text when it reaches the last checkpoint, or continue until text end.

UseExactCheckpoints (convoprints coordinator)

Boolean
Determines if the system should return intermediate results for the amount of text specified by the checkpoints, or for all available text.

Deviceprints Audio

AudioTooLoudThreshold (deviceprints processor)

Float
Number of saturated samples per minute that defines an audio segment as too loud. Must be greater than or equal to 0.

AudioTooNoisyThreshold (deviceprints processor)

Float
Energy difference (active-silent) threshold that defines an audio segment as too noisy (SNR in dB). Must be greater than or equal to 0.

AudioTooSoftThreshold (deviceprints processor)

Float
Energy threshold defining audio segments as too quiet. Must be greater than or equal to 0.

MinNetAudioLengthInMilliseconds (deviceprints processor)

Integer
Minimum audio length (in milliseconds) required for any audio-based operation. Must be greater than or equal to 100.

MinNonSpeechLengthInMilliseconds (deviceprints processor)

Integer
Minimum length (in milliseconds) of audio that is non speech (like noise, silence, etc) required for any deviceprint audio-based operation. Must be greater than or equal to 100.

StopOnFirstMatch (deviceprints coordinator)

Boolean
Specifies if the system should stop processing audio when a Match decision is reached (when set to TRUE), or continue until audio end (when set to FALSE). Default value is TRUE

StopOnLastCheckpoint (deviceprints coordinator)

Boolean
Specifies if the system should stop processing audio when it reaches the last checkpoint (when set to TRUE), or continue until audio end (when set to FALSE). Default value is TRUE

UseExactCheckpoints (deviceprints coordinator)

Boolean
Determines if the system should return intermediate results for the exact amount of net audio specified by the checkpoints (when set to TRUE), or for all available audio (when set to FALSE). Default value is TRUE

Deviceprints Checkpoints

CheckpointsForAuthentication (deviceprints coordinator)

Array
Checkpoints that the system uses for authentication.

netAudioMilliseconds

Milliseconds of net audio required to reach the checkpoint. Integer.

reliabilityThreshold

If reliability is below this value, the system cannot reach a decision. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

authenticThreshold

If risk is below this threshold, the decision will be Authentic. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

CheckpointsForFraudDetection (deviceprints coordinator)

Array
Checkpoints that the system uses for fraudster detection.

netAudioMilliseconds

Milliseconds of net audio required to reach the checkpoint. Integer.

reliabilityThreshold

If reliability is below this value, the system cannot reach a decision. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

fraudThreshold

If risk is above this threshold, the decision will be Fraud. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

Engagement

EngagementTimeoutSec

Integer
Engagement timeout in seconds. When the timeout is reached the system will close the Engagement. Must be less than or equal to 2592000. Must be greater than or equal to 0.

Licensing

Liveness Detection

LivenessDetectionPrompts (antispoofing coordinator)

Array
Prompts that the system will use for liveness detection.

expectedText

The expected text in the user’s response. String.

ivrInfo

The information that the IVR needs in order to specify the prompt. String.

NCSL

AudioSaturationLevel (ncslengine)

Integer
The NCSL saturation level Must be greater than or equal to 0.

EnableDTMFToneElimination (ncslengine)

Boolean
Determines if the system performs DTMF tone elimination.

EnableFastVAD (ncslengine)

Boolean
Enables acceleration of Voice Activity Detection processing.

EnableLowLatencyMode (ncslengine)

Boolean
Enables getting faster results in realtime text-independent operations.

EnableNoiseReduction (ncslengine)

Boolean
Enables NCSL noise reduction.

EnableSignalDCOffsetRemoval (ncslengine)

Boolean
Determines if the system will remove any Direct Current bias offset from audio segments.

GainControlMaxValue (ncslengine)

Float
Sets the maximum amplification level for gain control. Must be greater than or equal to 1.

GainControlTargetValue (ncslengine)

Float
Sets the target amplification level for gain control. Must be greater than or equal to 1.

IgnoreAfterDTMF (ncslengine)

Boolean
Ignore all audio after the first DTMF detection.

IgnoreAfterFax (ncslengine)

Boolean
Ignore all audio after fax detection.

IgnoreAudioAfterSilence (ncslengine)

Boolean
Ignore all audio after fax detection.

MinNetAudioBeforeIgnore (ncslengine)

Float
Minimum net audio (in seconds) required before silence audio termination is activated. Must be less than or equal to 5. Must be greater than or equal to 0.

SilenceLengthForIgnore (ncslengine)

Float
Specifies the minimum silence length (in seconds) after which audio is ignored. Must be less than or equal to 5. Must be greater than or equal to 0.

SlidingWindowSize (ncslengine)

Integer
Defines the sliding window size (in terms of the number of seconds of net audio) on which verification calculation is performed. Value zero means use all available audio. Must be greater than or equal to 0.

NTSL

DisableScoreNormalization (ntslengine)

Boolean
Determines if the biometric engine will perform score normalization.

EnableDebugLogs (ntslengine)

Boolean
Enables engine debug logs.

MinNumOfDistinctTokensToIdentify (ntslengine)

Integer
Minimum number of distinct tokens to Identify (it applies to Verify, Identify and DetectKnownFraudsters APIs). Must be greater than or equal to 1.

MinNumOfDistinctTokensToTrain (ntslengine)

Integer
Minimum number of distinct tokens to Train (it applies to ProcessText, Train, Verify, Identify and DetectKnownFraudsters APIs, when a Convoprint or TextModel are created). Must be greater than or equal to 1.

NVSL

AudioSaturationLevel (nvslengine)

Integer
The NVSL saturation level Must be greater than or equal to 0.

AudioSaturationLevel (nvslengineclustering)

Integer
The NVSL saturation level Must be greater than or equal to 0.

DisableScoreNormalization (nvslengine)

Boolean
Determines if the biometric engine will perform score normalization.

DisableScoreNormalization (nvslengineclustering)

Boolean
Determines if the biometric engine will perform score normalization.

EnableDTMFToneElimination (nvslengine)

Boolean
Determines if the system performs DTMF tone elimination.

EnableDTMFToneElimination (nvslengineclustering)

Boolean
Determines if the system performs DTMF tone elimination.

EnableFastSSD (nvslengine)

Boolean
Enables acceleration of Synthetic Speech Detection processing.

EnableFastSSD (nvslengineclustering)

Boolean
Enables acceleration of Synthetic Speech Detection processing.

EnableFastVAD (nvslengine)

Boolean
Enables acceleration of Voice Activity Detection processing.

EnableFastVAD (nvslengineclustering)

Boolean
Enables acceleration of Voice Activity Detection processing.

EnableLowLatencyMode (nvslengine)

Boolean
Enables getting faster results in realtime text-independent operations.

EnableLowLatencyMode (nvslengineclustering)

Boolean
Enables getting faster results in realtime text-independent operations.

EnableMusicDetection (nvslengine)

Boolean
Determines if the system performs music detection.

EnableMusicDetection (nvslengineclustering)

Boolean
Determines if the system performs music detection.

EnableMusicElimination (nvslengine)

Boolean
Determines if the system performs music elimination.

EnableMusicElimination (nvslengineclustering)

Boolean
Determines if the system performs music elimination.

EnableNoiseReduction (nvslengine)

Boolean
Enables NVSL noise reduction.

EnableNoiseReduction (nvslengineclustering)

Boolean
Enables NVSL noise reduction.

EnableSignalDCOffsetRemoval (nvslengine)

Boolean
Determines if the system will remove any Direct Current bias offset from audio segments.

EnableSignalDCOffsetRemoval (nvslengineclustering)

Boolean
Determines if the system will remove any Direct Current bias offset from audio segments.

GainControlMaxValue (nvslengine)

Float
Sets the maximum amplification level for gain control. Must be greater than or equal to 1.

GainControlMaxValue (nvslengineclustering)

Float
Sets the maximum amplification level for gain control. Must be greater than or equal to 1.

GainControlTargetValue (nvslengine)

Float
Sets the target amplification level for gain control. Must be greater than or equal to 1.

GainControlTargetValue (nvslengineclustering)

Float
Sets the target amplification level for gain control. Must be greater than or equal to 1.

IgnoreAfterDTMF (nvslengine)

Boolean
Ignore all audio after the first DTMF detection.

IgnoreAfterDTMF (nvslengineclustering)

Boolean
Ignore all audio after the first DTMF detection.

IgnoreAfterFax (nvslengine)

Boolean
Ignore all audio after fax detection.

IgnoreAfterFax (nvslengineclustering)

Boolean
Ignore all audio after fax detection.

IgnoreAudioAfterSilence (nvslengine)

Boolean
Ignore all audio after fax detection.

IgnoreAudioAfterSilence (nvslengineclustering)

Boolean
Ignore all audio after fax detection.

MinNetAudioBeforeIgnore (nvslengine)

Float
Minimum net audio (in seconds) required before silence audio termination is activated. Must be less than or equal to 5. Must be greater than or equal to 0.

MinNetAudioBeforeIgnore (nvslengineclustering)

Float
Minimum net audio (in seconds) required before silence audio termination is activated. Must be less than or equal to 5. Must be greater than or equal to 0.

ScoreCompensationMode (nvslengine)

String
Improves score stability based on SNR and speech length variable (application to mobile frameworks). One Of:

  • none
  • fixedFA
  • fixedFR
  • fixedRatioFAFR

ScoreCompensationMode (nvslengineclustering)

String
Improves score stability based on SNR and speech length variable (application to mobile frameworks). One Of:

  • none
  • fixedFA
  • fixedFR
  • fixedRatioFAFR

SegmentationConfiguration (nvslengine)

String
Configuration used for segmentation.

SegmentationConfiguration (nvslengineclustering)

String
Configuration used for segmentation.

SegmentationDiscardSilence (nvslengine)

Boolean
Determines if the biometric engine will discard silence when performing segmentation.

SegmentationDiscardSilence (nvslengineclustering)

Boolean
Determines if the biometric engine will discard silence when performing segmentation.

SegmentationEnableRefinementStep (nvslengine)

Boolean
Enables the speaker segmentation refinement operation that improves the diarization accuracy.

SegmentationEnableRefinementStep (nvslengineclustering)

Boolean
Enables the speaker segmentation refinement operation that improves the diarization accuracy.

SegmentationProducePerSpkAudioFeatures (nvslengine)

Boolean
Enables the API for speaker segmentation results that returns audio quality figures per speaker.

SegmentationProducePerSpkAudioFeatures (nvslengineclustering)

Boolean
Enables the API for speaker segmentation results that returns audio quality figures per speaker.

SegmentationSilenceToleranceForClustering (nvslengine)

Integer
Tolerance clustering parameter for advanced tuning of meeting speaker diarization. Must be less than or equal to 2000. Must be greater than or equal to -2000.

SegmentationSilenceToleranceForClustering (nvslengineclustering)

Integer
Tolerance clustering parameter for advanced tuning of meeting speaker diarization. Must be less than or equal to 2000. Must be greater than or equal to -2000.

SegmentationSimilarityHighPrecisionMode (nvslengine)

Boolean
When enabled, the biometric engine operates in high precision mode, increasing the accuracy of the segmentation similarity score.

SegmentationSimilarityHighPrecisionMode (nvslengineclustering)

Boolean
When enabled, the biometric engine operates in high precision mode, increasing the accuracy of the segmentation similarity score.

SegmentationSimilarityThreshold (nvslengine)

Float
Threshold for deciding whether multi-speaker audio segments have 1 or 2 speakers. Must be less than or equal to 100. Must be greater than or equal to -100.

SegmentationSimilarityThreshold (nvslengineclustering)

Float
Threshold for deciding whether multi-speaker audio segments have 1 or 2 speakers. Must be less than or equal to 100. Must be greater than or equal to -100.

SilenceLengthForIgnore (nvslengine)

Float
Specifies the minimum silence length (in seconds) after which audio is ignored. Must be less than or equal to 5. Must be greater than or equal to 0.

SilenceLengthForIgnore (nvslengineclustering)

Float
Specifies the minimum silence length (in seconds) after which audio is ignored. Must be less than or equal to 5. Must be greater than or equal to 0.

SlidingWindowSize (nvslengine)

Integer
Defines the sliding window size (in terms of the number of seconds of net audio) on which verification calculation is performed. Value zero means use all available audio. Must be greater than or equal to 0.

SlidingWindowSize (nvslengineclustering)

Integer
Defines the sliding window size (in terms of the number of seconds of net audio) on which verification calculation is performed. Value zero means use all available audio. Must be greater than or equal to 0.

Playback

MaxNumberOfFootprints (antispoofing processor)

Float
Maximum number of audio footprints to consider for playback detection. Must be greater than or equal to 0.

PlaybackCalibrationModelId (antispoofing processor)

String
Calibration Model for playback detection.

Post Call

EnrollAndConsentCleanupUponFraud

Boolean
Determine whether the system should cleanup the Person consent or the prints enrollments that been performed during the Engagement, upon Fraud post call Session decision.

PostCallDetectChannelPlaybackOnOrphanEnabled

Boolean
Enable channel playback detection after the call ended.

PostCallDetectFraudVoiceprintOnOrphanVoiceprintTag

String
The Voiceprint tag that the system uses with the associated Watchlist.

PostCallDetectFraudVoiceprintOnOrphanWatchlistId

String
The Watchlist to be used for Detect Known Fraudster operations.

PostCallDetectSyntheticSpeechOnOrphanEnabled

Boolean
Enable synthetic speech detection after the call ended.

ResetEngagementVerdictForPostCall

Boolean
Determine whether the system should reset the Engagement verdict upon Fraud post call Session decision.

LockDuration (convoprints coordinator)

String
Duration before an automatically locked Print is automatically unlocked. Use s,m,h,d as suffix to specify the time unit. Value zero means the Print will not be unlocked automatically.

LockDuration (deviceprints coordinator)

String
Duration before an automatically locked Print is automatically unlocked. Use s,m,h,d as suffix to specify the time unit. Value zero means the Print will not be unlocked automatically.

LockDuration (voiceprints coordinator)

String
Duration before an automatically locked Print is automatically unlocked. Use s,m,h,d as suffix to specify the time unit. Value zero means the Print will not be unlocked automatically.

MismatchesBeforeLock (convoprints coordinator)

Integer
Number of consecutive mismatches before a Print is locked. Value zero means the Print will never be locked. Must be greater than or equal to 0.

MismatchesBeforeLock (deviceprints coordinator)

Integer
Number of consecutive mismatches before a Print is locked. Value zero means the Print will never be locked. Must be greater than or equal to 0.

MismatchesBeforeLock (voiceprints coordinator)

Integer
Number of consecutive mismatches before a Print is locked. Value zero means the Print will never be locked. Must be greater than or equal to 0.

Redaction

RedactionForceTranscription (transcription coordinator)

Boolean
Specifies whether redact should transcribe even if a transcription is already present.

RedactionRegEx (transcription coordinator)

String
A regular expression used to detect the transcribed words that need to be redacted.

RedactionReplacement (transcription coordinator)

String
Pattern to use to replace a redacted word. One Of:

  • ReplaceWithEmpty
  • ReplaceWithAsterisks
  • ReplaceWithCount
  • ReplaceWithHash

Risk Engine

AuthenticThreshold

Integer
Risk assessments that score below this threshold indicate that results are authentic. Must be less than or equal to 1000. Must be greater than or equal to 0.

FraudThreshold

Integer
Risk assessments that score above this threshold indicate that results are fraudulent. Must be less than or equal to 1000. Must be greater than or equal to 0.

ReliabilityThreshold

Integer
If reliability is below this threshold, the decision is uncertain. Must be less than or equal to 100. Must be greater than or equal to 0.

RiskEngineCalibrationModelId

String
Risk engine Calibration Model that the system uses to calculate the risk.

Session

AdministrationSessionTimeoutSec

Integer
Administration Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

BackwardSearchJobSessionTimeoutSec

Integer
Backward search job Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

BackwardSearchSessionTimeoutSec

Integer
Backward search Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

CallCenterSessionTimeoutSec

Integer
Call center Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

ClusteringJobSessionTimeoutSec

Integer
Clustering job Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

ClusteringSessionTimeoutSec

Integer
Clustering Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

DigitalSessionTimeoutSec

Integer
Digital Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

FraudAnalystSessionTimeoutSec

Integer
Fraud analyst Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

IVRSessionTimeoutSec

Integer
IVR Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

OfflineMediaProcessingSessionTimeoutSec

Integer
Offline media processing Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

OtherSessionTimeoutSec

Integer
Other Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

PostCallSessionTimeoutSec

Integer
Post call Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

PreCallSessionTimeoutSec

Integer
Pre call Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

WEBSessionTimeoutSec

Integer
Web Session timeout in seconds. When the timeout is reached the system will close the Session. Must be less than or equal to 2592000. Must be greater than or equal to 0.

Supervisor

Text

TextModelCreationEnabled (convoprints processor)

Boolean
Specifies whether to create and save textmodels for future use.

Text Validation

ASRSpeechDetectionSensitivity

Float
Determines how the system should balance between detecting speech and noise. Zero means ignore all noise. One means interpret all noise as speech. Must be less than or equal to 1. Must be greater than or equal to 0.

ASRSuppressCallRecording

Boolean
Determines if the system should suppress call recording in the Nuance ASR service Call Log Aggregator (CLAG). By default call logs, metadata, and audio are collected.

ASRTopic

String
Specialized language model in the Nuance ASR service data pack. Must be less than or equal to 255.

TextValidationDecoyGrammar

String
Grammar template for decoy grammar generation (rejected sentences). Must be less than or equal to 32000.

TextValidationDecoySentences

String
List of sentences for decoy grammar generation such as sentences that the system should reject. If empty, decoy grammar is not generated internally. Must be less than or equal to 32000.

TextValidationDynamicGrammar

String
Grammar template for Text Validation using Nuance ASR service. Must be less than or equal to 32000.

TextValidationFraudThreshold

Float
Risk assessments that score above this threshold indicate that results are fraudulent. Must be less than or equal to 1000. Must be greater than or equal to 0.

TextValidationLanguage

String
Language code for Text Validation using Nuance ASR service. Must be less than or equal to 255.

TextValidationReliabilityThreshold

Float
If reliability is below this threshold, the decision is uncertain. Must be less than or equal to 100. Must be greater than or equal to 0.

Training

ConsistencyCheckThreshold (deviceprints processor)

Float
Threshold for checking consistency of audio created from a set of audio enrollment segments. Must be less than or equal to 1000. Must be greater than or equal to -1000.

ConsistencyCheckThreshold (voiceprints processor)

Float
Threshold for checking consistency of audio created from a set of audio enrollment segments. Must be less than or equal to 1000. Must be greater than or equal to -1000.

ConvoprintCalibrationModelId (convoprints processor)

String
Calibration Model for training Convoprints.

DeviceprintCalibrationModelId (deviceprints processor)

String
Calibration Model for training Deviceprints.

EnableConsistencyCheck (deviceprints processor)

Boolean
Determines if the system will detect inconsistent audio among enrollment segments.

EnableConsistencyCheck (voiceprints processor)

Boolean
Determines if the system will detect inconsistent audio among enrollment segments.

MaxNetAudioForEnrollMilliseconds (deviceprints coordinator)

Integer
Maximum total net audio (not including silence) allowed for a single Enrollment Segment. Must be greater than or equal to 1000.

MaxNetAudioForEnrollMilliseconds (voiceprints coordinator)

Integer
Maximum total net audio (not including silence) allowed for a single Enrollment Segment. Must be greater than or equal to 1000.

MaxWatchlistInvalidityResults (voiceprints processor)

Float
Maximum number of invalidity results returned by a watchlist check. Must be less than or equal to 100. Must be greater than or equal to 1.

MinNetAudioForEnrollMilliseconds (deviceprints coordinator)

Integer
Minimum total net audio (not including silence) required for a single Enrollment Segment. Must be greater than or equal to 0.

MinNetAudioForEnrollMilliseconds (voiceprints coordinator)

Integer
Minimum total net audio (not including silence) required for a single Enrollment Segment. Must be greater than or equal to 0.

MinNetAudioForTrainMilliseconds (deviceprints coordinator)

Integer
Minimum total net audio (not including silence) required for automatically training a Deviceprint. Must be greater than or equal to 100.

MinNetAudioForTrainMilliseconds (voiceprints coordinator)

Integer
Minimum total net audio (not including silence) required for automatically training a Voiceprint. Must be greater than or equal to 100.

MinNonSpeechForEnrollMilliseconds (deviceprints coordinator)

Integer
Minimum total non speech (including silence and noise) required for a single Enrollment Segment. Must be greater than or equal to 0.

MinNonSpeechForTrainMilliseconds (deviceprints coordinator)

Integer
Minimum total non speech (including silence and noise) required for automatically training a Deviceprint. Must be greater than or equal to 100.

MinSegmentsForOptimalTraining (convoprints coordinator)

Integer
Minimum number of Enrollment Segments required for training a Convoprint at an optimal level. Must be less than or equal to 30. Must be greater than or equal to 1.

MinSegmentsForOptimalTraining (deviceprints coordinator)

Integer
Minimum number of Enrollment Segments required for training a Deviceprint at an optimal level. Must be less than or equal to 30. Must be greater than or equal to 1.

MinSegmentsForOptimalTraining (voiceprints coordinator)

Integer
Minimum number of Enrollment Segments required for training a Voiceprint at an optimal level. Must be less than or equal to 30. Must be greater than or equal to 1.

MinTrainingSegments (deviceprints coordinator)

Integer
Minimum number of Enrollment Segments required for training a Deviceprint at a basic level. Must be less than or equal to 30. Must be greater than or equal to 1.

MinTrainingSegments (voiceprints coordinator)

Integer
Minimum number of Enrollment Segments required for training a Voiceprint at a basic level. Must be less than or equal to 30. Must be greater than or equal to 1.

UseModelsForTrain (convoprints processor)

Boolean
Determines if the Train should use existing Text Models or reprocess the whole text. Text Models usage is faster, while text reprocessing is more accurate.

VoiceprintCalibrationModelId (voiceprints processor)

String
Calibration Model for training Voiceprints.

Transcription

AutoPunctuate (transcription coordinator)

Boolean
Specifies whether to enable auto punctuation, if available for the language.

DlmUri (transcription coordinator)

String
Location of a DLM resource as a URN reference. e.g. ‘urn:nuance-mix:tag:model/names-places/mix.asr?=language=eng-USA’.

DlmWeight (transcription coordinator)

Float
Weight of DLM or builtin as a numeric value from 0 to 1. Default 0. Must be less than or equal to 1.0. Must be greater than or equal to 0.

FormattingScheme (transcription coordinator)

String
Keyword for a formatting type defined in the data pack. e.g. ’num_as_digits’. This value is if not specified in the transcription methods. If empty, a data pack dependent default is used.

InlineWordSet (transcription coordinator)

String
Inline wordset JSON resource. See documentation for the format. Default blank, meaning no inline wordset.

Language (transcription coordinator)

String
Language and country (locale) code as xx-XX, e.g. ’en-US’ or ’eng-USA’ for American English. This value is used if not specified in the transcription methods. If empty, ‘actual_language’ audio property is used.

NoInputTimeoutMilliseconds (transcription coordinator)

Integer
Maximum silence allowed while waiting for user input after recognition timers are started. Default (0) means server default, usually no timeout. Must be greater than or equal to 0.

SpeechDetectionSensitivity (transcription coordinator)

Float
A balance between detecting speech and noise (breathing, etc.), 0 to 1. 0 means ignore all noise, 1 means interpret all noise as speech. Default is 0.5. Must be less than or equal to 1.0. Must be greater than or equal to 0.

Topic (transcription coordinator)

String
Specialized language model in data pack. Case-sensitive, uppercase. This value is if not specified in the transcription methods. If empty, ‘GEN’ (generic) is used.

UtteranceEndSilenceMilliseconds (transcription coordinator)

Integer
Minimum silence that determines the end of an utterance. Default (0) means server default, usually 500ms or half a second. Must be greater than or equal to 0.

Voiceprints Audio

AudioTooLoudThreshold (voiceprints processor)

Float
Number of saturated samples per minute that defines an audio segment as too loud. Must be greater than or equal to 0.

AudioTooNoisyThreshold (voiceprints processor)

Float
Energy difference (active-silent) threshold that defines an audio segment as too noisy (SNR in dB). Must be greater than or equal to 0.

AudioTooSoftThreshold (voiceprints processor)

Float
Energy threshold defining audio segments as too quiet. Must be greater than or equal to 0.

MinNetAudioLengthInMilliseconds (voiceprints processor)

Integer
Minimum audio length (in milliseconds) required for any audio-based operation. Must be greater than or equal to 100.

StopOnFirstMatch (voiceprints coordinator)

Boolean
Specifies if the system should stop processing audio when a Match decision is reached (when set to TRUE), or continue until audio end (when set to FALSE). Default value is TRUE

StopOnLastCheckpoint (voiceprints coordinator)

Boolean
Specifies if the system should stop processing audio when it reaches the last checkpoint (when set to TRUE), or continue until audio end (when set to FALSE). Default value is TRUE

UseExactCheckpoints (voiceprints coordinator)

Boolean
Determines if the system should return intermediate results for the exact amount of net audio specified by the checkpoints (when set to TRUE), or for all available audio (when set to FALSE). Default value is TRUE

Voiceprints Checkpoints

CheckpointsForAnalyzeAudio (voiceprints coordinator)

Array
Checkpoints that the system uses for combined audio analysis.

netAudioMilliseconds

Milliseconds of net audio required to reach the checkpoint. Integer.

reliabilityThreshold

If reliability is below this value, the system cannot reach a decision. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

fraudThreshold

If risk is above this threshold, the decision will be Fraud. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

authenticThreshold

If risk is below this threshold, the decision will be Authentic. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

CheckpointsForAuthentication (voiceprints coordinator)

Array
Checkpoints that the system uses for authentication.

netAudioMilliseconds

Milliseconds of net audio required to reach the checkpoint. Integer.

reliabilityThreshold

If reliability is below this value, the system cannot reach a decision. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

authenticThreshold

If risk is below this threshold, the decision will be Authentic. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

CheckpointsForFraudDetection (voiceprints coordinator)

Array
Checkpoints that the system uses for fraudster detection.

netAudioMilliseconds

Milliseconds of net audio required to reach the checkpoint. Integer.

reliabilityThreshold

If reliability is below this value, the system cannot reach a decision. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.

fraudThreshold

If risk is above this threshold, the decision will be Fraud. Integer. Must be less than or equal to 1000. Must be greater than or equal to 0.