Amazon Connect application data source settings

Settings applicable for the application data source depend on the integration type.

Recorder Settings

Setting

Description

Seating Arrangement

For Application data sources, the Seating Arrangement is Free and cannot be changed. Free Seating indicates that employees do not have permanently assigned workstations. They are assigned an Employee ID and can log in from any location in the call center.

Minimum Session Length (seconds)

To prevent the retention of calls that are so short they are unlikely to contain usable data, specify a Minimum Session Length (seconds). Calls that are shorter than this length are deleted automatically. If set to 0 (zero), no calls are deleted based on this setting. The maximum value is 3600 (or one hour). This setting applies to active CTI Sessions or VOX Sessions. Inactive CTI Sessions can be retained using the Session Auditing Policy.

Long Call Duration (minutes)

This setting allows you to specify the length of a call, in minutes, after which the system will trigger an alarm. The system will also stop tracking the call from a CTI perspective, so in CTI-controlled application or performance mode environments this may cause loss of recording. Enter any number between 1 and 1440 (24 hours)—an alarm is raised in the cases where calls exceed this number of minutes. The default is 120.

The Integration Service runs maintenance checks every five minutes to close calls that last for more than the length of time specified as the Long Call Duration. An alarm will indicate that a call was closed because it was too long. These maintenance checks are not run more frequently to avoid imposing an additional load on the system. Therefore it may take up to five minutes to close a long call after it has passed the defined Long Call Duration threshold.

Long Hold Duration (minutes

This setting allows you to specify the maximum duration of a single hold in minutes. Any holds over this duration will raise an alarm. Enter any number between 1 and 3000, representing the number of minutes. The default is 30.

Session Auditing Policy

Defines the type of session/interaction that should be marked and kept in the system. “Disabled” (the default) only marks sessions/interactions with some kind of content. Two additional options will create a basic entry in the database for calls that occurred but were not recorded: “Missed Recordings” will mark calls that should have been recorded, but were not, while “Full Switch” will mark all sessions/interactions for which we receive CTI without recording (for example, calls that were blocked, or interception calls that were met with a busy tone or unanswered ringtone). You may then search for these types of interactions in the Portal.

Recorded employee segments marked by auditing will appear in playback once all Interactions in the related Contact are closed, after a delay of up to five minutes.

Default Text Language

Specify the default language assigned to text interactions if the language is not provided by the text source.

 

Default employee for interaction capture

In some situations, recordings do not have an employee associated to them. Examples include:

  • IVR recordings, where no employee or phone device is involved.

  • Back-office environments, where phones are shared and not assigned to a specific employee.

To ensure that these recordings can be replayed, associate a Default Employee for interaction capture to the data source.

How the Default Employee works

Assigning a default employee to a data source allows recordings without an explicitly associated employee to be replayed. When this option is configured:

  • Any recordings from the data source that do not have a specific user associated with them are automatically associated with the Default Employee.

  • This association provides access to otherwise unassigned recordings.

Important Behavior and Warnings

After an employee is selected as the Default Employee, an error message appears on the data source screen if that employee has been:

  • Deleted

  • Terminated

  • Moved to a different organization

This error appears the next time the data source is viewed or saved.

Setting

Description

Organization

Displays the organization to which the Default Employee for interactions belongs. The list of available employees is filtered based on the selected organization.

Find

Filters the Available Employees list by the entered search criteria to help locate a specific employee.

Go

Applies the search criteria entered in the Find field and refreshes the Available Employees list.

Available Employees

Lists employees in the selected organization who are eligible to be assigned as the Default Employee for Interactions. Employees with a configured end date are not displayed.

Add (>>)

Moves the selected employee from Available Employees to Default Employee. Only one employee can be assigned as the Default Employee.

Avoid choosing an employee who will soon move or transfer to a different organization. Employees with a configured end date cannot be selected.

Remove (<<)

Removes the employee from Default Employee and returns the employee to the Available Employees list.

Default Employee

Displays the employee currently assigned as the Default Employee.

Pagination Controls

Navigates through multiple pages of employees in the Available Employees list when the list spans more than one page.

Amazon Connect Settings

Setting

Description

Contact Trace Records Stream Name

This field is required only when the data source is associated to an Interaction Capture Service server role. In this field, specify the Kinesis data stream name that the Interaction Capture Service connects to for post-call recording retrieval. This field is blank by default. The field can have a minimum of 1 character and a maximum of 1024 characters.

Agent Events Stream Name

This field is required only when the data source is associated to a Recorder Integration Service server role. In this field, specify the Kinesis data stream name that the Recorder Integration Service connects to for real-time agent and call events. This field is required for real-time audio and screen recording. This field is blank by default. The field can have a minimum of 1 character and a maximum of 1024 characters.

Region

You must select a region from the Region drop-down list. In this drop-down list, select the Amazon Web Service (AWS) region hosting the Amazon Connect call center. This field is blank by default. If you select the Custom Region option, you must specify the custom region in the Custom Region field below.

Custom Region

Complete this field only when you have selected the Custom Region option from the Region drop-down list. You must enter only the region code for custom region, not the entire display name. An example of a Custom Region value is “ap-south-1” for the region "Asia Pacific (Mumbai) ap-south-1". This field is blank by default.

Specify Application Credentials

Select this check box to enable the use of specific application credentials. If you select this check box, you must provide the specific application credentials in the Access Key ID and Access Key Secret fields.

Access Key ID

This field is enabled only when the Specify Application Credentials check box is selected. When this field is enabled, enter the user-specified application access key ID. This field is blank by default. The field can have a minimum of 1 character and a maximum of 1024 characters.

Access Key Secret

This field is enabled only when the Specify Application Credentials check box is selected. When this field is enabled, enter the user-specified application access key secret. This field is blank by default. The field can have a minimum of 1 character and a maximum of 1024 characters.

Specify IAM Role

Select this check box to use a specific application IAM Role. If you select this check box, you must provide the specific IAM Role information in the Role ARN (Amazon Resource Name) and External ID fields. If this check box is not selected, no role is assumed.

Role ARN

This field is enabled only when the Specify IAM Role check box is selected. When this field is enabled, enter the user-specified application IAM Role ARN. This field is blank by default. The field can have a minimum of 1 character and a maximum of 1024 characters.

External ID

This field is enabled only when the Specify IAM Role check box is selected. When this field is enabled, enter the user-specified application IAM Role External ID. This field is blank by default. The field can have a minimum of 1 character and a maximum of 1024 characters.

Role Session Name

When authenticating using an IAM role, you can explicitly configure the role session name. If this name is not configured, then the adapter generates a role session name.

Pause/Resume Recording

For a post-call recording deployment where Amazon Connect is handling the audio recording, enable this setting to pass pause and resume recording commands through to Amazon Connect. The commands are passed through such that screen and audio recordings will both be masked during the same time periods. This setting must be enabled for call recovery.

This can also be configured in a real-time recording solution where Amazon recording is being used as a fallback mechanism. In these environments, pause and resume should also be enabled such that the backup copies recorded in Amazon can also be paused/resumed at the appropriate time periods.

The pause and resume commands must be sent to the Interaction Capture Control API (formerly RIS API), which passes the commands to Amazon Connect.

Amazon Connect Instance ID

This is the instance ID of the Amazon Connect instance. The instance ID is required to send pause and resume recording commands.

Amazon Connect Multi-party Call Mode

Amazon Connect Multi-party Call Mode - Enable this option if your Amazon Connect instance supports multi-party calls and enhanced monitoring for voice. This setting must match your Amazon Connect configuration for recordings to work correctly.

Proxy Settings

When the customer has a proxy host for their Amazon Connect, Ring Central RingEx, Zoom Meetings, or Zoom Phone systems, in the Proxy Settings section, configure the following:

  1. Proxy: Select this option when the tenant uses a proxy host for connection to the internet.

  2. Proxy Scheme: Specify the protocol used by the proxy server to authenticate and manage client requests, such as HTTP or HTTPS.

  3. Proxy Host: Enter the fully-qualified domain name (FQDN) of the proxy server.

  4. Proxy Port: Enter the port number used for communications with the proxy server. The specified port cannot be in use by any other application or adapter on the local server.

  5. Proxy Username: Enter the username the data source will use to authenticate with the proxy server.

  6. Proxy Password: Enter the password for the specified username.

  7. Non Proxy Hosts: Enter the domain names (FQDN) of the host servers that should bypass the proxy server, separated by commas. Requests to these hosts will connect directly to the internet instead of through the proxy. You can use the wildcard * to specify multiple server domains. For example, *.example.com will match all subdomains of example.com.

Post-call Cloud Settings

Internal. The Post‑Call Cloud settings are for the External Cloud Recording Configuration service, and apply to supported Application data sources (currently Amazon Connect). Configuration is performed by Verint Services.

The Post‑Call Cloud Settings section is always visible on the Application data source, regardless of which recording service is deployed.

When using the External Cloud Recording Configuration service, you must configure the Post‑Call Cloud Settings, and no adapter is required.

When using the Recorder service (on-premises deployments and some cloud deployments), the Post‑Call Cloud Settings are ignored, even if configured. To capture post‑call audio in these deployments, you must deploy and configure the appropriate adapter (for example, the Amazon Connect Polling Adapter).

Do not rely on the Post‑Call Cloud Settings when the Recorder service is in use.

The Post‑Call Cloud Settings include the following fields:

  • Application Name: Specifies the DynamoDB table name provided by Verint Services for this deployment. The post‑call capture service uses this value for checkpointing and post‑call event tracking.

    This value must exactly match the value supplied by Verint Services.

  • Record Unknown Agents: Determines whether the system processes post‑call interactions when the agent identified in the interaction record cannot be matched to a configured Verint agent.

    • When enabled, the system captures post‑call audio and metadata for unmapped or unrecognized agents.

    • When disabled, these interactions are discarded.

  • Block Fallback Audio: Controls whether post‑call audio is stored when the real‑time recording is missing.

    • When enabled, the system blocks ingestion of fallback audio and saves only post‑call metadata and tags.

    • When disabled, the system stores fallback audio when required.

Associated Installations

Associated Integration Service Installations: Select the server role that will provide the Recorder Integration Service for this data source. The integration service captures media in real-time, such as agent screen activity.

Associated Interaction Capture Installations: Select the server role that will provide the Interaction Capture services for this data source. The Interaction Capture Service retrieves the media recorded by the external system, which can include post-call audio, video, chat, or email, depending on the capabilities of the external system.

Advanced Settings - Key Value pairs

The following Key and Value pairs are optional settings.

Do not use both the EnableAwsRegionalEndpoints and the AwsStsRegionalEndpoint key at the same time. The AwsStsRegionalEndpoint setting overrides the EnableAwsRegionalEndpoints. So, you must choose one or the other.

Key

Value

EnableAwsRegionalEndpoints

False (default): The option is disabled.

When disabled, the data source uses the AWS global service for AWS Security Token Service (AWS STS), DynamoDB, Kinesis, and S3.

Global requests for STS go to the endpoint https://sts.amazonaws.com.

 

True: The option is enabled.

When enabled, the data source uses standard regional endpoints for AWS Security Token Service (AWS STS), DynamoDB, Kinesis, and S3.

AWS recommends using Regional AWS STS endpoints instead of the global endpoint.

AwsStsRegionalEndpoint

Important: This setting overrides the EnableAwsRegionalEndpoints.

Enter a specific regional endpoint for AWS Security Token Service (AWS STS).

Example:  

https://sts.us-east-1.amazonaws.com

Amazon.KinesisStreamsEnhancedFanOut

False (default): Disables the use of Amazon Kinesis Data Stream Enhanced Fan Out.

True: The Amazon Kinesis Data Streams are configured to use the Enhanced Fan Out feature. Next, enable the required permission, as listed in the row "Kinesis data streams using Enhanced Fan Out" in IAM User Permissions.

For more info, see Amazon Kinesis Data Streams Adds Enhanced Fan-Out and HTTP/2 for Faster Streaming (AWS News Blog).

SwapAudioChannels

By default, when you playback a voice conversation, the agent's voice is on the Blue channel (top) and the customer's voice is on the Green channel (bottom). This is known as speaker separation.

Verint Interaction playback showing red and blue voice channels

If your Amazon Connect contact center is configured to swap these streams, add the SwapAudioChannels key and set it to True. Wait a few minutes for WFO to update all the active jobs on the Recorder. Then check that new calls have the agent and customer on the correct channels.

BlockFallbackRecording

The purpose of BlockFallbackRecording is to manage how call recordings are processed during fallback scenarios. When this setting is set to True:

  • No fallback recovery: Both manual and automatic recovery methods are disabled.

  • No audio ingestion: Post-call audio recordings are not ingested.

  • Metadata-only focus: Only session-related metadata, such as tagging details are captured.

This setup focuses on retaining metadata while making sure no audio recordings are ingested after the call. It is designed for situations where privacy or compliance rules require excluding audio data.