This article explains how to export Genesys Cloud recordings and associated metadata to an Amazon Simple Storage Service (AWS S3) where it can be retrieved by Call Journey and processed through its VoiceAI service. The VoiceAI service provides advanced speech and text analytics and delivers insights into Agent Performance, Customer Experience (emotion, sentiment, negative experience, etc), Call Drivers, Trends, and Risk & Compliance.
Prerequisites
Genesys Cloud license
An AWS S3 storage account.
S3 provider will need to know the Genesys Cloud Organization ID when creating the S3 bucket.
Admin access to Genesys Cloud
You will need to know the following credentials when configuring the Genesys Cloud integration:
S3 Bucket name
Role ARN (the customer AWS role that Genesys Cloud is authorized to access).
Note |
---|
If Call Journey is providing the S3 bucket, you will need to provide your Genesys Organization ID and Genesys Region to Call Journey. Call Journey will then create the S3 bucket and provide you with the credentials to insert into the Genesys Cloud configuration. |
Note |
---|
Depending on your Genesys Cloud settings, calls that involve multiple agents (ie transfers) may either be saved as one recording (where both agents are on the recording) or as two separate recordings, each with one agent, the same conversationid but different recordingids. |
Summary of Key Steps (details are below)
(assumes Call Journey is providing the S3 bucket)
Task | Who | |
---|---|---|
1 | Provide the following information to Call Journey
| Client |
2 |
| Call Journey |
3 |
| Client |
4 | Provide the following information to Call Journey
| Client |
5 | Create Wordbench account and folders Configure platform to retrieve and process calls from S3 bucket. | Call Journey |
Other Important Information
If not done previously, Call Journey will now automatically redact the IVR and any Audio Events (hold times).
Calls with no start time and date, no queue ID or no user ID (agent UID) in the metadata file uploaded to S3 will not be processed.
NB: there will be cases where the user ID is provided in the metadata file but the Genesys api will return no agent name. These calls will still be processed.
Call routing:
If call routing is based on queue names, then the routing will be based on the first queue listed in the metadata file.
If the first queue listed is not a provided parameter, then the second queue will be used (if one is provided).
Depending on your Genesys Cloud settings:
Conversations with multiple segments/recordings (typically occurring where a call is transferred to another agent) may now be shown separately in Wordbench
Each segment will have its own unique recording ID, but have the same conversation ID.
All segments of the conversation can be located in Wordbench by searching on the conversation ID.
Calls where the agent puts the customer on hold while making inquiries with other agents or departments but then returns to the customer may not result in multiple segments/recordings but may show multiple agents on the call. This is a feature of the Genesys Cloud platform.
Setting Up the AWS S3 Integration
The configuration involves four key steps to be completed within Genesys Cloud:
Create an Integration (this instructs Genesys Cloud WHERE to send your recordings ie to AWS S3).
Create a Policy Rule (this instructs Genesys Cloud WHAT calls to send to AWS S3)
Give Call Journey Access to the Genesys api
Provide Call Journey with Segmenting Details (for multi-folder setup)
Install the S3 Integration
Log into Genesys Cloud and select Admin and then Integrations.
Add an Integration.
Search for the AWS S3 listing and select Install.
Give a name to your Integration and then select the Configuration tab.
Enter the S3 Bucket name and then select the Credentials tab:
Select Configure and enter the Role ARN. Select OK and Save the settings.
Back on the Integrations page, you should see your newly created Integration.
Under Status, select Active/Inactive to activate the integration and select Yes to confirm.
That completes the Integration set up.
Create a Policy Rule
Select Admin and then Policies.
You will find Policies under the Quality heading or use the search box to search for Policies.
Create a new policy.
You now have to configure a Policy Rule to instruct Genesys Cloud WHAT calls to send to AWS S3 storage.
Enter a name for the Policy Rule
Decide the criteria that each call needs to meet in order to be sent for exporting to AWS S3.
Ensure you select Export Recording With Integration and enter the name of the Integration you created above.
The settings below show Inbound Calls from the DevTest queue are being sent to the “AWS S3 Bulk Export Integration” integration.
Select Save to finish.
Any call meeting the criteria set out in the Policy Rule will be exported to AWS S3.
Note |
---|
If you created the S3 bucket, make sure to send your AWS S3 credentials to Call Journey so that they can connect and retrieve the recordings and metadata. |
Give Call Journey Access to the Genesys API
Two steps to providing Call Journey with access to the Genesys api:
Create a Role with the Correct Permissions
Create an OAuth Client
Create a Role with the Correct Permissions
To change or add a role and permissions in Genesys Cloud, please follow the steps below.
Login to your respective Genesys Cloud portal
Click Admin and then Under People and Permissions, click Roles/Permissions
.
Click Add Role
Enter a Name and optional Description in the fields.
5. Click Permissions tab
6. Assign the following Permission sets by clicking the tick boxes as shown in the images below.
Recording > Recording > View
Routing > Queue > View
Routing > Wrap-Up Code > View
Analytics > conversationDetail > View
7. Assign the correct Division to role by clicking Change Membership button
8. In the pop-up window, click the toggle button “
Set divisions for (n) assigned person“
(assign role to yourself)
9. From the “Add Divisions” textbox field, search the appropriate division to be assigned to role and click the save button
Create an OAuth Client
Login to your respective Genesys Cloud portal
Click Admin and then Under Integrations, click OAuth
.
Click Add client: the Client Details tab appears.
Set App Name to a descriptive name of the app. This name is shown when someone authorizes this OAuth client.
(Optional) Type a brief description of the app in the Description box.
4. Set the Token duration time - the time until tokens created with this client expire. Accept the default duration, or enter a value between 300 and 172800 seconds. This sets the lifetime of the token to a maximum of 2 days or less.
5. Choose “Client Credentials” option under Grant Types. Grant Types set the way an application gets an access token.
6. Click the Roles tab and assign the specific role created for Genesys S3 integration to enable access to OAuth client.
7. Click Save. Genesys Cloud creates a Client ID and a Client Secret (token).
8. Provide Call Journey with the Client ID and Client Secret
Provide Call Journey with Segmenting Details
The default integration will process all calls to a single folder in Wordbench.
If you wish to segment your calls into multiple folders on Wordbench, you will need to provide Call Journey with the relevant criteria to route the calls based on the metadata provided in the S3 bucket.
The most common metadata fields used for segmenting calls are also the most common fields used when creating the Policy Rule:
Queue Name/GUID
Call direction
Call duration
Info |
---|
The metadata file exported to AWS S3 provides queue names, agent names and wrap up codes in the form of GUIDs. For example:
|
If using any field with a GUID (eg queues), the best approach is to export the queue GUIDs and their plain English names into a csv or Excel file and insert the relevant Wordbench folder name against each entry.
Provide Call Journey with this “mapping” file along with any other relevant criteria.
As an example, you may have two routing rules:
For the Wordbench SupportInbound folder use:
Queues:
6b4b33d3-90a0-4241-bb93-5610c01264bc
9t6c2213-78a0-3278-xx45-7878b98762fg
Call direction: Inbound
Duration greater than 60 seconds
For the Wordbench SalesOutbound folder use:
Queues:
7h8v22w2-67a0-1278-gg94-8456c01264bc
8456c01g-5690-3028-jk45-9328b99332yt
Call direction: Outbound
Duration greater than 300 seconds
Note |
---|
If calls have multiple values within a field (eg multiple queue GUIDs), the first listed field value will be used. |
Info |
---|
Call Journey will use the Genesys api to retrieve the plain English names associated with all GUIDs and use these within Wordbench. |