Conversation Sources define the systems from which Quality AI ingests and processes conversations for AutoQA and Conversation Intelligence.
Quality AI supports both Human Agent and AI Agent conversation sources. The Automation AI conversation source enables Quality AI to process conversations from Automation AI Experience Flows. When enabled, AutoQA evaluates AI Agent conversations using evaluation forms assigned to Experience Flows, while Conversation Intelligence analyzes conversations for sentiment, topics, and intents. Evaluate conversations handled by AI-assisted human agents, including conversations recorded through Agent AI on supported third-party desktops.
Agent AI also supports ingestion of third-party desktop conversations recorded through Agent AI when the required conversation metadata is provided.
The Conversation Sources page helps you:
- Enable or disable conversation sources.
- Configure and manage voice and chat queues.
- View queue details, including queue name, queue ID, system-generated ID, and assigned agents.
- Add or edit queues.
- Assign agents to Agent AI and Quality AI Express queues through the UI.
This centralized configuration simplifies conversation ingestion, routing, and management while ensuring the metadata required for accurate analytics and reporting is available.
Access Conversation Sources
Navigate to Quality AI > Configure > Conversation Sources.
Supported Conversation Sources
Quality AI supports three conversation sources, each designed for specific deployment scenarios and organizational needs.
Automation AI
Automation AI processes conversations handled by Automation AI Experience Flows using AutoQA and Conversation Intelligence.
| Attribute | Description |
|---|
| Integration Method | Native Automation AI integration |
| Data Processing | Automatically processes AI Agent conversations |
| Configuration Requirements | Enable the Automation AI conversation source |
| Use Case | Evaluate and analyze AI Agent conversations handled by Automation AI Experience Flows |
Prerequisites
Enable Contact Center AI (CCAI) or Agent AI from Quality AI > Configure > Conversation Sources before enabling Automation AI.
When Enabled
- Enable AutoQA and Conversation Intelligence for the app.
- Create and enable a default evaluation form with default evaluation metrics.
- Assign the default evaluation form to all Experience Flows in the current workspace.
- Make Experience Flows available for evaluation form assignment.
- Process AI Agent conversations using AutoQA and Conversation Intelligence.
Contact Center AI (CCAI) ingests real-time conversation data from the CCAI platform for evaluation and analytics.
| Property | Details |
|---|
| Integration method | Direct API connection |
| Data processing | Real-time conversation ingestion |
| Configuration | None (automatic integration) |
| Use case | Evaluate conversations from Kore.ai Contact Center AI deployments |
When Disabled
Stop processing incoming interactions from CCAI on third-party desktops.
Limitation
- Configure queues and assign agents through existing CCAI workflows and APIs.
- Queue management through the Conversation Sources UI isn’t supported.
Agent AI
Agent AI processes human-agent conversations enhanced with AI assistance, including conversations recorded through Agent AI on third-party desktops.
| Property | Details |
|---|
| Integration Method | Structured data feeds |
| Data Processing | Real-time processing with metadata validation |
| Configuration | Add queues with Agent Queue names and Queue IDs |
| Use Case | Evaluate conversations handled by AI-assisted human agents |
When Disabled
Stop processing incoming interactions from Agent AI on third-party desktops.
Webhook API Requirements for Conversation Events
For third-party desktop conversations recorded through Agent AI, the Agent AI Webhook API must include the following mandatory fields (participant custom field objects):
| Field | Description |
|---|
agentEmail | Platform or Quality AI onboarded agent email address |
queueId | Quality AI onboarded queue ID |
queueName | Quality AI onboarded queue name |
These above fields are mandatory for Quality AI evaluation of third-party desktop conversations.
Quality AI Express
Quality AI Express imports conversations from external sources through AWS S3 Connectors for evaluation and analysis. It supports service records, chat logs, emails, audio recordings, and related metadata files.
| Property | Details |
|---|
| Integration Method | Amazon Web Services (AWS) S3 Connectors |
| Data Processing | Batch file processing |
| Configuration | Add queues with agent queue names and queue IDs |
| Supported data | Customer service records, chat logs, email interactions, audio recordings, and structured metadata files |
| Use case | Evaluate conversations imported from third-party Contact Center as a Service (CCaaS) platforms |
Integration
Use AWS S3 Connectors as the primary mechanism for importing conversations into Quality AI.
Configuration Requirements
Before You Begin
Before configuring Conversation Sources:
- Assign the appropriate user permissions.
- Verify that the required metadata structures exist.
- Confirm that the integration endpoints are available.
Before Activating Quality AI Express
Complete the following steps before activating Quality AI Express:
- Remove all active Playbook metrics from evaluated form queues.
- Remove all active Dialog Task metrics from evaluated form queues.
If you skip these steps, the system blocks activation and displays a warning.
For AWS S3 configuration, see AWS S3 Connectors.
Conversation Sources Interface
The Conversation Sources page displays all enabled conversation sources and their associated queues.
When Enabled
- Display Connectors in the Quality AI navigation panel after enabling Quality AI Express.
- Refresh or re-sync queues to view agent assignments made through public APIs.
Interface Elements
| Element | Description |
|---|
| Enable/Disable Toggles | Start or stop conversation ingestion for each source. |
| Queue List | Displays the configured queues. |
| Name | Displays the queue name. |
| Queue ID | Displays the user-defined queue identifier. |
| System Generated Queue ID | Displays the system-generated queue identifier. |
| Agents | Displays the number of assigned agents. |
| Actions | Edit or delete queues. |
| Add Queue | Create a new queue and assign agents. |
Queue Management
Configure queues for Agent AI and Quality AI Express to route conversations correctly.
Add a Queue
-
Navigate to Quality AI > Configure > Conversation Sources.
-
Enable Agent AI, Quality AI Express source, or both.
-
Select Add Queue.
-
Enter the following details:
| Field | Description |
|---|
| Name | Enter a descriptive name for the queue |
| Queue ID | Enter a unique identifier for the queue |
| System Generated Queue ID | Displays the system-generated queue ID for internal tracking |
| Agents | Assign agents using the searchable list |
-
Select Save to start or update conversation ingestion for the queue.
Conversation Agent–Queue Mapping
This describes the agent-to-queue mapping for both Quality AI Express queues and Agent AI queues. Using the UI, agents can map directly to Express queues, and the system successfully saves the configuration.
To enable interaction evaluation and Kore score validation, you can attach each mapped queue to the relevant evaluation form. This setup validates scheduled interactions ingested via Amazon S3, Public APIs, and Agent AI queues.
The system displays interactions only when all the following conditions apply:
- The queue maps to an evaluation form.
- The agent maps to the queue using the UI.
- The supervisor or agent viewing the data belongs to the same queue.
When any of these conditions fail, the system doesn’t display the interactions.
Applies To
- Agent AI
- Quality AI Express
Limitations
- Agent-to-queue mapping through the UI applies only to Agent AI and Quality AI Express.
- Contact Center AI (CCAI) queue configuration, routing, and agent assignment continue to follow existing CCAI workflows.
- Assign agents created through platform-level user administration or public APIs.
For third-party desktop conversations recorded through Agent AI, the webhook API must provide the configured queueId, queueName, and agentEmail. Queue mapping and evaluation depend on these values.
Chat Script Timestamp Format controls how Quality AI Express parses timestamps in chat conversation files during ingestion. Each chat script contains multiple timestamps per message. Correct configuration ensures accurate message sequencing, duration calculations, and time-based analytics.
Enables you to configure the timestamp format at the Quality AI Express app level, and it applies to all ingested chat files. The system persists the selected format across sessions and displays it consistently in the Conversation Sources tab.
- Orders messages chronologically
- Calculates conversation duration and response times
- Identifies hold periods and transfer events
- Generates time-based analytics and reports
Incorrect timestamp parsing causes sequencing errors and inaccurate metrics.
Unix Timestamp
Represents the number of seconds since the Unix epoch.
Quality AI supports Unix timestamps with start and end time validation. This format limits flexibility for third-party desktop integrations.
- Format: Integer or decimal number
- Examples:
1735574400 or 1735574400.523
The ingestion process validates start and end times, which may limit flexibility for demo environments and third-party desktop integrations.
Offset Timestamp
Represents the elapsed time from the start or end of a conversation.
Quality AI supports message-level offset timestamps without validating the conversation start or end time. This format applies only to chat conversation ingestion and does not affect voice conversation ingestion.
- Format: Integer or decimal number
- Examples:
0 (start), 45 (45 seconds in), 120.5 (2 min 0.5 sec in)
This format doesn’t validate start and end dates-timestamps are used only for reporting and filtering. This change applies only to chat ingestion.
Conversation Source Behavior
| Conversation Source | Behavior |
|---|
| Contact Center AI (CCAI) | Stops processing CCAI interactions. |
| Agent AI | Stops processing Agent AI interactions. |
| Quality AI Express | Stops file-based conversation ingestion. |
Access Control
| Role | Permissions |
|---|
| Auto QA access | Enable or disable Quality AI Express, modify conversation source settings, and save configuration changes. |
| Limited access users | View-only interface, can’t save configurations |
| Agent-level access | Enable or disable conversation sources, can’t save settings |
- Compare functionality
- Contact Center Efficiency
- Agent Efficacy (including CSAT)
- Agent Performance Monitor (Occupancy, Playbook Adherence)
- Customer Experience (Average Wait Time, NPS Score)
- Interaction Details (Dispositions)
