Duplicate-Genesys Cloud Audio Connector Integration
Overview
The Genesys Cloud Audio Connector Integration enables iX Hello VNext to operate as a third-party voice bot within Genesys Cloud using the Genesys Audio Connector (AudioHook Protocol v2).
This integration replaces the previously planned SIP/BYOC trunking architecture and eliminates the need for SIP endpoints, RTP media handling, and transfer node configurations. Instead, VNext operates as a secure WebSocket (WSS) server, allowing Genesys Cloud to stream audio directly to VNext for real-time conversational processing.
Key Benefits
Simplified Infrastructure
Removes the need for SIP trunks, BYOC configuration, and RTP media management.
Reduces deployment and maintenance complexity.
Enhanced Reliability
Uses secure bidirectional WebSocket communication over port 443.
Simplifies firewall and network configuration.
Seamless Agent Handoff
Genesys Architect continues to control the overall call flow.
VNext manages only the bot interaction and returns conversation outcomes to Architect for agent routing.
Native Context Sharing
Customer and call context are exchanged through AudioHook session variables.
Eliminates manual parsing of SIP UUI headers.
How the Integration Works
High-Level Call Flow
A customer call enters a Genesys Architect flow.
Architect invokes the Call Audio Connector action.
Genesys establishes a secure WebSocket connection with VNext.
Audio is streamed bi-directionally between Genesys and VNext.
VNext processes the conversation using:
Speech-to-Text (STT)
Natural Language Understanding (NLU)
Text-to-Speech (TTS)
VNext captures customer intent and relevant business information.
When the conversation ends, VNext returns the configured output variables to Genesys.
Architect resumes execution and routes the call to the appropriate agent queue or workflow.
Administrator Guide
Configure a Genesys Audio Connector Connection
Step 1: Create and Publish the Voice Bot
Log in to the iX Hello V2 platform.
Create a new Voice Bot or open an existing bot.
Configure:
Speech-to-Text (STT)
Dialog Flow
Text-to-Speech (TTS)
Test the conversation flow.
Publish the bot.
Note the WebSocket endpoint details provided for the deployment.
Step 2: Configure AudioHook in Genesys
Sign in to Genesys Cloud as an administrator.

Navigate to the AudioHook or integration configuration section.
Create a new AudioHook integration.
Enter the iX Hello V2 WebSocket endpoint.
Configure:
Connection settings
Audio streaming parameters
Security and authentication details (if applicable)
Save the configuration.
Step 3: Create a Call Flow in Genesys Architect
Open Genesys Architect.
Create a new inbound call flow or edit an existing flow.
Add the required call routing logic.
Configure the call flow to invoke the AudioHook integration.
Associate the AudioHook connection with the desired routing path.
Save and publish the flow.
Step 4: Configure Bot-to-Agent Transfer
Design escalation logic in the iX Hello V2 bot.
Define scenarios requiring agent transfer, such as:
Complex requests
Customer escalation requests
Authentication failures
Unsupported intents
Configure the Transfer Node within the bot flow.
Ensure transfer metadata is generated before escalation.
Step 5: Configure Conversation Summary
Add a Flow Summary node in the bot flow.
Define a summary prompt for capturing conversation details.
Configure the session variable to store the generated summary.
Validate that the summary is generated before call transfer.
Example Summary Prompt
Plain Text1Generate a concise summary of the customer interaction including:2- Reason for contact3- Account details gathered4- Actions performed5- Outstanding issues6- Suggested next stepsShow more lines
Step 6: Configure UUI Data
User-to-User Information (UUI) is used to transfer interaction context back to Genesys.
Configure the bot to generate and pass information such as:
Customer ID
Account Number
Case Number
Intent
Interaction Summary
Session Variables
Example UUI Payload
Plain Text1CustomerID=1001232AccountNumber=ABC4567893Intent=BillingSupport4Summary=Customer reported duplicate charges and requires specialist assistance.Show more lines
Step 7: Configure Agent Desktop Display
Ensure the Genesys agent interface is configured to display transferred data such as:
Customer identification details
Account information
Conversation summary
Bot-collected context
This enables agents to continue the conversation without asking customers to repeat information.
Step 8: Publish and Test
Publish the Genesys call flow.
Place a test call.
Verify:
Audio is streamed successfully.
Bot responses are received.
Summary generation is working.
Transfer requests are successful.
UUI information reaches the agent.
Validate end-to-end customer experience.
Conversation Context Transfer
What Information Can Be Passed?
Account Number
Identify customer account
Customer ID
Link customer records
Intent
Explain reason for contact
Summary
Provide interaction overview
Session Variables
Share collected information
Case Reference
Associate existing cases
Flow Designer Guide
Flow designers can define how information is exchanged between Genesys and VNext during a conversation.
Input Variable Mapping
Input mapping allows Genesys call context to be injected into VNext flow variables.
Supported Standard Fields
ANI
Caller phone number
DNIS
Dialed number
conversationId
Unique Genesys conversation identifier
Custom Variables
Architect variables can also be mapped to flow variables, such as:
Customer Account Number
Product Information
Case ID
Customer Segment
Language Preference
Example
accountNumber
{{account_id}}
customerType
{{customer_type}}
preferredLanguage
{{language}}
Output Variable Mapping
Output mapping determines which VNext variables should be returned to Genesys when the conversation ends.
Common Output Variables
{{intent}}
Customer's identified intent
{{account_id}}
Account reference
{{summary}}
AI-generated conversation summary
{{resolution_status}}
Conversation outcome
Example
intent
Queue routing decision
summary
Agent screen-pop
account_id
CRM lookup
resolution_status
Workflow disposition
Bot Conversation Lifecycle
Architect initiates the Audio Connector session.
VNext conducts the customer conversation.
Flow logic collects customer information and determines intent.
VNext generates a conversation summary.
The Disconnect Event is triggered when the interaction ends.
Mapped output variables are returned to Genesys.
Architect resumes the flow and executes the next action.
Genesys Architect Guide
After the Call Audio Connector action completes, Architect automatically resumes execution.
Typical Post-Bot Processing
Step 1: Read Returned Variables
Architect receives values such as:
intent
summary
account_id
resolution_status
Step 2: Set Participant Attributes
Use the returned values to populate participant attributes and downstream systems.
Step 3: Route the Call
Based on the identified intent or disposition:
Transfer to the appropriate queue
Route to a specialized agent
Continue self-service workflows
End the interaction if fully resolved
Security and Authentication
The integration supports multiple layers of security.
API Key Authentication
Each incoming connection can be validated using an API key configured in VNext.
HMAC Signature Validation
Requests can also be authenticated using HMAC signatures generated with a shared client secret.
Organization-Level Access Control
Administrators can explicitly allow traffic only from approved Genesys Organization IDs.
Security Benefits
Prevents unauthorized access
Protects customer conversations
Ensures secure communication between platforms
Supports enterprise compliance requirements
Data Privacy and Compliance
PII Redaction
Before conversation data is returned to Genesys:
VNext generates an AI-powered conversation summary.
The summary is processed through the VNext PII Redaction Service.
Sensitive customer information is removed or masked.
The sanitized summary is sent back to Genesys.
This helps protect customer privacy while still providing agents with useful conversation context.
Technical Considerations
Audio Streaming
The integration supports:
Real-time bidirectional audio streaming
PCMU 8kHz audio format
AudioHook Protocol Version 2
Supported Processing Flow
Customer Speech
The customer speaks during the call.
Genesys streams the audio to iX Hello VNext through the Audio Connector.
Speech-to-Text (STT)
Incoming audio is converted into text.
The transcribed text becomes the input for conversational processing.
Natural Language Understanding (NLU)
VNext analyzes the transcribed text.
Customer intent, entities, and contextual information are identified.
Flow Execution
The configured VNext flow executes the appropriate business logic.
Actions may include validating customer information, retrieving data, invoking integrations, or determining the next conversational step.
Text-to-Speech (TTS)
The bot's response is converted from text into speech.
Natural-sounding audio is generated for the customer.
Response to Customer
The generated audio is streamed back to Genesys.
The customer hears the response in real time, creating a seamless voice interaction.
Session Data Exchange
Unlike SIP-based integrations, session information is exchanged through AudioHook JSON payloads.
Benefits
No SIP UUI dependency
No 256-character UUI limitation
Structured JSON data exchange
Easier context management
Note: Although larger data payloads are supported, concise summaries are recommended to improve agent usability and readability.
Summary Model Selection
Flow designers can select the LLM used to generate AI conversation summaries.
Examples include:
GPT-4o-mini
Other supported models available within VNext
This allows organizations to optimize:
Cost
Performance
Summary quality
Troubleshooting
Connection Fails
Verify:
WSS endpoint URL is correct
API key is configured properly
HMAC client secret matches
Genesys Organization ID is authorized
Variables Not Appearing in VNext
Verify:
Input mappings are configured correctly
Variable names match the Architect configuration
Required session variables are being sent from Architect
Variables Not Returned to Genesys
Verify:
Output mappings are configured
Variables contain valid values at disconnect
Architect is expecting the returned attributes
Audio Quality Issues
Verify:
Stable network connectivity
Correct PCMU 8kHz audio format
WebSocket connectivity over port 443
Service health of STT and TTS providers
Conclusion
The Genesys Cloud Audio Connector integration provides a modern, scalable, and secure approach for connecting iX Hello VNext voice bots with Genesys Cloud. By leveraging the AudioHook protocol and WebSocket-based audio streaming, organizations can simplify infrastructure, improve reliability, securely exchange customer context, and deliver seamless transitions between automated and human-assisted experiences. This architecture enables VNext to focus on intelligent conversational engagement while Genesys continues to orchestrate call routing and agent experiences.
Last updated
Was this helpful?