Use Admin Panel Configurations to manage application-level settings that control how your Conversational AI application behaves and responds. These settings help you configure timeout handling, fallback behavior, voice bot capabilities, feature access, and advanced AI settings.

Timeout Settings

Configure how the application responds when a user does not provide input within a specified time period. You can set reminder messages and define actions to keep the conversation active.

For more information, see Global Timeout Settings.

Fallback Settings

Configure the default response when a user's message does not match any configured intent and no local fallback is available. The application can prompt users to rephrase their query or provide a predefined response.

Voice Bot Settings

For Voice Bot applications, you can configure the following settings:

• Global Repeat Intent: Select an intent that allows users to request the bot to repeat its last response.

• Speech Profile: Choose the speech profile to be used for voice interactions.

• Accumulate User Replies: Specify how long the system should collect multiple user utterances before sending them for intent detection.

  

Barge-in Settings

Allow users to interrupt an ongoing voice response and speak before the bot finishes.

You can configure:

• Minimum completion percentage of the bot response before interruptions are allowed.
• Minimum bot response duration before user input is accepted.
• Time (in milliseconds) after which the bot stops speaking when a barge-in is detected.

The system applies the highest configured value across these settings.

Global Hold Settings (Voice Bot)

Global Hold Settings allow the voice bot to manage situations where users want to place a conversation on hold.

Configure the following:

• Hold Intent – Intent that triggers the hold action.
• Wait Duration – Time the system remains on hold while waiting for a response.
• Hold Message – Message played when the hold intent is detected.

Trigger Application Without Rule Setup

Select the Live Chat application where you want to activate the bot without creating routing rules.

Feature Suite

The Feature Suite section provides access to additional capabilities through configurable toggles.

• Typing Indicator: Displays a typing indicator while the bot prepares a response. You can configure the indicator text directly from this section.

• Stream Response: Streams LLM-generated responses in real time within Live Chat. This feature is supported for LLM-powered capabilities such as Dynamic Workflows and Smart FAQ.

• Bot Response Delay: Adds a delay of 1–5 seconds before bot responses to create a more natural conversational experience.

• Context Switching: Allows the application to switch automatically between dialog flows when a new issue type is detected. For more information, see Simplified Context Switching.

• Prioritize Context Switching Over Local Configuration: When enabled, context switching takes precedence over local fallback configurations. The recommended setting is Off.

• Advanced Nodes:Provides access to advanced node types for more complex use cases.

Note: For Smart FAQ, when Stream Response is enabled:

• Responses are streamed in real time as they are generated.
• Output guardrails and post-processing are bypassed for generated responses.
• Configuration of the Bot Response variable is not required because the node publishes the response directly.

Additional Settings

Configure advanced AI and application-level settings from this section.

• Guardrail: Select the guardrails that should be applied to both user inputs and AI-generated responses.

• PII Masking Template: Choose a predefined masking template to automatically redact sensitive personal information from conversations.

• LLM Provider and Model: Select the LLM provider and model used to generate AI responses.

• Reset Context Keywords: Define keywords that reset the conversation context and start a new interaction. For example, entering Hi can initiate a new conversation flow.

• Advanced Modules: Enable access to advanced modules in the Manage section. Recommended only for specific use cases.

• Deployment Settings: Provides access to deployment and priority management settings. Recommended for advanced configurations.

• Golden Test Sets for ML Models: Enables the Golden Test Set module for validating and testing intent models.

• Save as Draft for Dialogue Trees: Allows users to save canvas changes as drafts without affecting the currently deployed version.

​