First impressions > Recurring areas and elements > AI Chat

AI Chat configuration is managed through environment-based settings and integration-level configurations, enabling flexible deployment across development, staging, and production environments.

SAMCloud Integration

The SAMCloud integration enables AI-enhanced License Management capabilities such as document parsing and access to enriched knowledge services.

The following environment variables must be configured to enable SAMCloud integration:

SAMCloud__Url: "<SAMCLOUD_URL>"

SAMCloud__ApiKey: "<SAMCLOUD_API_KEY>"

SAMCloud__MaxAttempts: 5

MaxPollingAttempts: 6

PollingDelaySeconds: 10

Without SAMCloud, License Management remains operational but lacks automated document analysis and AI-supported enrichment services.

Technology Catalog Configuration

The Technology Catalog integration provides structured product and technology metadata used for AI-assisted analysis and License Management insights.

Catalog__Url: "<CATALOG_SERVICE_URL>"

Catalog__ApiKey: "<CATALOG_API_KEY>"

Catalog__MaxAttempts: 3

This configuration ensures that product normalization and enriched metadata are available for AI-driven evaluations and reporting.

AI Chat /LLM Integration

The AI Chat feature requires a configured AI provider endpoint to process natural language requests and generate responses.

AiConfig__Url: "<AI_SERVICE_URL>"

AiConfig__ApiKey: "<AI_API_KEY>"

AiConfig__Model: "<MODEL_NAME>"

Without a configured AI provider, the AI Chat interface remains available but cannot generate responses.

Voice Mode Configuration (Speech Services)

Voice Mode is an optional capability that must be explicitly configured before it becomes available in the AI Chat interface. Voice features depend on external speech services for both output (Text-to-Speech) and input (Speech-to-Text).

Configuration Scope:

Voice configuration is managed via the Voice Control integration settings. The integration defines the endpoints and credentials used by the backend-controlled voice layer.

The configuration consists of two independent service definitions:

•Text-to-Speech (TTS) for reading assistant responses aloud.

•Speech-to-Text (STT) for converting spoken input into text (Dictate Mode).

Both services must be configured for full voice functionality. If only one service is configured, the corresponding voice feature remains unavailable.

Required Parameters:

For each service, the following parameters are required:

Text-to-Speech (TTS):

•URL: Speech service endpoint used for TTS requests.

•API Key: Credential used to authenticate requests.

•Model / Voice: Voice model used for synthesis (e.g., en-US-JennyNeural).

Speech-to-Text (STT):

•URL: Speech service endpoint used for STT requests.

•API Key: Credential used to authenticate requests.

•Model: Recognition model used for transcription (e.g., scribe_v1).

Activation and Availability:

Voice Mode controls appear in the AI Chat interface only after the Voice Control integration has been successfully configured and activated.

•Voice Mode requires a configured TTS service.

•Dictate Mode requires a configured STT service.

Deployment Notes (Docker)

In the provided docker-compose.yml, no Voice Mode specific environment variables are defined. Voice configuration is therefore expected to be managed through the Voice Control integration settings rather than deployment-time environment configuration.

Security Considerations

API keys and credentials must not be committed to source control. It is recommended to use a secure secrets management solution (e.g., Vault) for production environments.

AI Chat Configuration & Deployment

AI Chat Configuration & Deployment