Search Manual / Version 2512.0
Table Of ContentsTo enable the Semantic Search feature, ensure that your CoreMedia license includes the required feature flag. Please contact CoreMedia support for licensing details.
The Embedding Service is configured via Spring Boot properties in both the Content Feeder and the Studio Server. These properties specify which embedding model to use and how to connect to the service.
Multiple embedding service implementations are supported:
Amazon Bedrock Nova (default): Offers multimodal embeddings, supporting both text and images.
Amazon Bedrock Titan: Provides text-only embeddings, fully integrated with Spring AI.
Shared AWS Connection Properties: The embedding service is integrated via Spring AI. The following AWS connection properties are required for both models:
spring.ai.bedrock.aws.regionspring.ai.bedrock.aws.access-keyspring.ai.bedrock.aws.secret-key
Amazon Bedrock Nova: To enable Amazon Bedrock Nova use the following properties:
ai.model.embedding=bedrock-novaAll properties starting with
ai.bedrock-nova.*
For a complete list of Nova configuration properties, see the Content Feeder in Deployment Manual and Studio Server in Deployment Manual property references in the Deployment Manual.
Amazon Bedrock Titan: This model is fully integrated via Spring AI. Use the following properties:
spring.ai.model.embedding=bedrock-titanAll properties starting with
spring.ai.bedrock.titan.*
For details, refer to the Spring AI Bedrock Titan documentation.
For further configuration details, see the following sections:
Content Feeder: Section 4.2.3.8, “Configuring Semantic Search for Content Feeder”
Studio Server: Section 4.4.4, “Configuring Semantic Search for Studio Server”
