Translate queries into an Azure search query with the Azure Query Translator

The Azure Query Translator user experience tuning stage translates the search query into an Azure search query.

To configure your Azure Query Translator stage, do the following:

1. In the SmartHub administration portal, click General Settings.

2. In the User Experience Tuning section, click ADD QUERY TUNING or ADD RESULTS TUNING.

3. In the User Experience Tuning field, select Azure Query Translator. Various new fields will appear on the right side of the User Experience Tuning window.

4. In the Name field, enter a name for your tuning stage.

5. Complete the Azure Query Translator specific fields. Refer to the following table for more information:

   Parameter	required/Optional	Description
   Search Fields	Required	A comma-delimited list of searchable fields. The query runs only against these fields. Hit-highlighting is automatically enabled for the fields in this list.
   Search Mode	Required	This setting allows the administrator to specify the search mode that will be used when running a search query. The following search mode options are available: ALL: This search mode will return results that match every keyword in the search query. This is the search mode that is used by default. ANY: This search mode will return results that match any keyword in the search query. This search mode is recommended for conversational search scenarios.
   Ratings Sort Field	Required	This is the name of a column in the search index, which is used to sort documents when the rating is enabled. The default value is escbase_lastupdateutc.
   Minimum Index Coverage	Required	This is the name of a column in the search index, which is used to indicate the percentage of the index that must be available to service the query before it can be reported as a success. The default value is 100.
   Scoring Profile	Optional	This is the name of an existing Scoring Profile created for the used index. For more information about Azure Scoring Profile check here. To learn how to configure a Scoring Profile in Azure, see below.
   K Parameter for Vector Search	Required if you are using hybrid or pure vector searches	This is the number of the nearest neighbor results that are retrieved based on their similarity to the query vector. The value of the k parameter must be between 1 and 10000. By default, this field is set to 1000
   Enable Pure Vector Search	Optional	If this field is enabled, you will be able to submit pure vector searches to Azure AI search. If this field is disabled, a hybrid search query (includes both keyword and vector search) is submitted to Azure AI search. This field is enabled by default. If your search query does not have vector embeddings, this setting is ignored. If you are using hybrid search, you should also specify your search mode through a query scripting stage.
   Enable Semantic Search	Optional	Enable or disable semantic search for queries.
   Semantic Configuration Name	Required (only if the semantic search is enabled)	This is the name of the semantic configuration what is going to be used when performing queries. If this is not set, the default semantic configuration will be used. If no default semantic configuration is set in Azure, requests will fail.
   Maximum number of answers		A value between 1 and 5 determining the maximum number of answers that should be returned for a query. The default value is 5.
   Minimum number of stop words		A value greater than 0, determining the minimum number of stop words necessary for a query to be eligible for semantic search. The default value is 1.
   Minimum number of words		A value greater than 0, determining the minimum number of words necessary for a query to be eligible for semantic search. The default value is 3.
   Allowed source IDs	Optional	This is a list of GUIDs delimited by commas or semicolons representing the source IDs allowed to perform semantic search.

6. Click Accept.