Language and region support for QnA Maker
The language of a knowledge base affects QnA Maker's ability to auto-extract questions and answers from sources, as well as the relevance of the results QnA Maker provides in response to user queries.
QnA Maker supports question/answer extraction in any language page, but the effectiveness of the extraction is much higher for the following languages, as QnA Maker uses keywords to identify questions.
Query matching and relevance
QnA Maker depends on language analyzers in Azure search for providing results. Special re-ranking features are available for En-* languages that enable better relevance.
While the Azure Search capabilities are on par for supported languages, QnA Maker has an additional ranker that sits above the Azure search results. In this ranker model, we use some special semantic and word based features in en-*, that are not yet available for other languages. We do not make these available, as they are part of the internal working of the ranker.
QnA Maker auto-detects the language of the knowledge base during creation and sets the analyzer accordingly. You can create knowledge bases in the following languages. Read this for more details about how QnA Maker handles languages.
Language analyzers, once set, cannot be changed. Also, the language analyzer applies to all the knowledge bases in a QnA Maker service. If you plan to have knowledge bases in different language, you should create them under separate QnA Maker services.