The Document Intelligence application provides support for documents in different languages.

The language model configured for a use case allows Document Intelligence to detect data in documents, make predictions for field values, and further train the model in the selected languages.

Select the language model when creating a document extraction use case. For more information, see Create a document extraction use case.

The following table lists the languages supported for Document Intelligence along with the models used to support them.

Table 1. Document Intelligence languages and models
Model Languages
Default English, French, German, Spanish, Portuguese, Dutch, Italian, Czech, Danish, Finnish, Norwegian, and Swedish.
OCR for Japanese and Chinese Chinese (simplified) and Japanese