概述

The languages that UiPath® Document Understanding^TM models can process are primarily determined by the capabilities of the OCR engine. If the OCR engine is able to recognize a specific language, then the UiPath Document Understanding models can be trained to process documents in that language. However, the out-of-the-box pre-trained models only support a limited number of languages. If you need to process languages that are not supported out of the box, you will have to further train these models. Additionally, consider the following guidance:

• 要在中文文档上训练模型，请使用 DocumentUnderstanding 包或中国发票包。
• To train a model on Hebrew documents, use either the DocumentUnderstanding package or the InvoicesHebrew package.
• 要在日语文档上训练模型，请使用 DocumentUnderstanding 包或日本发票包。
• To train a model on Latin script documents, use any package except for InvoicesChina, InvoicesHebrew, or InvoicesJapan.