German Language Detection

[PhilippJindraBS]

Description of the bug | 错误描述

Hallo,
First of all thank you for MinerU.

BUT the german language detection is miserable.

The German "Umlaute" are not recognized ... like ö, ü, ä, ß.
I have set the language to "german" but nothing really changed.

For exmaple:
This is in the text: Köln Dünnwald
This is the output: K8ln Dünnwald

In addtion:
this is in the text: Köln-Dünnwald
This is the output: Koln-Dunnwald

Maybe you have a solution for this problem.

Thank you!

How to reproduce the bug | 如何复现

You can try it with any document that has the german language.

Operating system | 操作系统

Windows

Python version | Python 版本

3.10

Software version | 软件版本 (magic-pdf --version)

0.10.x

Device mode | 设备模式

cuda

[myhloli]

Clearly, PaddleOCR does not perform well in scenarios other than Chinese and English. We plan to incorporate additional OCR methods in the future to improve recognition quality for non-Chinese and non-English texts. However, due to limited development resources, this process may take some time.