Most words in Romanian are missing characters

[vladandrew]

🔎 Search before asking

• I have searched the PaddleOCR Docs and found no similar bug report.
• I have searched the PaddleOCR Issues and found no similar bug report.
• I have searched the PaddleOCR Discussions and found no similar bug report.

🐛 Bug (问题描述)

The Romanian language is not fully supported, as diacritics (ăâîșț) are not properly handled. This makes the resulting text unusable since most words are missing characters.

🏃‍♂️ Environment (运行环境)

OS: Ubuntu 24.04.1 LTS
Environment: conda
Python: Python 3.11.5
Install: pip
python -m pip install paddlepaddle==3.0.0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
pip install paddleocr
RAM: 128GB
CPU: AMD Ryzen Threadripper PRO 5965WX 24-Cores
CUDA: None

🌰 Minimal Reproducible Example (最小可复现问题的Demo)

For the image below, run PaddleOCR using:

paddleocr ocr -i monitorul_oficial_sample.png --lang ro --use_doc_orientation_classify False --use_doc_unwarping False --use_textline_orientation False --save_path ./output

The resulting text has many words that are incorrectly extracted due to missing characters (judector vs judecător). A snippet of the resulting JSON:

        "— judector",
        "neconstituionalitate a dispoziiilor art. 327 lit. b) i ale",
        "Gheorghe Stan",
        "— judector",
        "art. 328 alin. (3) din Codul de procedur penal, excepie",
        "Livia Doina Stanciu",
        "— judector",
        "ridicat de Gheorghe Cureleac într-o cauz penalà în care",
        "Elena-Simina Tnsescu",
        "— judector",
        "autorul excepiei a fost trimis în judecatà pentru svârirea unor",
        "Varga Attila",
        "— judector",
        "infractiuni.",
        "5. În motivarea excepiei de neconstituionalitate, autorul",