wrong text recognition

[mila6452]

If I run the following code in python 3.12:

import easyocr

print(f"EasyOCR verion: {easyocr.__version__}")

reader = easyocr.Reader(['en'],gpu=False,verbose=False)
texts = reader.readtext('_screenshot_image.png')

print(texts)

The output is:
EasyOCR verion: 1.7.2
[([[np.int32(17), np.int32(3)], [np.int32(95), np.int32(3)], [np.int32(95), np.int32(23)], [np.int32(17), np.int32(23)]], '900/9 900', np.float64(0.9994190484841114))]

As you can see the first '9' character is missing.

What did I wrong? Or how can I change the code that the picture will be correctly recognised?
Thx

[mila6452]

I have tried your web demo (with 'en' only) and it gave the correct result.