ElGarash / Al-Ajwad

Al Ajwad is a graduation project submitted to the Department of Computers and Systems Engineering, Minia University as partial fulfilment for a B.Sc. degree. It is an ASR model trained to recognize the Tajweed rules of The Holy Quran recitation.
https://huggingface.co/spaces/abdoeid/Al-Ajwad
Creative Commons Zero v1.0 Universal
2 stars 1 forks source link

Text normalizer #8

Closed AbdulrhmnGhanem closed 10 months ago

AbdulrhmnGhanem commented 10 months ago

A text normalizer should be run on the model output before doing any postprocessing steps.

def nomalize(text: str) -> str:
   ...

example = "الذين كذبوا بالكتاب الذين كذبوا بالكتاب وبما أرسلنا به رسلنا فسوف يعلمونا"
assert normalize(example)  == "الذين كذبوا بالكتاب وبما أرسلنا به رسلنا فسوف يعلمونا"
abdoei commented 10 months ago

You can test this function implementation here