aho/welcomecenterbot
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

normlizer-fix5

Browse Source
This commit is contained in:
Untone 2024-09-27 11:18:18 +03:00
parent e241e14764
commit 66d0bba13c
1 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
nlp/normalize.py
View File

@ -42,12 +42,13 @@ def normalize(text):
Normalize English text to resemble Russian characters. Normalize English text to resemble Russian characters.
""" """
# Segment the text first # Segment the text first
t = segment_text(text.replace(" ", " ")) t = segment_text(text.replace(" ", " ").replace(" ", " ").replace(" ", " "))
t = t.lower()
if is_russian_wording(t): if is_russian_wording(t):
# Normalize the text by replacing characters # Normalize the text by replacing characters
normalized_text = ( normalized_text = (t
t.lower()
.replace("e", "е") .replace("e", "е")
.replace("o", "о") .replace("o", "о")
.replace("x", "х") .replace("x", "х")
@ -82,4 +83,3 @@ if __name__ == "__main__":
normalized_output = normalize(input_text) normalized_output = normalize(input_text)
print(normalized_output) print(normalized_output)