whisper_normalizer_basic_and_english_text_cleaning_quickstart.py

1from whisper_normalizer.basic import BasicTextNormalizer
2from whisper_normalizer.english import EnglishTextNormalizer
3
4# Initialize the normalizers
5basic_normalizer = BasicTextNormalizer()
6english_normalizer = EnglishTextNormalizer()
7
8# Example raw text
9text = "I'm a teacher at a high school, and I've been working there for 10 years. It's $5.00."
10
11# Perform basic normalization (removes punctuation, lowercase, etc.)
12basic_cleaned = basic_normalizer(text)
13print(f"Basic Normalized: {basic_cleaned}")
14
15# Perform English-specific normalization (standardizes spellings, numbers, and contractions)
16english_cleaned = english_normalizer(text)
17print(f"English Normalized: {english_cleaned}")