Supernova Nepali Normalizer

Python text normalization library from Supernova11c.

Install

pip install supernova-norm

Quick start

from supernova_norm import normalize

text = "  नेपाली   भाषा  "
print(normalize(text))

Class-based usage

from supernova_norm import SupernovaNepaliNormalizer

normalizer = SupernovaNepaliNormalizer()
print(normalizer.normalize("“नेपाल”—सुन्दर देश"))

Normalization rules

• Unicode NFC normalization
• Removal of zero-width non-joiner and joiner characters
• Standardization of selected dash and quotation marks
• CRLF and carriage-return conversion to line breaks
• Tabs converted to spaces
• Repeated spaces and excessive blank lines reduced
• Leading and trailing whitespace removed

None returns an empty string. Other inputs are converted to strings.

Project

Source: https://huggingface.co/Supernova11c/Supernova-Nepali-Normalizer-V2