VietNormalizer: An Open-Source, Dependency-Free Python Library for Vietnamese Text Normalization in TTS and NLP Applications
Hung Vu Nguyen, Loan Do, Thanh Ngoc Nguyen, Ushik Shrestha Khwakhali, Thanh Pham, Vinh Do, Charlotte Nguyen, Hien Nguyen
arXiv preprint
Abstract
Presents VietNormalizer, an open-source, zero-dependency Python library for Vietnamese text normalization aimed at Text-to-Speech (TTS) and NLP applications. The rule-based pipeline handles numbers, dates, times, currency, percentages, acronyms and loanword transliteration without neural-network dependencies, and has been used in the VietSuperSpeech data pipeline for ASR preprocessing.
Keywords
APA-style Citation
Hung Vu Nguyen, Loan Do, Thanh Ngoc Nguyen, Ushik Shrestha Khwakhali, Thanh Pham, Vinh Do, Charlotte Nguyen, Hien Nguyen (2026). VietNormalizer: An Open-Source, Dependency-Free Python Library for Vietnamese Text Normalization in TTS and NLP Applications. arXiv preprint.
BibTeX
@misc{Nguyen2026vietnormalizer,
author = {Hung Vu Nguyen and Loan Do and Thanh Ngoc Nguyen and Ushik Shrestha Khwakhali and Thanh Pham and Vinh Do and Charlotte Nguyen and Hien Nguyen},
title = {VietNormalizer: An Open-Source, Dependency-Free Python Library for Vietnamese Text Normalization in TTS and NLP Applications},
year = {2026},
note = {arXiv preprint},
url = {https://arxiv.org/html/2603.04145v1}
}