NOTICE

Indic Text Normalization
Copyright 2025 Kenpath Technologies Pvt Ltd

This product is a derivative work based on NVIDIA NeMo Text Processing.

================================================================================

This product contains code from NVIDIA NeMo Text Processing
Copyright 2020-2024 NVIDIA Corporation

Original repository: https://github.com/NVIDIA/NeMo-text-processing
Licensed under the Apache License, Version 2.0

The base text normalization framework, including the English language module
(en/), graph utilities, and core WFST architecture are derived from NVIDIA
NeMo Text Processing.

================================================================================

Portions of this software are derived from Google's text normalization grammars.
Copyright 2015 and onwards Google, Inc.

Specifically, the base graph utilities in en/graph_utils.py contain code
originally developed by Google and subsequently incorporated into NVIDIA NeMo
Text Processing.

================================================================================

Kenpath Technologies Pvt Ltd has contributed:
- Comprehensive Indic language support for 19 Indian languages:
  Hindi (hi), Bengali (bn), Marathi (mr), Telugu (te), Kannada (kn),
  Bhojpuri (bho), Magahi (mag), Chhattisgarhi (hne), Maithili (mai),
  Assamese (as), Bodo (brx), Dogri (doi), Gujarati (gu), Malayalam (ml),
  Punjabi (pa), Tamil (ta), Nepali (ne), and Sanskrit (sa)
- Language-specific taggers and verbalizers for each semiotic class
- Native script digit handling and verbalization
- Cultural and linguistic adaptations for Indic languages
- Test cases and validation data for Indic languages

================================================================================

This software is licensed under the Apache License, Version 2.0.
See the LICENSE file for the full license text.