license: cc-by-sa-4.0
language:
- tig
Tigre Data Lexicon (tigre-data-lexicon)
Overview
This repository contains the Tigre Data Lexicon, a specialized linguistic resource designed to support the development of speech and language technologies for Tigre, an under-resourced Semitic language. This lexicon serves as a foundational component for bridging the gap between written text and spoken language, facilitating advancements in Artificial Intelligence (AI) and Natural Language Processing (NLP) for the Tigre community.
The primary file in this repository, tigre-data-lexicon.txt, provides a structured mapping of Tigre vocabulary, which is essential for training and refining acoustic and language models.
Key Benefits & Applications
This lexicon provides critical infrastructure for various speech technology tasks. Its primary benefits include:
- Improving Text-to-Speech (TTS) Systems: Enhances the naturalness and intelligibility of synthesized speech by ensuring accurate pronunciation.
- Consistent Phonetic Mapping: Establishes a standard for phonetic representation, reducing ambiguity in grapheme-to-phoneme conversion.
- Enhancing Speech Recognition (ASR): Improves the accuracy of Automated Speech Recognition systems by providing a verified vocabulary baseline.
- Low-Resource Technology Development: Serves as a vital building block for creating tools in a low-resource linguistic environment.
- Broad Application Utility: Enables integration across diverse speech applications, from virtual assistants to educational tools.
- Linguistic Research Support: Assists researchers in analyzing the phonological and morphological structures of the Tigre language.
- Scalable Vocabulary Coverage: Designed to allow for easy expansion, accommodating larger vocabularies and domain-specific terminology as data grows.
Dataset Structure
The repository is organized as follows:
tigre-data-lexicon/
├── README.md # This file
└── tigre-data-lexicon.txt # The core lexicon file
File Description
tigre-data-lexicon.txt: A text file containing the lexicon entries. Each line typically maps a word to its phonetic decomposition or pronunciation, suitable for direct ingestion by training scripts (e.g., Kaldi, Espnet, or Coqui).
How to Use
Manual Download You can clone this repository or download the files directly:
git clone https://huggingface.co/BeitTigreAI/tigre-data-lexicon
Loading the Lexicon (Python Example)
You can parse the lexicon file using standard Python I/O:
def load_lexicon(file_path):
lexicon = {}
with open(file_path, 'r', encoding='utf-8') as f:
for line in f:
parts = line.strip().split()
if len(parts) >= 2:
word = parts[0]
phonemes = parts[1:]
lexicon[word] = phonemes
return lexicon
Example Usage
lexicon_data = load_lexicon("tigre-data-lexicon.txt")
print(f"Loaded {len(lexicon_data)} entries.")
Licensing
This dataset is licensed under the Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0).