nairabarseghyan/Armenian-tokenizer
A student project implementing multiple tokenization methods (BPE, WordPiece, SentencePiece, tiktoken) for Armenian language using Wikipedia data. Includes custom tokenizer implementations, trained models, and evaluation notebooks.
A demonstration repository for a pre-trained Armenian text embedding model, showcasing applications in text classific...
A Python project that creates a morphological augmentation layer for Western Armenian language processing. It integra...
A project exploring Armenian language autocomplete using multiple NLP approaches including Word2Vec, LSTM, BERT trans...
AzgIntel is an AI-powered NLP pipeline for Armenian text classification using a BERT-based multilingual model. It pro...
A rule-based morphological analyzer for Modern Eastern Armenian built with uniparser-morph. It performs full morpholo...
A Western Armenian language model project using transformer architecture, implemented in Jupyter notebooks. Includes ...