These projects are all focused on developing and evaluating AI models and applications specifically for the Armenian language, covering areas like text generation, speech processing, and sign language translation.
ArmBench-LLM is a specialized evaluation framework for benchmarking large language models on Armenian language tasks....
A project exploring Armenian language autocomplete using multiple NLP approaches including Word2Vec, LSTM, BERT trans...
A specialized evaluation benchmark for testing text embedding models on Armenian language tasks including semantic te...
A Western Armenian language model project using transformer architecture, implemented in Jupyter notebooks. Includes ...
An end-to-end pipeline for Armenian-to-English speech and text conversion, featuring Armenian automatic speech recogn...
A project to create a bilingual Western Armenian-English large language model using QLoRA fine-tuning on Qwen 2.5 1.5...
AzgIntel is an AI-powered NLP pipeline for Armenian text classification using a BERT-based multilingual model. It pro...
A robust Speech-to-Text (STT) service with multi-provider API integration (Google Gemini, ElevenLabs, Whisper, HiSpee...
A full-stack Armenian speech recognition system with React frontend, Spring Boot backend, and Azure deployment. Inclu...
A real-time Armenian Sign Language translator using MediaPipe for hand landmark extraction and an LSTM neural network...
A real-time Armenian sign language translation tool using MediaPipe for hand landmark extraction and an LSTM neural n...
A machine learning project for detecting AI-generated text in Armenian. It includes a custom dataset, fine-tuned tran...
A personal playground repository for testing and comparing Armenian speech recognition models. Contains Jupyter noteb...
Armo-GPT is a bilingual Armenian-American AI chatbot platform featuring a sophisticated persona system with four dist...
A speech-to-text model for Armenian language using Wav2Vec2-BERT, fine-tuned on the Armenian Common Voice dataset. Th...
ArmSpeechTT is a fine-tuned Whisper model for Armenian speech-to-text, trained on Common Voice data with a webcam dem...