These projects all apply machine learning and computer vision techniques to process, generate, or understand the Armenian language, spanning tasks like handwriting recognition, speech recognition, text generation, and character detection.
A deep learning project implementing convolutional neural networks (CNNs) to classify handwritten Armenian alphabet c...
A Jupyter Notebook implementing a ResNet-based deep learning model for Armenian handwritten character recognition, tr...
ArmenianWhisper is an open-source summer project focused on creating an AI-powered pipeline for Armenian speech recog...
A computer vision project for detecting and locating Armenian letters from drone imagery. It appears to be a contest ...
This project implements a deep learning system for detecting and recognizing handwritten Armenian characters using a ...
A project implementing an autonomous UAV navigation system that recognizes Armenian alphabet letters from aerial imag...
This repository implements an unconditional Denoising Diffusion Probabilistic Model (DDPM) trained on a custom datase...
This repository implements a machine learning model to convert English phonetic spellings of Armenian words into Arme...
A Jupyter Notebook project implementing a CNN for Armenian handwritten letter recognition using PyTorch. It classifie...
A research project exploring how unsupervised pre-training on untranscribed audio from different language families af...
A tutorial repository for building an Armenian Automatic Speech Recognition (ASR) system using the Kaldi toolkit. It ...
A research project from the National Polytechnic University of Armenia that develops an end-to-end system for generat...
A research project aiming to develop a state-of-the-art Automatic Speech Recognition (ASR) model for the Armenian lan...
A complete Jupyter Notebook project for handwritten Armenian character recognition using a CNN built from scratch in ...
A complete implementation of a decoder-only GPT model trained from scratch on Armenian text. The project includes cor...