Muhammad Shoaib Farooq and Ali Abbas from the Department of Computer Science at the University of Management and Technology, Lahore, conducted a study to generate Urdu poetry using deep-learning techniques. They utilized various models, including Long Short-term Memory Networks (LSTM) and Gated Recurrent Unit (GRU), to understand, analyze, and generate human-like language. The data used for this study was collected from the Rekhta website, which contains a mix of Urdu poems and Ghazals. The main goal was to create a model that generates pure Urdu poems using complete data, not sampled data. The generated poems showed good accuracy.
Publication date: 26 Sep 2023
Project Page: Not Provided
Paper: https://arxiv.org/pdf/2309.14233
