Text Summarization and Multilingual Text to Audio Translation using Deep Learning Models

Different technologies has been combined in this work to read the research paper or review paper. This work introduces a system that can automatically summarize research or review papers into audio files in Hindi and English, making it easier for users to understand complex literature. The study makes a major contribution to encourage inclusivity and accessibility in the scholarly domain by bridging the linguistic gap in academic resources. The system enables users to upload PDF documents and preprocesses textual data by using natural language processing techniques. Optimal performance was attained by experimenting with different pretrained models, such as facebook/bart-large-cnn for English summarization and facebook/mbart-large-50-one-to-many-mmt for Hindi to English translation. Furthermore, using the facebook/mms-tts model made it easier to convert text to audio in both languages. This system offers an easy-to-use solution that will be especially helpful to people who might have trouble understanding English. In addition to improving literature accessibility, the implementation of the system will help researchers and students save important time.