Authors:
J. Angelin Jeba, S. Rubin Bose, R. Regin, S. Suman Rajest, Md Mahdi Hasan

Addresses:
Department of Electronics and Communication Engineering, CEG Campus, Anna University, Chennai, Tamil Nadu, India. Department of Electronics and Communication Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. Department of Research and Development & International Student Affairs, Dhaanish Ahmed College of Engineering, Chennai, Tamil Nadu, India. Department of Management, St. Francis College, Brooklyn, New York, United States of America. jebaangelin@gmail.com, rubinbos@srmist.edu.in, reginr@srmist.edu.in, sumanrajest414@gmail.com, mhasan6@sfc.edu

Abstract:

Tamil transcript summarizing improves Tamil video information extraction for study, learning, and accessibility. However, complex grammar, dialects, colloquialisms, and lack of linguistic resources and tools require specific approaches to ensure accurate and successful summarization. A new methodology that smoothly blends natural language processing (NLP), translation, and text-to-speech capabilities is designed to extract crucial insights from abundant internet video footage. The application effectively collects video transcripts using the YouTube Transcript API library and spacy for NLP for extractive summarization. Users can choose from Small, Medium, or Large summary lengths. Translating between English and Tamil is smooth, and the gTTS library powers text-to-speech features for summaries in both languages. The Tkinter library-built interface has a search tool for easy summary navigation. Real-time performance monitoring shows CPU and memory use and summary execution times for efficiency measurement. Hover effects for buttons and multithreading for audio file management improve user experience. Users wishing to extract key insights from YouTube videos can benefit from this tool’s versatility, multilingual support, and simplified summarizing.

Keywords: Tamil Transcript Summarization, Natural Language Processing (NLP), Multimodal Integration, Extractive Summarization, YouTube Transcript API, Video Summarization, Real-Time Processing.

Received on: 19/10/2023, Revised on: 11/12/2023, Accepted on: 05/01/2024, Published on: 07/03/2024

DOI: 10.69888/FTSCL.2024.000179

FMDB Transactions on Sustainable Computer Letters, 2024 Vol. 2 No. 1, Pages: 26-39

• Views : 84
• Downloads : 8

Download PDF