Article’s

This paper presents an AI-based system for automated podcast transcription and topic segmentation. The system accepts audio uploads, performs speech-to-text transcription, segments content into topics, extracts keywords, generates summaries, and produces PDF reports through a Flask web application. The architecture integrates audio preprocessing, transcription, NLP-based topic segmentation, and report generation to improve accessibility and content discovery. Experimental evaluation demonstrates efficient processing and useful summaries for educational and media applications.