Article
Video Metadata Generation & Classification
The Video Metadata Generation and Classification project is a web-based application developed to automatically generate metadata and classify videos based on their content. With the rapid growth of video content, manually generating metadata has become a time-consuming process. To address this problem, the proposed system automates the process of video analysis using speech recognition and keywordbased classification techniques. Users can upload a video through a Flask-based web interface, where the uploaded video is first previewed. The system then extracts the audio using FFmpeg and converts the speech into text using Google Speech Recognition, which implements Automatic Speech Recognition (ASR). The extracted text is analyzed to generate relevant keywords and classify the video into predefined categories such as Sports, Food, Education, and General using a Keyword-Based Classification Algorithm. The system is developed using Python, Flask, HTML, CSS, FFmpeg, and the SpeechRecognition library. It provides a simple, efficient, and cost-effective solution for automated video organization and retrieval, making it suitable for educational platforms, digital libraries, media organizations, and multimedia content management systems.
Full Text Attachment
