Voice-Based-Image-Retrieval-System (1)

This project proposes a voice-based image retrieval system using multimodal deep learning. It allows users to search for images using natural spoken queries. Spoken input is transcribed with Whisper and encoded with CLIP for cross-modal alignment. Retrieved images are re-ranked with Stable Diffusion to improve semantic relevance. The system is designed for real-time performance and accessibility, especially for visually impaired users. It combines voice input, image retrieval, and diffusion-guided ranking in one platform. The architecture includes a React frontend, FastAPI backend, and vector database for fast similarity search. This solution fills a gap not addressed by existing text-only or voice-limited systems.This work was conducted at Arab International University (AIU),Syria. The official website of the university is: https://www.aiu.edu.sy.This work is licensed under CC By 4.