PromoGenerator

The project is a web application that allows users to upload a series of images to automatically generate a promo video for exhibitions or collections, tailored for small to medium local museums. The backend, built with Flask, processes the uploaded images by leveraging OpenAI’s GPT-4 for image description generation and TTS (Text-to-Speech) to produce corresponding narration. The video is created by combining the images with the generated audio using the MoviePy library. The final result is a video, which can be downloaded or viewed directly in the browser.

The application demonstrates creativity by combining advanced AI technologies—OpenAI for natural language processing and speech generation—with video production in a seamless workflow. The project has practical applications for museums looking to create engaging, automated promotional content for their exhibitions. Its real-world impact lies in empowering smaller institutions that may lack the resources to produce high-quality video content, thus improving their online presence and audience engagement. The project aligns perfectly with the hackathon’s AI productivity theme, as it harnesses AI to significantly streamline and automate the traditionally time-consuming process of video production, saving museums both time and costs.