Louis Debes

How can an AI system analyze the content of a spoken sentence and match it to a specific 
image? In this project, a multimodal AI system should be developed that maps images and 
spoken language into a shared semantic space. 
The goal is to retrieve corresponding images from a dataset using spoken descriptions as 
input. For example, a spoken query such as “Two people are walking across the street” 
should return images with similar visual content.