Multimodal technique for analyzing audio and visual data improves performance of machine-learning models

Researchers from MIT, the MIT-IBM Watson AI Lab, IBM Research, and elsewhere have developed a new technique for analyzin