Create Music Visualizations with AI from MIT

Mariano Salcedo, a master's student in the new Music Technology and Computation Graduate Program at MIT, is designing an AI to visualize and express music and other sounds. His research focuses on neural cellular automata (NCA), which merge classical cellular automata with machine learning techniques to generate images that can regenerate. When paired with a musical stimulus, these images can 'show' sounds in action.

Salcedo has developed a web interface that allows users to adjust the relationship between the energy of the music and the NCA system, creating unique visual performances using any audio stream. He aims for the visuals to complement and enhance the listening experience.

Last year, Salcedo, the Alex Rigopulos Fellow in Music Technology and Computation, earned a BS in artificial intelligence and decision making from MIT, where he explored signal processing in machine learning. Now, he is one of five master's students in the inaugural cohort of the program.

The program, directed by Professor Eran Egozy, is a collaboration between MIT Music and Theater Arts and the School of Engineering. It invites practitioners to study and develop new computational approaches to music and includes a speaker series that exposes students to music industry professionals and researchers.

Salcedo has been selected to deliver the student address at the 2026 Advanced Degree Ceremony, which is a great honor and responsibility for him. He hopes that his work will help change the landscape of AI, promoting a more diverse and inclusive approach to technology development. His passion for music resurfaced during his time at MIT, where he began DJing and creating soundscapes using technology.