Headphones Use AI to Isolate Specific Speaker’s Voice

Engineers at the University of Washington have developed an AI-powered headphone system called "Target Speech Hearing" that allows users to focus on a single speaker in noisy environments by looking at them for just a few seconds. Presented at the ACM CHI Conference, this technology isolates the target speaker's voice using machine learning, even as the user moves around. This innovative system, which builds on previous "semantic hearing" research, offers promising applications for personal audio experiences in crowded places. The code is available for further development, though the system is not yet commercially available.

article

1 comment