Generating realistic 3D sound from ordinary videos using AI and visual cues
Peer-Reviewed Publication
Updates every hour. Last Updated: 2-Apr-2026 16:15 ET (2-Apr-2026 20:15 GMT/UTC)
Researchers have developed an AI technique that adds realistic 3D sound to ordinary videos recorded with a single microphone. By using visual information in the video, the system makes sounds appear to come from the correct directions, even when no spatial audio was originally recorded.
Drug resistance has accelerated in recent years with the emergence of deadly bacteria and “superbugs.” In response to this global health crisis, UC San Diego biologists have developed a new CRISPR-based technology capable of removing antibiotic-resistant elements from populations of bacteria.