Tuesday, 20 October 2026Support us

Aube.

News of progress
PrototypeSingle source

Screen sound from where the action is, in a prototype

Languages for this article
Original · ENESFRITPT

Originally written in English. 5 languages available; yours is one click away.

Generated infographic: 99.9% — Screen sound from where the action is, in a prototype
AI-generated illustration

On a 100-inch screen in a Korean lab, a voice can now come from the exact point where a face appears. Behind the panel, 80 voice coil motors (VCMs), small electromagnetic actuators, shake the panel in a precise pattern. Only the area where the sound should originate is driven to vibrate. The panel becomes the loudspeaker.

The problem being solved is a mismatch that every cinema and living room knows. Conventional large displays use speakers installed apart from the screen, so the sound of an object seen on the left can arrive from the right. The team, which works at the Korea Institute of Machinery and Materials, says this weakens immersion. Earlier panel-vibration systems had a different limit: their sound points and panel sizes were fixed, which restricted both the frequency range and the screen sizes they could serve.

The method works by solving the inverse of a physical problem. The team first measures how the panel responds to vibration, then calculates the voltage to send to each actuator so that only the target area moves. To keep this stable in real time, they built a "selective excitation algorithm" that leaves out actuators that transmit little vibration, and a way to join successive sound signals so that the output does not stutter. AI then identifies where each sound source sits and how large it is, separates one voice or instrument from a mixture, and moves the sound-emitting zone to follow it. In performance tests with an acoustic camera, which maps where sound comes from, the sound appeared at the intended points and moved with the source in the video.

What changes, and for whom? According to the researchers, it is expected to be applied in media art, live performance, exhibitions, digital signage, large LED displays and in-vehicle screens, where a visual and its sound must match. Modules can be combined to reach a required screen size, and several modules driven together can produce low-frequency effects similar to a large woofer. The team also plans to extend the approach to modular LED tile displays. The sources give no delivery date, no price and no commercial partner, so this remains a demonstration, not a product. The next test that matters is whether the 99.9% accuracy holds outside the lab, with real content and real rooms.

99.9%Target-zone selection accuracy reported in the team's performance tests

Sources — read the originals(Paris time)

Phys.org — TechnologyEN
0000

Read next

Comments

Loading the thread…

Sign in to leave a comment. Sign in