□ A core audiovisual convergence display technology has been developed that allows sound to be heard from the exact location of a person or object shown on the screen. Without separate speakers, the technology vibrates the display panel itself to generate sound from a selected area of the screen. It can also move the sound-emitting location in real time as the sound source moves across the screen.
□ A research team led by Dr. Dongjoon Kim, Senior Researcher at the AX Convergence Research Center of the Korea Institute of Machinery and Materials (KIMM, President Seog-Hyeon Ryu), developed a modular-grid-based audiovisual convergence display technology that precisely controls multiple vibration actuators attached to the rear of a display to emit sound from desired locations on the screen.
□ The technology controls the “vibration field,” or the distribution of vibrations across the panel surface, using 80 voice coil motors (VCMs) installed on the rear of the panel. After analyzing the panel’s vibration response characteristics, the system calculates the voltage signals to be applied to each actuator using an inverse problem method, allowing only the desired area of the panel to vibrate selectively. By changing the sound-emitting zone in real time according to the movement of a sound source in the video, the system can create the effect of sound moving together with the on-screen source.
□ Performance testing demonstrated a target-zone selection accuracy of 99.9% when reproducing sound from designated areas of the screen. The difference in vibration amplitude between activated and non-activated zones reached as high as 35 dB, demonstrating that vibration and sound can be effectively concentrated in specific areas of the display.
□ Conventional large-scale displays typically use speakers installed separately from the screen, causing a mismatch between where an object appears on the screen and where its sound appears to come from, which can reduce immersion. Existing panel-vibration speaker technologies also have limitations because the sound-emitting locations and panel sizes are fixed, restricting the available frequency range and scalability to different screen sizes.
□ KIMM overcame these limitations through a modular-grid concept that combines turning the display itself into a speaker with a modular display architecture. The screen itself generates sound while selectively vibrating only the specific area where the sound source appears. Modules can be combined or separated according to the required screen size, enabling the technology to scale from home TVs to exhibition displays, digital signage, and large-scale LED displays. Driving multiple modules simultaneously can also produce low-frequency sound effects similar to those of a large woofer.
□ To address computational instability that can occur during vibration-field control, the research team developed a “selective excitation algorithm” that excludes actuators with low vibration transmission effectiveness from the calculation. The team also applied a technique that smoothly connects successive acoustic signals, reducing sound interruptions and noise that may otherwise occur during real-time computation.
□ The research team further integrated AI-based sound source separation and localization technologies. AI automatically identifies the position and size of sound sources in the video and separates a desired sound from a mixture of multiple sounds. The separated sound is then reproduced in real time from the corresponding area of the screen, automatically aligning the on-screen location of the sound source with the direction from which the sound is heard.
□ The research team built an approximately 100-inch prototype combining a transparent LED display film with an aluminum panel and installed 80 VCMs on the rear of the panel. Performance tests using an acoustic camera confirmed that sound was generated from the intended locations on the screen and that the sound-emitting location moved in accordance with the movement of the sound source in the video.
□ By aligning the location of an on-screen sound source with the actual direction from which the sound is heard, the technology can provide immersive content that integrates visual and auditory experiences. It is expected to be applied in a wide range of fields, including media art, performances and exhibitions, digital signage, large-scale LED displays, and in-vehicle displays.
□ Dr. Dongjoon Kim, Senior Researcher at KIMM’s AX Convergence Research Center, said, “The modular-grid technology aligns the position of a sound source on the screen with the direction from which the actual sound is heard, creating an immersive audiovisual experience that has been difficult to achieve with conventional displays.” He added, “We plan to expand the technology to modular LED tile displays and further develop it into a next-generation core technology that can be applied across a wide range of display applications.”
□ Meanwhile, the research team has obtained a Korean patent (No. 2741448) for the related “Modular Acoustic Radiation System,” and an SCI-indexed paper presenting the research results has been published. The research was supported by KIMM’s in-house creative and challenge-oriented research program and institutional research program.

###
The Korea Institute of Machinery and Materials (KIMM) is a non-profit government-funded research institute under the Ministry of Science and ICT. Since its foundation in 1976, KIMM is contributing to economic growth of the nation by performing R&D on key technologies in machinery and materials, conducting reliability test evaluation, and commercializing the developed products and technologies.