USD
GBP
EUR
The ability to direct sound to a specific listener without headphones-known as directional sound-has long been a field of research in audio engineering. However, achieving this typically requires large and complex sound sources, and the audio signal tends to be audible anywhere along the beam path.
A new method developed by a Penn State team overcomes these limitations, according to a new study published in the Proceedings of the National Academy of Sciences.
They combined a compact array of ultrasonic transmitters with a specially designed two-dimensional structure, called a metasurface, that manipulates the properties of waves to produce self-bending ultrasonic beams. These beams are not only inaudible to the human ear but also capable of bypassing obstacles. When two such beams intersect, they interact to produce sound within the audible range, confined to a spot just a few centimeters wide, what the researchers call the "audible zone."
"The key innovation is that sound is produced only where the two beams intersect, which allows us to precisely deliver audio to a specific location while keeping the beams themselves from producing audible sound," explained Jiaxin Zhong, a postdoctoral scholar in acoustics at Penn State and the paper's lead author.
Previous studies have shown that audible self-bending beams can bypass obstacles, but due to the long wavelength of audible sound, the sound source usually needs to be on the meter scale, and the signal can be detected anywhere along the beam path. The research team at Pennsylvania State University has come up with a new technology that uses ultrasonic beams that are inaudible to the human ear and can be generated using smaller hardware.
To achieve self-bending beams, they 3D-printed a 16 x 8 cm grid-like metasurface and placed it in front of the ultrasonic transmitter array, precisely controlling the phase of the output beam. "These metasurfaces act like an acoustic lens, manipulating the wavefront and bending the beam as it propagates," Zhong said.
The research team demonstrated that two such beams could pass around a dummy's head and intersect in front of the dummy's face. When the two sound waves interact, they produce a secondary wave with a frequency equal to the difference between the original waves. By using a pair of beams at 40 kHz and 39.5 kHz, the researchers created a sound zone just a few centimeters wide in front of the dummy's head with a frequency of 500 Hz, well within the human audible range.
By varying the frequency of one of the beams, the researchers showed they could use the same metasurface to generate six octaves of audio over a frequency range of 125 Hz to 4 kHz. While these experiments primarily created simple, single-frequency tones, the team also demonstrated that the method could be applied to a 9-second excerpt of the "Hallelujah Cantata" from Handel's Messiah, which contains a range of fluctuating frequencies.
Zhong Jiaxin pointed out that the main problem with the current method is that when the two beams interact, distortion is generated, causing the audio signal to become chaotic. However, the researchers believe that this problem can be solved by adopting new signal processing techniques, including deep learning algorithms that can learn to compensate for distortion.
Currently, the trajectory of self-bending beams is fixed, which means the location of the sound source must be precisely set to avoid obstacles. Zhong Jiaxin said that the team hopes to eventually create reconfigurable beams that can dynamically adjust to avoid different objects. "We envision using adaptive processing algorithms to dynamically adjust the beam trajectory in real time, allowing the beam to intelligently bypass obstacles based on environmental feedback," he said. "This will make the technology more versatile in real-world applications, because in real-world environments, the location of obstacles may move or change."
Zhong Jiaxin also stated that this technology has a wide range of potential applications. For example, it can create personalized audio in public places, such as museum audio guides, without the need for headphones; in cars, different passengers can listen to different audio without interfering with each other; and private audio zones can be created for confidential information. Furthermore, by projecting audio that cancels out the existing sound field, the technology can also achieve localized noise cancellation.
More information: Jing, Yun, Audible enclaves crafted by nonlinear self-bending ultrasonic beams, Proceedings of the National Academy of Sciences (2025). DOI: 10.1073/pnas.2408975122. www.pnas.org/cgi/doi/10.1073/pnas.2408975122
Information source: IEEE.ORG
The article is reprinted from 21dB Acoustics