CNMAT and Meyer Sound Collaborate on Spherical Loudspeaker Project

Meyer Sound is supporting a University of California Berkeley’s Center for New Music and Audio Technologies (CNMAT) effort to create a loudspeaker array capable of mimicking the kinds of radiation patterns exhibited by acoustical musical instruments. That project is the subject of a paper titled “A Compact 120 Independent Element Spherical Loudspeaker Array with Programmable Radiation Patterns” that the CNMAT group presented at the recent Audio Engineering Society (AES) convention in Paris.

CNMAT director David Wessel points out, “It is exceedingly rare to find a company that values, much less supports, early-stage research.” He adds, “Commercial partners are vital to getting our work out into the world, where our research can be brought to fruition in new products. We are fortunate to be associated with Meyer Sound, which is one of the few companies to fully understand how working with us strengthens their future as much as it does ours. We get to combine our expertise with theirs and consult with the likes of John and Perrin Meyer. The synergy really is exciting for both of us.”

The principle being investigated in the loudspeaker array project has to do with the ways that acoustical instruments radiate sound very differently from standard loudspeaker arrays. The radiation pattern of instruments varies with frequency and is different for each instrument. In contrast, loudspeakers are typically designed to opposite goals: they are intended to be as free as possible from having an identifiable sonic character or exhibiting variations with frequency, and should be completely consistent from unit to unit. The difference between instrument behavior and that of existing loudspeaker arrays becomes glaringly obvious in musical situations encompassing both acoustic and electronic instruments, where the sounds usually do not blend smoothly.

The CNMAT project seeks to create a compact loudspeaker array that resides in a single location and has a radiation pattern controlled by onboard DSP. After initial experiments that simply combined existing commercial self-powered loudspeakers into a tight cubic array, research turned to creating spherical prototypes. The most recent prototype, which is reported on in the AES paper, is a 10-inch-diameter icosahedron incorporating 120 1.25-inch drivers developed by Meyer Sound, each driven from a separate audio channel, plus circuit boards for all of the control and class-D amplification functions. The array is controlled through Gigabit Ethernet.

Wave field synthesis is currently a hot research topic, however the majority of research has been focused on transducers composed of multiple transducers in planar arrays. This spherical loudspeaker is unique in that it’s a tightly packed spherical array, making it possible to create three dimensional wave fields using WFS. Both planar and spherical loudspeakers have interesting possibilities for future enhancements including early reverberation enhancement.

Research related to the project was presented by CNMAT researcher Peter Kassakian to an IEEE International Conference on Acoustics, Speech, and Signal Processing last year, as well as in papers presented to two earlier AES conventions. In addition to Meyer Sound, support for the project has come through a UC Discovery Grant from the Industry-University Cooperative Research Program (IUCRP), which forms partnerships tying together the University of California, industry sponsors, and the State of California.

“Our mission is to break new ground and discover how the technologies we create can be made useful,” says CNMAT research director Adrian Freed. “The spherical loudspeaker array project is a new approach to sound reproduction that we feel could have potent applications for room acousticians, musicians, and the research community, just for a start. It’s really a thrill to do this work, but we are acutely aware that pure research can only happen with support, and want to express our gratitude to Meyer Sound, the IUCRP, and, of course, the University of California for believing in what we’re doing enough to invest themselves in it. We think it will end up being a good investment for everyone, and we get to have a lot of fun opening a new world in the process.”

The paper was presented during the Loudspeakers and Sound Reinforcement papers session, with an associated poster presentation following the next day. Freed reports that the project received avid interest in Paris. “I had great success with the speaker,” he says. “The poster presentation was supposed to last from 9-10:30 AM, but I ended up talking continuously until 5 PM. I even spent a couple of hours talking with (audio research authority) Stanley Lipschitz about it! There is lots of interest, including from someone who wants to use one right away to do room measurements by combining spherical microphone array measurements with synthesized harmonics from the speakers.”

Convention attendees were particularly excited that there was a functioning prototype, rather than simply theoretical research to report on. Several researchers speaking to Freed commented on the problems they had encountered with dodecahedral and icosahedral loudspeakers they had built and used, and how heartened they were to see a speaker of this type that solves those problems and gives unprecedented individual control over each driver.

After stirring up the AES convention, Freed took the prototype to IRCAM (Institut de Recherche et Coordination Acoustique et Musique), the famous French computer music center located at the Centre Pompidou, where Wessel and Freed both worked in the past. As at the AES, Freed started a one-hour presentation at 3 PM and answered the last question six hours later.