How well does a speaker work in a given space? That question is way more complicated than you might think. Not only do speakers perform differently in settings with different surfaces, shapes and furnishings, but it turns out that humans are pretty bad at comparing those varied acoustic environments. Faced with more than two choices, our brains are simply unable to keep the unique sounds of each in mind.
But in the basement of a Danish university, that problem vanishes into thin air — thanks to a new system that can accurately reproduce the sounds of any space. With the help of 43 speakers, a specially designed room and a new method of recording and reproducing sound, researchers can now make their lab echo any environment on Earth, from gigantic concert halls to the claustrophobic interiors of cars.
“We wanted to try to bring real life into the lab,” says Neo Kaplanis, an audio researcher and PhD student at Aalborg University, where the lab is located. Kaplanis is a Tonmeister — a sound recording expert — who is also employed by Bang & Olufsen, the high-end speaker manufacturer. (Though the lab was put together along with the company, its work is not proprietary.) For his PhD project, he decided to tackle a problem that has stumped the audio world since speakers were invented.
The issue stems from sound waves’ tricky tendency to fill a space three-dimensionally. Since they consist of waves that travel from a specific source into space, they run into plenty of obstacles as they go. As the energy of each wave travels, it hits air, then bumps into reflective and muffling surfaces. That’s why only high-end cars have top-of-the-line speaker systems — it simply costs too much to test speakers inside the cars themselves again and again.
Despite advances in headphones, these tiny devices are still incapable of reproducing the spatial environment of sounds that come from far away and super close. Similarly, headphones can’t transmit the exact experience of bass frequencies that, when played through normal speakers, reverberate throughout the body.
But the Danish lab comes armed with a multi-microphone array and a new type of recording — one that takes a kind of acoustic fingerprint of a space as it records sounds. The recordings don’t just capture sounds but record exactly where sound reflections come from within a space. Kaplanis also designed a computer program that plays back sounds from those precise locations in an anechoic, or sound absorptive, room.
“You can reproduce the sound field in the system,” explains Kaplanis. “The sound quality is so real.” When people visit the lab and enter, they do so without the help of sight or context, so their brains trick them into thinking that they’re in the kind of space re-created by the speaker array and the computer program. “When I recorded rooms I knew already, it felt just like the real thing.”
It’s hard to imagine, but for listeners, the basement lab can become anything from a massive cathedral to an intimate room. Think of it as sound plus space — an experience that’s immersive and even unsettling at times. You can hear some samples here.
So what’s the point? The commercial applications are clear: Being transported to another place via sound could one day be used to create more realistic gaming experiences, and similar technology is already being deployed in at least one Finnish club that’s an audiophile’s fantasy. And being able to test sound in different acoustic environments without physically going there could cut costs for speaker manufacturers and audio makers in search of great acoustics.
But there’s more to the system, says Kaplanis’s adviser, Søren Bech. (Though 80 percent of his time is spent as a senior research adviser for Bang & Olufsen, he’s a professor at Aalborg University, too.) Both adviser and student are part of the DREAMS ITN project, a multidisciplinary consortium devoted to figuring out ways to control and remove reverb in different applications that help humankind. That can range from making car speakers better (a challenge that can thus far be accurately modeled only through Kaplanis’s system) to making sounds clearer for hearing-impaired people.
What might seem like a cool party trick could one day allow researchers to, say, hone the acoustic environments of stadiums or concert halls or create better acoustic testing for people with hearing challenges. “This system could be used to answer really complex research questions,” says Kaplanis, who points out its potential use in helping study how people respond to sound on a psychological and physiological level. Then, he grins. “But it also sounds amazing.”