Psychoacoustics helps reproduce sound the way the brain prefers to hear it

[From The New York Times, where the article includes a multimedia feature; information about Audyssey’s latest product, Personal Surround, is available here]

[Image: Tyson Yaberg of Audyssey Laboratories listened to an experimental system at the University of Southern California. Audyssey’s goal is to make dens and living rooms sound like concert halls and movie theaters. Photo by David Ahntholz for The New York Times.]

Sound, the Way the Brain Prefers to Hear It

By GUY GUGLIOTTA
Published: September 5, 2011

LOS ANGELES — There is, perhaps, no more uplifting musical experience than hearing the “Hallelujah” chorus from Handel’s “Messiah” performed in a perfect space. Many critics regard Symphony Hall in Boston — 70 feet wide, 120 feet long and 65 feet high — as just that space.

Some 3,000 miles away, however, a visitor led into the pitch-blackness of Chris Kyriakakis’s audio lab at the University of Southern California to hear a recording of the performance would have no way to know how big the room was.

At first it sounded like elegant music played in the parlor on good equipment. Nothing special. But as engineers added combinations of speakers, the room seemed to expand and the music swelled in richness and depth, until finally it was as if the visitor were sitting with the audience in Boston.

Then the music stopped and the lights came on. It turned out that the Immersive Audio Lab at U.S.C.’s Viterbi School of Engineering is dark, a bit dingy, and only 30 feet wide, 45 feet long and 14 feet high.

Acousticians have been designing concert halls for more than a century, but Dr. Kyriakakis does something different. He shapes the sound of music to conform to the space in which it is played. The goal is what Dr. Kyriakakis calls the “ground truth” — to replicate the original in every respect. “We remove the room,” he said, “so the ground truth can be delivered.”

Dr. Kyriakakis, an electrical engineer at U.S.C. and the founder and chief technical officer of Audyssey Laboratories, a Los Angeles-based audio firm, could not achieve his results without modern sound filters and digital microprocessors.

But the basis of his technique is rooted in the science of psychoacoustics, the study of sound perception by the human auditory system. “It’s about the human ear and the human brain, and understanding how the human ear perceives sound,” Dr. Kyriakakis said.

Psychoacoustics has become an invaluable tool in designing hearing aids and cochlear implants, and in the study of hearing generally. “Psychoacoustics is fundamental,” said Andrew J. Oxenham, a psychologist and hearing expert at the University of Minnesota. “You need to know how the normally functioning auditory system works — how sound relates to human perception.”

The field’s origins date back more than a century, to the first efforts to quantify the psychological properties of sound. What tones could humans hear, and how loudly or softly did they need to be heard?

Pitch could be measured in hertz and loudness in decibels, but other phenomena were not so easily quantified. Human hearing can discern the movement of sound with a surprising degree of accuracy. It can distinguish timbre, the difference between a clarinet and a saxophone. It can remember patterns of speech, to immediately identify a friend in a phone call years after last hearing the voice. And a parent can effortlessly sift the sound of an infant’s cry from the blare of a televised football game.

Finally there were the imponderables, things we do with our hearing simply because we can. “Everyone knows the sound of a bowling ball as it rolls down the alley,” said William M. Hartmann, a Michigan State University physicist and former president of the Acoustical Society of America. “What is it about that sound that we can identify?”

For much of the 20th century, engineers devoted themselves to developing acoustical hardware like amplifiers, speakers and recording systems. After World War II, scientists learned how to use mathematical formulas to “subtract” unwanted noise from sound signals. Then they learned how to make sound signals without any unwanted noise.

Next came stereo. By recording two tracks, engineers could localize sound for the listener. “Simple enough,” said Alan Kraemer, chief technological officer for SRS Labs, an audio company in Santa Ana, Calif. “If something’s louder on one side, you’ll hear it on that side.”

But stereo had no real psychoacoustics. It created an artificial sense of space with a second track, but did so by dealing with only one variable — loudness — and enhanced human perception simply by suggesting that listeners separate their speakers.

The digital age changed all this, allowing engineers to manipulate sound in ways that had never been tried before. They could create sounds that had never existed, eliminate sounds they did not want and use constant changes in filter combinations to deliver sound to listeners with a fidelity that had never before been possible.

Digital technology has led to innovations that have been critical in improving sound reproduction, in tailoring hearing aids for individual patients and in treating hearing impairment and developing cochlear implants — tiny electronic devices linking sound directly to the auditory nerve of a deaf person.

“Hearing aids are not the same as glasses,” said Dr. Oxenham, at the University of Minnesota. “It’s never been just about hearing sound; it’s also about understanding sound and separating it from background noise. We can help with microprocessors. Without them it would have been impossible.”

Despite recent advances, however, psychoacoustics has shown engineers that they still have a long way to go. No machine can yet duplicate the ability of the human ear to understand a conversation in a crowded restaurant. People with cochlear implants have “a terrible time” with background noise, Dr. Oxenham said. They also have trouble with pitch perception and distinguishing the sounds of different instruments. “Hearing loops,” which are transmitters that broadcast sound signals directly to a receiver in a hearing aid, are catching on in concert halls, places of worship and even subway booths.

“The technology is really being strained,” said Dr. Hartmann, at Michigan State. Because of psychoacoustics, “we know so much more, and therefore we can do so much more,” but “there is so much more to do.”

One factor that slows the pace of innovation, Dr. Hartmann suggested, is that the human auditory system is “highly nonlinear.” It is difficult to isolate or change a single variable — like loudness — without affecting several others in unanticipated ways. “Things don’t follow an intuitive pattern,” he said.

It was this anomaly, in part, that led Dr. Kyriakakis in the 1990s to venture into psychoacoustics. He, his U.S.C. film school associate Tomlinson Holman, and their students were trying to improve the listening qualities of a room by measuring sound with strategically placed microphones.

“Often our changes were worse than doing nothing at all,” Dr. Kyriakakis recalled. “The mic liked the sound, but the human ear wasn’t liking it at all. We needed to find out what we had to do. We had to learn about psychoacoustics.”

The trick was to establish a baseline for what sounded best, and there was no guidebook. So Dr. Kyriakakis and his students went to Boston Symphony Hall in 1998 to conduct a series of sound tests and to record the “Messiah.”

At that time, acousticians had long known that a shoebox-shaped concert hall like Boston’s offered the best sound, but what was important for Dr. Kyriakakis was to know why the human ear and the human brain that processed the signal felt that way.

Back in Los Angeles, his team began a series of simple experiments. Listeners were invited into the labs to hear the Boston tests and music and to rate the sound, using a scale of 1 to 5. Researchers shifted the sound to different combinations of speakers around the room.

Statistics showed that speakers directly ahead, combined with speakers 55 degrees to either side of the listener, provided the most attractive soundstage. The “wide” speakers mimicked the reflection from the side walls of the concert hall by causing the sound to arrive at the listener’s ears milliseconds after the sound from the front. Sound from other angles did not have as great an effect.

Next, the team asked listeners what combination of speakers gave the best impression of “depth of stage.” Here again, statistics showed a clear preference for speakers in front of listeners and high above them. This sound — also slightly delayed — gave the ear and the human brain a sense of where the different instruments were on a bandstand.

With these results as his template, Dr. Kyriakakis founded Audyssey. His idea was to make dens and living rooms sound like concert halls and movie theaters. Microprocessors made it possible to filter sound to minimize distortion and add the delays that make the music sound nearly perfect to the human ear from anywhere in the room.

Audyssey’s first product, MultEQ, started with a five-speaker configuration, but for a full concert-hall-like effect, it now offers what Dr. Kyriakakis calls an “11.2” system: three speakers in front of the listener; two elevated speakers; two wides; two speakers slightly behind the listener, and two speakers directly in back. Audio-video receivers with Audyssey’s latest MultEQ technology cost $1,000 to $2,000.

For the unsophisticated listener, top-of-the-line sound equipment by itself seems good enough, but it is not like the psychoacoustically adjusted version. A video of the Eagles singing “Hotel California” sounded nice to a visitor until Audyssey’s hardware director, Andrew Turner, pointed out that there was no bass when the volume was low. He flicked a switch and the bass returned, enriching the music with startling effect.

“At the concert itself, where there was a big room with a lot of high-volume sound, you could hear the low tones,” he said. “But here in the studio, your brain is filtering them out as irrelevant at low volume. So you have to restore them.

“It’s pure psychoacoustics.”

This entry was posted in Presence in the News. Bookmark the permalink. Trackbacks are closed, but you can post a comment.

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

*
*

  • Find Researchers

    Use the links below to find researchers listed alphabetically by the first letter of their last name.

    A | B | C | D | E | F| G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z