Sound Advice on the Audio Dynamics of Live Events

Crowds of people, like this one at Convening Leaders 2015, tend to exhibit what is known as the “cocktail-party effect,” in which people hear up to three different conversations at the same time in a crowded space. But it becomes mentally taxing after a while to discern just one. (Jacob Slaton)

Colby Leider, Ph.D., is associate professor and director of the music-engineering technology program at the University of Miami’s Frost School of Music. He teaches courses in environmental acoustics, acoustic ecology, architectural acoustics, loudspeaker design and analysis, and digital-signal processing. When he’s not in the classroom, Leider consults with architects on acoustical treatments for a variety of large venues — including a private, 4,000-square-foot museum project in Nicaragua that he had just started work on when Convene spoke with him in March. We asked Leider to apply his acoustic expertise to the spaces that meetings inhabit.

Why is sound such a problem at many meetings?

Some buildings, when they get beyond a certain size — that is true especially in convention halls and auditoriums — once a certain volume or threshold is exceeded, then maintaining an appropriate intelligibility quotient becomes increasingly difficult. There are about 10 different ways that we can measure or even predict the intelligibility of a space or speech.

What are those measurements a function of? Architectural design? Building materials?

It’s mostly construction and a function of the volume of the room — [measured in] cubic meters or cubic feet, the total surface area of the room. And the most important part is the materials that are on the walls, on the floor, or on the ceiling. Is there, for example, an acoustical tiled ceiling? Is the floor marble or polished concrete? Or is it carpeted?

For all of these different materials, there are tables of absorption coefficients. Let’s say you’re going to have a conference in Beijing at the Marriott. As long as I can get raw numbers of the dimensions of the room, what the walls, the floor, and the ceiling are made of, I can predict through a computer program [whether it meets] our proposed standard for a “Good Housekeeping Seal of Approval” for intelligibility.

How does the number of people in the room and the type of situational environment come into play? A museum, for example, needs to be more hushed than a space holding a social-networking event.

It’s definitely a function of the number of people in the room. Each human body absorbs [sound]. Absorption is measured in units of sabins — named after the first acoustician of record, Wallace Sabine, who was a physicist at Harvard in the early 1890s. The regents of Harvard came to him and said, “We have this museum, Fogg Hall. It sounds horrible. We cannot have a conversation in it.” They [wanted him to] fix it. He said there was no theory about how sounds propagates indoors, and [they instructed him to] develop one. This was his lifelong research — developing these equations as it characterized rooms and essentially speech and intelligibility.

[Those equations have] lasted 115 years now, the science that he started. Each person has and exhibits typically between three and five sabins of absorption. The more people in the room, the more “dead” we call the room; the more reverberant it becomes, and reverberation is definitely the enemy of intelligibility. It helps in terms of having more people in a room as far as absorption. But if they’re all talking, then all bets are off, because they’re generating so much more noise than they’re absorbing.

“We need to develop a standard.”
— Colby Lieder

Crowds of people tend to exhibit what is known in the psychoacoustic literature as the so-called “cocktail-party effect.” The brain has a really good neuroplasticity to adapt to different environments. Try this sometime at a restaurant: See how many different conversations you can hear at the same time. The brain can usually pick out three different conversations. But it becomes very mentally taxing after a while to discern just one. That process of discerning is called “sound-source segregation” or “stream segregation.” It actually makes you physically tired after a while, because you’re causing your brain to work overtime. It’s much better to have a very absorptive room [when there are] a lot of people trying to talk to each other.

These are all aspects that are dependent on the space or the building itself, but not really within a meeting organizer’s control. What options do meeting professionals have to improve sound quality at meetings?

That’s a good question. [We need to develop] a standard that’s agreed upon by an organization, such as the Audio Engineering Society [AES]. This could actually become an AES standard. For example, one of the standards that the AES came up with was the compact disc [CD] back in 1983. They standardized what a compact disc looks like, how many tracks you can fit on it, how long it goes, and all of these sorts of things.

For the rooms that don’t meet the standard that we would propose, there are different kinds of treatments that could be brought in. Those treatments can either be permanent or temporary, such as putting absorber panels up on the wall, and a few other tricks so the standard is met. Or, they would have to hire acousticians to come in and work with an architect to make a permanent fix, which is usually a good investment.

Sound can do one of three things: It either can be absorbed, or it can be reflected, or it can be diffused — scattered like a diamond scatters light. The hardest, the most expensive problem to treat is that of sound isolation, the leakage from one room to another. If you have ever been to a movie theater that is not THX-certified, it’s very likely that you’ll hear bleed-over from the adjacent theater right next to you. But in those [theaters] that have been certified, you generally won’t hear any bleed at all. The reason for that is purely due to the amount of mass separating those walls. Ideally, it would be a concrete wall separating the rooms, or a double-fitted wall. If that isn’t available, then the venue could elect to do a less-expensive and temporary fix, which would be to hang heavy velour curtains, or to actually build another wall — a wall between a wall — which can really help as well.

What about at convention centers where session spaces are separated by air walls?

I would borrow from the technology of the recording studio, [which] has dealt with these kinds of problems for the last 75 or so years. We use a gobo — basically portable walls that are on rollers, about four inches thick. They’re filled with an insulating material like Owens Corning insulation.

So not all temporary fixes are expensive?

No. It would definitely be more expensive if you wanted the venue to find a permanent solution to sound insulation and noise abatements, and increasing the intelligibility. The other thing we haven’t talked about yet is that most venues I’ve seen for conventions do sound reinforcement. That’s the placement of loudspeakers for amplification of [the presenters]. They often place the loudspeakers in the absolute worst position. That is a very simple fix. I have computer software that can be aligned and take a few measurements in a room. We can then show, very quickly, the acoustically best place to place the speakers to maximize coverage for the audience and to make sure everyone can hear what’s being said.

Can you give an example of bad loudspeaker placement?

It’s a very common thing [to place] a left and a right loudspeaker at the very front of a large room. A lot of rooms aren’t capable of supporting inferior playback. If the room is primarily meant for speech and not for music, it makes much more sense to have a single speaker front and center. The other thing that sometimes happens is that the AV personnel — and I’ve seen this happen at different convention centers — will place a podium microphone in the absolute worst spot. If you incorrectly place a podium microphone, or if you use the wrong kind of microphone, which also happens quite a bit, there is an effect called “comb filtering.” The microphone is receiving the direct signal from the human that is speaking. But it’s also picking up, a few microseconds later, a reflection — the first order of reflection of that speech bouncing off the podium itself and then hitting the microphone. Those two signals can sometimes cancel each other out. It actually makes certain syllables and certain consonants diminished in amplitude. It’s like destroying the speech.

What are your thoughts about open-air learning environments, where sound bleed is especially a problem?

That’s a great question. [Relative to this environment] is the open-office plan. Probably the biggest practitioner of open-office plans are startups in the Bay Area — Google and Yahoo! It’s much less expensive. It’s much more social. You can fit more people than you can in a cubicle setup or especially in a closed office with the door kind of setup.

But there is some research that indicates worker productivity in open-office plans can be diminished by as much as 20 percent — owing to acoustic leakage, lack of concentration, increased desire to socialize with the person next to you. It’s difficult to concentrate with hearing everyone else talking all of the time. The most effective thing that can be done is for workers in open-office plans to put on headphones or to put in earplugs. Earplugs that cancel and that diminish the amplitude of sound after about 35 decibels are very inexpensive, maybe a dollar a pair.

Would you recommend that as a remedy for the open-air meeting environment, as well?

That would be the least expensive way to fix both of the problems. The other option [goes back to] gobos. And it would be much better to have [furniture with] material like fiberglass or wool; something that absorbs sound. It’s going to help a lot.

Is part of the problem that many adults — young and old alike — are experiencing hearing loss?

[Hearing loss] is definitely going to become an increasing problem. Just walking around in Miami’s restaurants and nightclubs, [I’ve heard how] federal noise guidelines are, on a daily basis, being trampled on. The Occupational Safety and Health Administration, OSHA, has a set of noise guidelines. If you just Google “OSHA noise exposure,” they dictate what is a safe decibel level for a human being to listen to and for how long. For example, you can listen to 85 decibels, which is pretty loud. That’s about as loud as you would want your stereo in your home to play. But if you listen to that for eight hours without interruption, it will cause damage. I was at a friend’s birthday party at a nightclub recently in South Beach. I was measuring the decibel level at a continuous 105 decibels, which after about one or two hours caused irreversible, permanent hearing damage in the 30- and 20-year-olds that were in there dancing.

I guess the point is if that’s going to become a fact of our population — that we are routinely damaging our hearing as a species — then I think it’s going to make the need for careful placement of microphones and loudspeakers for amplification of the human voice that much more important.

Listen Up

That open-air room set is murder on your attendees’ concentration. Your keynoter’s voice is getting lost in the cavernous depths of the theater space. And the music at your cocktail reception is way too loud. But don’t worry — good sound design is as easy as listening to the audiology experts, meeting organizers, and AV professionals we’ve assembled.

• The Complete Guide to Audio at Meetings
• How to Create Good Sound Design in Open-Air Spaces
• Can Attendees Improve Acoustics at Meetings?
• Commerce + Creativity Does not Equal Cacophony
• Scene and Heard: 10 Tips for Getting the Best Sound Mix at Events