Meeting Topic: Concert Audience Engagement - The Perception of Closeness
Moderator Name: David Weinberg
Speaker Name: Dr. David Griesinger, Acoustic Consultant
Other business or activities at the meeting: Election of new Treasurer - Brett Takacs
Meeting Location: National Public Radio Headquarters, Washington DC
Engage Us With Sound
by Fred Geil (AES-DC Secretary) and David J. Weinberg (AES-DC Chair)
On 21 April 2010, about 30 members and guests gathered at NPR headquarters for a technical presentation by David Griesinger on Concert Audience Engagement — The Perception of Closeness.
Weinberg introduced Dr. Griesinger (AES Fellow and former Lexicon Principal Scientist), who has spent decades researching the correlation between how we hear and our preferences in concert hall acoustics. His work has led to surround algorithms in Lexicon and Harman professional and consumer surround and reverberation processors. The objective of his research is to help architects and acoustic consultants design concert halls and opera houses that acoustically add excitement to the listening experience.
Engagement is a subset of the overall subject: The Relationship Between Audience Engagement and Our Ability to Perceive the Pitch, Timbre, Azimuth and Envelopment of Multiple Sources. His set of slides is at www.DavidGriesinger.com.
Engagement, a term conferred on the phenomenon by Asbjørn Krokstad, relates to the sound of a performance that stimulates focused attention, as if the performance is just for you; sounds perceived as close to the listener and more accurately situated are more emotionally involving than those that sound far away — certainly inherited from early-human's survival reaction. In addition, Griesinger has found that "envelopment — a sense of the hall -- is enhanced when it is possible to precisely localize instruments."
His talk comprised the biophysics of how we hear, the psychology of aural engagement, and concert hall acoustics — good and bad. These three topics helped verify several of his "radical concepts."
The critical issue is early reflections: the amount, the initial time delay, and the frequency content relative to the direct sound:
• Reflections in the time range of 10-100ms reduce clarity, engagement, and envelopment.
• Reflections off the back wall of a stage or shell decrease clarity because their level is relatively high and they arrive from the same direction as the direct sound.
• The direct-to-reverberant ratio above 700Hz (D/R700+), if above a critical threshold, can enhance the listening experience.
• Side-wall reflections are desirable in the front of a hall, but reduce engagement in the rear seats, which lack enough of the direct sound for an adequate D/R ratio.
• Reflections above 700Hz directed toward the audience close to the sound sources will reduce the reflected energy to listeners seated farther back, thereby improving their D/R700+ ratio. Coffers, niches, and/or open ceiling reflectors help, when they are sufficiently large and deep.
Griesinger discussed the physics of hearing — how we are able to hear with astounding resolution not explainable by the traditional concept of critical bands, the basilar membrane's limited frequency resolution, and the ~1kHz maximum rate of auditory-nerve pulses. He presented a model of the hearing mechanism: "a possible neural network that detects phase information in harmonics over the vocal formant range [1000-4000Hz]. These harmonics carry more information about the sound source than the fundamentals. The neural network is capable of detecting the fundamental frequency of each source to high accuracy (1%) and then separating signals from each source into independent neural streams that can be separately analyzed for pitch, timbre, azimuth, and distance." This mechanism explains our ability to detect the pitch, timbre, azimuth and distance of several sources simultaneously, all of which depend on harmonic coherence, and degrade in the same way with an increase in reverberation. With too many early reflections and too much early reflection energy, the source is perceived as distant; thus engagement is reduced as well as the ability to localize several sources.
Griesinger described "a simple mathematical equation that predicts the threshold for detecting the localization (azimuth) of a sound source in a reverberant field," which he claims is a good predictor of direct sound perception. The equation enables calculation of the D/R ratio based on the hall's impulse response. He has experimental data that validates his equation.
Another important property of sound in a hall is envelopment, engendered by reverberation. It is important that the direct sound is separately perceived so that the brain can sense them as two distinct but related sound streams. This initial time delay between the direct and reverberant sound (quantified as when the buildup — not the decay -- of reverberation reaches the level of the direct sound at that location) is key to achieving good sound in a hall, and also explains the aurally perceived difference between Boston Symphony Hall and the Amsterdam Concertgebouw, two venues that otherwise exhibit similar acoustic parametric values. Griesinger showed measurements taken in each venue with binaural microphones that pick up sound at the eardrums.
He tied all this together as he discussed the concept of engagement in terms of the concert hall experience. The goal is to perceive the sound sources as close, yet with enveloping reverberation. This is achieved in a few halls, and not achieved in many others. He used sound clips recorded at various locations in various halls to illustrate strong and weak engagement.
One validation of Griesinger's concept is that generally preferred symphonic and operatic recordings are often made with a combination of relatively close-miked direct sound mixed with distant-miked hall sound — generating the mix of engagement with envelopment he seeks in concert halls.
Griesinger summarized what he believes is required for good concert-hall sound:
• There must be the ability to hear the direct sound over as much of the hall as possible. When the D/R700+ ratio is less than 0dB in the first 100ms after the note begins the direct sound is inaudible. To achieve engagement the direct sound needs to be stronger, the reverberation needs to be weaker, or the time delay between the direct sound and the reverberation buildup needs to be longer so that less of the reverberation falls within the 100ms window.
• Hall shape does not scale. If a great 2000-seat hall is scaled to 1000 seats, in a higher percentage of those seats the time gap is too small for engagement. If scaled to 500 seats, an even lower percentage of seats will be "good".
One characteristic of good halls is the use of diffusing elements such as the niches and alcoves in Boston Symphony Hall, which serve to direct much of the high-frequency early and late reflections to the audience near the front (where they are absorbed), so that the direct sound is better preserved for the rear seats (retaining a good D/R700+ ratio).
In small halls it is common practice to improve the sound by installing a shell behind the performers, or simply having a wall behind the players. Griesinger related an experience in Williams Hall, New England Conservatory, where a curtain was drawn across the rear of the stage, which eliminated the early reflections that previously had been compromising the direct sound and the D/R ratio.
Griesinger emphasized that to maximize good sound over a wide range of seats:
• Maximize engagement by ensuring that the brain can perceive direct sound as distinct from the reflected energy.
• Achieve envelopment by preserving direct sound in the presence of adequate reverberation.
• The audibility of direct sound depends on the D/R700+ ratio and the time delay of reflections in the first 100ms. The optimum D/R700+ ratio depends on the hall size; it should be higher in smaller halls. This is especially true for opera houses.
• Small halls need absorption or diffusion of early reflections more so than large halls.
• Current hall measurements ignore both the D/R700+ ratio and the time delay between direct sound and reverberation buildup. Hopefully the missing measurements can be calculated with his localizability equation and be useful in hall analysis and design.
Several audience questions were explored, and Griesinger exhibited his home-built binaural eardrum microphone apparatus.
Preceding the presentation was a short business meeting during which Brett Takacs (audio consultant at Miller, Beam & Paganelli) was elected by acclamation to be the new AES-DC Treasurer. Special thanks were extended to the NPR staff who set up the high-quality surround-audio and video equipment for the demonstrations and presentation.
