Study Hall

Acoustic Essentials for Architects – Part V

A presentation of acoustical terminology and concepts relating directly to the design and construction of an architectural space. Low-tech descriptions, explanations, and examples.

This is an ongoing series guiding architects, and others, through the world of acoustics. In Parts I, II, III & IV we’ve defined architectural acoustics; explained what sound is; highlighted the three main types of sound propagation; offered a baseline acoustics lexicon; described echo, reverberation and resonance in some detail; provided an overview of the three acoustical tools; answered the question: Where does all the unused sound go; described and explained the good and bad influences of room geometry; and introduced the concept of acoustic shadowing.

This installment offers five short sections to help broaden your basic understanding of acoustics: Section 10 delves a little deeper into the importance of speech intelligibility. Section 11 outlines three simple, non-technical acoustical tests you may find useful; Section 12 lays out the differences between internal and external noise; Section 13 defines and characterizes the various means and methods of variable acoustics; and Section 14 glances into psychoacoustics.

This commentary is based on fundamentals and the author’s experience – not perfection. It covers many of the basics, while also exploring several modern and esoteric matters. You’ll be introduced to interesting and analytical subjects; some you may know, some you may never have considered.

If you’ve just joined us you might wonder: Why offer this material to the pro-audio community when the target audience is architects? The simple answers: Most pro audio practitioners have to work in acoustic environments that are out of their control because of past decisions made by an architect or owner. This series should help explain what and why those rooms are behaving the way they do. When complete, the goal is to transform this lengthy commentary into a reference book that architects can easily access.

Also consider that there are probably many new and veteran sound techs that have yet to accumulate a solid acoustical foundation. Most certainly, there are knowledge gaps on both sides of the designer/user relationship.

Hopefully this information will be of interest to many, and that you’ll find value that can be shared with others, including the owners and architects you may interact with on a regular basis.

Because of its scope and length, this commentary is being serialized over the course of many episodes. It’s a good idea to start from the beginning as each new installment builds on its predecessors. For example: The Section, reference, figure numbers, and acronyms sited in each publication continue from the previous issues.


10.1 When it comes to the good and bad of acoustics and sound reinforcement, few subjects receive more attention than speech intelligibility. Simply put, the phrase “speech intelligibility” references the average* person’s ability to understand what is being said when another person is speaking their native language. (* Requires average hearing and intelligence, and the absence of an unfamiliar dialect or thick accent.)

10.2 Loudness vs. clarity: When talking to new clients, one of the most common complaints we hear is: “Our people can’t hear the pastor (or presenter) who is preaching (or teaching).” Nine times out of ten, they can “hear”, but they can’t “understand” what’s being said. This is a perfect illustration of loudness versus clarity. Good speech intelligibility and good clarity are pretty much one in the same, and neither requires excessive volume or loudness.

10.3 The English language is almost entirely dependent on the quality and clarity of the consonants communicated within each spoken word; not the vowels. Here’s a written example. See if you can figure out this phrase using only the vowels. “_ _e  _e_ _ _  _ea  _ _e_ _ _  _y  _ _e  _ea  _ _o _ e, _ _i _ e  _ _e  _ai_  i_  _ _ai _  _o_ _i_ue_.” (See 10.4.B below for the complete phrase.)

A. This next sentence contains only consonants. “Th_ q_ _ck  br_wn  f_x  j_mps  _v_r  th_  l_zy  d_g.” With a little effort, most people can figure out most or all the words in the second sentence. (See 10.4.C below for the complete phrase.)

B. When consonants are not clearly received, words are missed or confused with other words. Sometimes this can be a significant problem. Examples of poor speech intelligibility arise quickly when two or more consonants rhyme. We’ve all experienced this at one time or another. The letters P, B, T, and D are easily confused, and M and N often get crossed up, as do F and S.

10.4 STI and %ALcons

A. Speech clarity and intelligibility are so important that real, objective tests have been devised and established as industry standards. Those standards are called the Speech Transmission Index, and the Articulation Loss of Consonants. STI and %ALcons are covered more specifically in Part II, Section 4.7.

B. From Section 10.3 above: “She sells sea shells by the sea shore, while the rain in Spain continues.” Doesn’t make any sense you say? Wasn’t meant to.

C. From Section 10.3.A above: “The quick brown fox jumps over the lazy dog.” How did you do?

10.5 “Speech intelligibility is directly impacted by the level of background noise, the reverberation time, and the size and shape of a room.” [34]


11.1 There are three simple tests that can be used to quickly judge the most basic acoustical merits of a room. These tests can’t and shouldn’t replace objective testing that’s done by a professional acoustician, but they will give anyone interested a rough idea of the acoustic conditions of an existing room. These are called: “The Two Talker”, “The Hand Clap”, and “The Room Mode Analyzer” tests.

11.2 The Two Talker test: This is used to get an initial impression of these acoustic qualities: speech intelligibility and clarity; room noise; critical distance (the point at which you hear equal amounts of both direct and reverberant sound); and, potentially, the need for a sound reinforcement system.

A. The procedure goes like this: Two people face each other, spaced about three feet apart. The beginning location is not overly significant. For this example let’s start the process near the front edge of the stage, platform, or lectern location in the room.

B. The person standing closest to the stage (person P1) faces out toward the audience area. The other person (P2) should face P1. P1 begins by speaking a short list of five to 10 random words, each only having one or two syllables. Pick words that contain the consonants P, B, D, T, M, N, F and S. Speak at a normal, conversational speaking volume, and never alter your speech volume, enunciation or cadence.

C. Next, P2 repeats the words back to P1. Presuming P1 receives the first list back from P2 without any errors, P2 should then move away from P1 by about four paces, or about 10’. P1 always stays in his/her original position.

D. Next, P2 should deliver a new set of random words to P1, with P1 repeating the new list back to P2. If P2 receives the word list back without error, P2 moves away from P1 by four more paces.

E. Continue this process of exchanging words, with increased spacing, until either P1 or P2 begins to struggle to hear or understand the other person, or makes obvious errors in the words they send back. If about 1 in 5, or 2 in 10 of the words are received inaccurately, you’ve reached the point of “Poor to Bad” intelligibility.

F. At this point stop the test. There’s a good chance you’ve reached a virtual boundary [Figure 59] we call “critical distance”. This means you’re hearing an equal amount of direct and reverberant energy when the other person is talking. Critical distance is further defined in Section 11.2.H below. There’s no point continuing further. Make note of how far apart you’re now standing. Also make note of how much of the room is still unevaluated.

    1. If P2 makes it all the way to the furthest seat in the room, without a noticeable loss of intelligibility, it means the environment is free of excess reverberation or noise.
Figure 59: Notice the gradual change in density between the red and blue dots as the sound wave moves away from the loudspeaker. Also, be aware that the critical distance changes somewhat, based on frequency. Graphic courtesy of Lenard Audio Institute. [35]

G. Presuming you’ve not reached critical distance, you may find a point at which the other person’s voice is just too soft to be easily and clearly heard. If this happens, it’s an indication that a sound reinforcement system may be needed to amplify or reinforce the spoken word.

    1. If the room is noisy, you may reach a point where you can’t understand the other person because the background noise is equal to, or louder than their voice.
    2. If the room is not too reverberant or noisy, you are probably evaluating a fairly small room, with reasonably good acoustics.

H. “Critical Distance is the distance from the sound source, where the direct and reverberant sound energies become equal. The more reverberant a room is, the closer the point of Critical Distance is to the sound source [human or loudspeakers]. The more absorbent a room is, the further the Critical Distance is from the sound source. Critical Distance is different at all frequencies. For good acoustic design the Critical Distance should be as far as possible from the sound source, and the resultant reverberation minimal and even at all frequencies. Direct sound from the loudspeaker system diminishes in level as a function of distance (inverse square law), whereas reverberation constantly spreads throughout the room. Because there is new incoming sound from the loudspeakers, reverberation keeps building up until the new incoming sound equals the sound absorbed (steady-state). When the reverberant sound becomes 12dB or greater than the direct sound, all intelligibility is lost.” [35]

11.3 The Hand Clap test: This activity can be done solo. It is used to evaluate the presence of problematic echoes, and provide a rough idea of a room’s reverberation time (Tmid).

A. The procedure for this test is no more difficult than having the ability to produce a single, loud, hand clap. Slapping the fingers only (no palm) of one hand against the palm of the other hand works best for me. Figure 60

Figure 60: Maybe you’ve noticed? When it comes to applauding a performance, it seems that everyone has their own way of clapping. Clapping your hands for maximum energy and pop requires a little bit of technique and practice. Try using fingers only of one hand to strike the palm only of the other. This, and using about a 45 degree rotational offset, seems to work best for most people. Photo credit: Thinkstock/Imagebank

B. Walk around the venue and occasionally stop and clap your hands together one time. Then just listen. In almost all locations, you’ll likely hear either reverberation, echo(es), or both.

    1. If you hear a smooth, obvious, reverberant tail (decay) after the hand clap, without any obvious slap or flutter echo, try to silently count to yourself the number of seconds it takes for the reverb tail to become inaudible. Count up from zero, as in “zero Mississippi (at the point of hand clap impact), one Mississippi, two Mississippi,” etc. If you get to more than about 1.5 “Mississippis”, this is an indication the room probably has too much reverberation for good to excellent speech clarity and intelligibility. Move around the room and repeat as needed. The goal is to get a rough idea of the rooms Tmid, which is what most people are describing when they say a room has some number of seconds of reverberation.
    2. It’s fairly common to hear both echo and reverberation. When you easily hear both, it usually indicates that the Tmid is under about 2.0 seconds, but greater than about 1.0 second. If this is the case, treatment specifications should prioritize minimizing the echoes.
    3. It’s also common for reverb to completely mask echoes – flutter or slap. When a room is treated to reduce reverberation, that treatment may “uncover” residual echoes. Keep an eye on the parallel walls when considering mounting locations for treatment. If possible, absorptive or diffusive materials should be placed on at least one of each pair of parallel walls/surfaces.
    4. If you hear echo, with little or no reverb, look around for any large, hard, parallel surfaces you may be standing between. These echoes are symptomatic of acoustical problems that will, at the very least, reduce clarity and intelligibility.

11.4 The Room Mode Analyzer test: This exercise requires a sound system capable of easily reproducing a very low frequency, such as 60 Hz, and a sine wave tone generator. It’s used to uncloak and reveal a room’s modal characteristics.

A. This test can be very easy if you have access to a venue’s sound system, and a sound tech to help with the implementation. Ask to have a 60Hz sine wave tone run through the system. It doesn’t have to be loud, though the louder the tone, the more dramatic the experience.

B. Slowly walk around the room and listen. This test is so interesting and powerful the outcome is hard to describe. You just have to experience it at least once in your life. See Part II, Section 5.6.C for my metaphorical attempt at describing what you’ll hear.

    1. Next, ask to have the tone generator changed to another low frequency, such as 100 Hz. Again, walk around and listen to what happens. Every frequency will have a different modal pattern [Figure 61] of peaks and valleys. Geographically, this translates to peaks and valleys that change locations every time the frequency is changed.
    2. Next, find one location and stand or sit still while the sound tech slowly changes the frequency – up or down – in 10 Hz steps. If 10 Hz resolution isn’t available, use whatever frequencies are available between 40 Hz and 315 Hz. At the location you’ve chosen, each frequency change will bring a different result (change in volume). Now move to another location only two or three paces away and listen again to the same sequence of frequencies. In most rooms, the second location will vary greatly from the first. The loudest frequencies in the first may be the softest in the second. Any combination of variables in frequency and location will produce different results. And, don’t be shocked when you experience this: What’s too loud in one location will become near silence a few feet away. Pure audio magic!
    3. The experience will be similar in most rooms, though in some it will be more dramatic; others less so based on room geometry and applied acoustic treatments. In a really good room, the variations in what you hear will be more subtle, which is our goal. Sadly, perfect rooms don’t actually exist.
Figure 61: While only showing a small room, this graphic does a good job of illustrating the dramatic changes in level (volume) for a 63 Hz tone. The lime green bands represent the approximate median level, while the range, from loudest (red) to softest (blue) is 30 dB. By moving only a few feet – in any direction – a listener could easily claim they’re hearing way too much, or too little VLF content. Graphic courtesy of the Olive Tree Labs Suite @

C. The underlying causes of this unique listening adventure are the modal nodes and anti-nodes [36] that are resonating throughout the room, and changing based on the stimulus frequency; your position in the room; the room’s dimensions and ratios; the architectural finishes; and the acoustic treatments applied. (See Part II, Section 5.5 for more on room resonance and modes.)


12.1 The methods and materials used to minimize or contain noise are generally divided into two classes, based on the need to control or mitigate either internal or external noise. To review, noise is sound energy that is unwanted, uncomplimentary and/or distracting to the intended/desired audio content you’re trying to hear.

A. Noises can either be short or long in duration. Regardless of length, it’s still noise.

B. A room’s ambient noise floor, and its speech intelligibility scores, go hand in hand.

12.2 Signal-to-noise (S/N) ratio: As the name implies, this is the ratio of direct, desired program material (signal) to the total, aggregate, ambient noise package (noise).

A. The S/N ratio is expressed using dB as the unit of measurement. If the program material averages 85 dB SPL, and the ambient noise measures 65 dB, then it is said that the S/N ratio is 20 dB.

B. General industry guidelines suggest that a 10 dB S/N ratio is the absolute minimum allowable for any speech comprehension. This is not meant to imply that 10 dB is remotely acceptable, but rather, it has been shown to be the minimum value that would result in some minimal speech intelligibility, under the most extreme emergency conditions.

C. When possible, most acousticians and audio system designers are looking for a minimum S/N ratio of 50 dB, or greater.

D. Coherent early reflections and moderate reverberation are not noise.

12.3 When reverberation and echo become noise:

A. Given a venue that’s acoustically well matched for the program material being presented, appropriate amounts of reverberation are coherent and complementary to the desired program sounds. In excess, reverberation can distract from the quality and clarity of the direct sounds, thereby becoming much like any other ambient noise.

B. When other background noise(s) is introduced into a highly reverberant space, the signal-to-noise ratio becomes even worse.

    1. All noises mix together, regardless of their source, tonality, quantity, or location of origin. When combined, and sustained by excess reverberation, these noises can create a din of random sound, making even the most articulate performance indistinct.

C. Unintended echo, be it flutter or slap, is another form of noise.

12.4 Examples [Figure 62] of common structural, mechanical, and environmental noises:

A. People talking, coughing, sneezing, clapping, crying and/or being generally restless

B. A noisy sound system, electronic instruments, or in-room production equipment

C. Water features, fountains, and pumps. Especially water that splashes

D. Excessive reverberation, slap, and flutter echoes

E. HVAC hum, vibration, and wind-velocity noise

F. Electrical transformer hum, and lighting buzzes and hums

G. Elevators

H. Outside road, rail, airplane, or other environmental noise

I. Neighboring or adjoining rooms – above, below, and on all sides – that transmit loud or noisy activities

Figure 62: This graphic does a good job outlining the most common external noise pathways. The VB and PB absorption treatments described in Part III, Section 6.3.D are ineffective tools when damping or blocking noise transmission is needed. Graphic courtesy of

12.5 Internal Noise:

A. Noise that’s created, and stays within the envelope of a given room, is internal noise. The first four items listed in Section 12.4 above are examples of internal noise.

    1. Internal noises can and should be removed or reduced at their source.

B. Unwanted noise(s), which cannot be mitigated at the source, can be minimized somewhat, using absorptive VB (velocity based) treatments on the walls, floor, ceiling, etc.

    1. Once noisy sound energy appears in a room, only absorptive or damping treatments are effective for any degree of mitigation.

12.6 External Noise:

A. External noise, which is transmitted into a room, is just as problematic as internally generated noise. The last five items listed in Section 12.4 above are examples of external noise.

B. Depending on the type of noise, and its location of origin, blocking it from entering a building can often be more challenging and expensive than any of the internal acoustical treatments mentioned earlier.

C. The most effective way to stop noise from getting into a room is to identify the potential problem before any construction begins. At this point, an acoustical consultant can evaluate the potential problems, and specify the appropriate construction materials and methods needed to block all, or most of the unwanted sound.

    1. If you are working with an existing structure, blocking external noise can be quite complicated. Because they are misapplied, many of the most common and “obvious” VB and PB (pressure based) acoustical treatments are ineffective, and a waste of money.
    2. Typically, blocking out external noise requires frequency-dependent vibration isolation or pathway blocking, not sound absorption.

12.7 Sound Transmission Class (STC):

A. Many construction materials have an STC rating. The STC rating defines how effective the material is at blocking various speech-range frequencies (125 Hz – 4 kHz) that might try to pass through it, and the mounting method required to achieve those specific results.

B. Material STC ratings are not generally relevant to the management of internal room acoustics. They are, however, very relevant to controlling sound that tries to get in or out of a specific room.

C. STC is covered in more detail in Part II, Section 4.8.


13.1 The phrase variable acoustics (VA) means the acoustical characteristics of a room can be changed to better match the requirements of a performance or presentation. This is a very useful feature.

13.2 Historically, the acoustics of most rooms have been static and unchangeable, other than when people occupy the space in varying quantities. Beginning with Wallace Clement Sabine, in the early 20th century, acousticians began studying the qualities and properties of various finish materials, and how they affected a venue’s acoustics. It didn’t take long before people started trying to manipulate and control the reverberation time in rooms.

13.3 VA is a topic that’s most often considered for implementation when designing a performing arts venue, though many other facilities are candidates too. Concert halls, houses of worship, and multi-purpose ballrooms are examples.

A. More often than not, the goal in these performance venues is to build and maintain a space that can support the widest possible range of events and activities. In almost all cases, venues that can effectively accommodate a diverse genre of audio and musical styles are the ones booking the most calendar dates; ultimately making them the most commercially-viable and profitable.

13.4 Over the past 50 years or so, concert halls and other performance venues have been installing mechanical, VA systems. The most common applications and techniques included:

A. Drapes, which are manually opened or closed to expose or cover various reflective surfaces. The application is passively subtractive; providing variable percentages of VB absorption, without the use of any audio electronics.

    1. Using drapes alone, the reverberation times can never be greater than what the room offers when the drapes are fully retracted (opened). Conversely, the reverberation times can never be less than what the room offers with the drapes fully deployed (closed).
    2. More modern variations use motorized drape pulley systems, or other motorized [Figure 63] assemblies.

B. The addition of purpose-built, auxiliary, reverberation chambers.

    1. These chambers are empty, highly-reflective annexes that share a common moveable, or shuttered, wall with the main venue.
    2. The shutters are manually opened or closed to introduce more or less volume into the venue’s boundaries. The application is passively additive, as the added volume is intended to garner longer reverberation times, without the use of any audio electronics. This technique has fallen out of favor, as the intended results tend to be minimally-effective, and are usually cost and/or space prohibitive.

C. To offer a greater range of adjustment, some venues have employed both drapes and auxiliary reverberation chambers.

D. While drapes and auxiliary chambers are seen as useful advancements over most static, acoustical options, they are passive solutions that have several notable drawbacks:

    1. Cost of construction
    2. Loss of usable space
    3. A small range of variability
    4. Slow and/or labor-intensive to adjust
    5. Somewhat difficult to repeat exact settings and conditions
    6. Neither of these solutions offer the ability to optimize the T60 Slope Ratio (TSR) characteristics of the room. See Part II, Section 5.4.C.2 for a refresher on the TSR thesis.

E. With these shortcomings, it’s a wonder any owner would commit to the cost of this “new” technology. However, the need and demand for such acoustic flexibility was clear, and justifiable, when weighed against owning an unused venue; having only a single, static, acoustical trait.

Figure 63: The evoke 1 VA system. This is a modern version of the passively subtractive draping technique noted above. These slotted wooden panels are motorized to open and close as needed. When closed, they present a fully reflective surface, with a very small amount of horizontal scattering. When open. sound energy passes through the slots and reaches VB absorptive material within. To vary the absorption coefficients of each bank of modules, the percent of openness can be precisely set and recalled. Another significant advantage is that rapid variability, with positional uniformity and consistency, is now possible. Photo courtesy of FlexAcoustics [37]

13.5 One of the most interesting technologies available today is described as “electronic variable acoustics” (EVA). Consider these points:

A. Apart from acoustics, all aspects of the technical systems – i.e. sound, video, and lighting – in performing arts centers, concert halls, houses of worship, or other entertainment venues, are adjustable via real-time, human interaction. Now, we have access to tools and technologies that bring architectural acoustics under our control too.

B. In recent years the high-level implementation of digital audio signal processing has allowed manufacturers to explore the management and manipulation of room acoustics. This is done using powerful micro-processors, elaborate software applications, and secondary microphones and loudspeaker systems.

    1. To be clear, these are additive systems, meaning they add full, natural-sounding reverberation into a room to enhance the acoustical support necessary for purely acoustic instruments and voices. They do not and cannot remove any reverberation that naturally occurs in a room. To take full advantage of such systems, a room must start with a Tmid no greater than one second. And ideally, an “Optimal” TSR grade.

C. Now, using presets, changes can be made quickly, easily, consistently, and with a much wider range of Tmid and TSR parameters, than ever before imagined. This technology has sufficiently matured to the point where the best EVA solutions are now a viable option, with few if any, discernible, unnatural artifacts.

D. Cost will still be an issue for some, as the leading brands can be quite expensive to properly implement. Contact your favorite acoustician for more info and budgetary estimates.

13.6 To help illustrate why VA and EVA are gradually gaining greater traction, consider these wide-ranging types and styles of music, which are best experienced when performed in a room with an optimal reverberant profile. Examples include:

A. Spoken word – Short reverb, at or below a 1.00 second Tmid.

B. String or wind ensemble – Medium-long reverb time, between 1.50 and 2.00 second Tmid.

C. Unamplified jazz ensemble – Medium length reverb, in the range of 1.25 to 1.50 second Tmid.

D. Rock, pop, blues, funk, country, hip-hop, soul, contemporary Christian, or other up tempo, rhythmic music – Medium-short reverb, between 1.00 and 1.25 second Tmid, combined with a Good to Optimal TSR grade.

E. Large choir, pipe organ, and/or orchestra – Medium-long, to long reverb, between 1.75 to 4.00 second Tmid.

F. Dramatic acting and musicals – Medium length, between 1.25 and 1.50 second Tmid.

G. Without the implementation of an EVA system there are very few, if any, venues that can support anything close to this diverse range of acoustical requirements.

13.7 As we approach the second quarter of the 21st century, owners, architects, and acousticians are implementing a radically more ambitious approach to VA, and architectural geometric expression: “shape-shifting” room volumes. The referenced animated clip shows how the Lindemann Performing Arts Center, at Brown University, uses movable walls, seating, flooring, and ceiling reflectors to suit their various application and acoustic needs. [38] [39]

A. Initial testing of the Lindemann PAC shows the average range of Tmid times, for five of the many possible configurations, to be between 1.33 and 2.10 seconds.

B. The Perelman Performing Arts Center, at the World Trade Center in New York, is another recent example of this new paradigm shift in AA. [39]

C. Remember, room volume is a key component in determining the average reverberation times in any room.

13.8 Here are some basic VA and EVA takeaways for your consideration:

A. Old -school mechanical VA systems are expensive to integrate, labor intensive to operate, and provide only minimally-effective variation.

B. Modern EVA systems are also fairly expensive to integrate. Nevertheless, they provide a much greater range of adjustments, and can be quickly and easily modified by recalling one of many presets. Presets are established during the initial EVA system commissioning.

C. Shape-shifting VA designs are also very expensive to create, and require significant time and manpower to adjust.

D. There are no electronic products/solutions on the market that can reduce or eliminate excess reverberation. Therefore, to properly implement an EVA system, you must start with a room that has a very short Tmid, i.e. one second or less. And, to fully optimize for all or most of the musical styles listed above, a Good to Optimal TSR grade should also be a high priority.

    1. Starting with a Tmid that’s less than 1.00 second is okay too. Longer is problematic because, if needed, you can never get back to a shorter Tmid than the venue naturally presents with the EAV system turned off.
    2. If an existing room has too much reverberant energy (read: reverberation or modal tails that last too long) it will first require the addition of the appropriate amounts of absorptive and/or diffusive materials – to achieve the desired Tmid baseline – before an EVA system can be considered and effectively deployed.

14.1 No commentary on architectural acoustics (AA) would be complete without touching on the subject of psychoacoustics. The human brain is such an amazing computational tool that it’s easy to overlook its role in audio perception.

14.2 “Psychoacoustics is the branch of psychophysics involving the scientific study of sound perception and audiology – how the human auditory system perceives various sounds. More specifically, it is the branch of science studying the psychological responses associated with sound (including noise, speech, and music). Psychoacoustics is an interdisciplinary field including psychology, acoustics, electronic engineering, physics, biology, physiology, and computer science.” [40]

A. Though this subject has been extensively researched and documented, only a few core concepts will be expressed below.

14.3 Time is one of the most important elements related to the quality of sound perception. As highlighted in Sections 2.5.B and 4.10, the speed of sound is extremely slow. So slow in fact that when listening to identical audio content, the average listener can hear timing differences, between distant and nearby loudspeakers, if they are offset by as little as 34 feet. In some cases even less. As a result, time plays a critical role in the discussion of how sound is perceived.

A. Remember, it takes 30 ms for sound to travel about 34′. That’s thirty, one-thousandths of a second. Beyond roughly 30 ms the brain gradually loses its ability to integrate the two loudspeakers, or sound sources, as one; thus we begin perceiving them as two discrete signals. If not electronically-corrected in the time domain, such asynchronous signals can easily reduce quality, tonality, intelligibility and clarity.

B. When the same two loudspeakers are moved from a front-to-back orientation, to a side-by-side orientation, a completely different set of psychoacoustic phenomena occurs; the phantom center.

14.4 Phantom Center

A. The concept of a phantom center loudspeaker [Figure 64] is classic psychoacoustics in action. Most people have experienced this without knowing or understanding that it’s happening. The experience: Set up (or visualize) a stereo listing environment, configured as an equilateral triangle created by two speakers and your head, each separated by six feet or so. Each speaker should be facing you and positioned at the same height as your ears.

Figure 64: The white speaker in the center doesn’t actually exist. But, when a monaural audio signal propagates equally from both the left and right speakers, it will appear to be coming out of a single speaker in the middle. Graphic courtesy of (41)

B. Next, play a monaural audio signal (like a person talking) that propagates equally from the two speakers. What you should notice is the talker appears to be positioned directly between the two speakers. The voice won’t sound like it’s coming from either of the individual speakers, but rather from a point directly in the middle.

    1. Headphones and earbuds can also be used, but the phantom center image moves from being in front of you, to feeling like it’s centered inside your head.
    2. Simply changing one or more of the following items on the left or right side: signal, volume, timing, and/or polarity, will create a significantly different listening experience. For example, if the polarity is reversed on either of the two channels, the phantom center sound will completely disappear.
    3. Because everything is equal, including the time arrival, your brain doesn’t distinguish the sound as coming from two discrete locations. It integrates the two sounds into what seems to be a single phantom speaker, located between the two actual speakers.
    4. It doesn’t matter if your eyes are open or closed, the phantom center does not move or change. Only if you move your head slightly to one side or the other will you begin to hear the sound move with you. Move a little right and the phantom center moves a little right too. Move too much, in either direction, and you may think that only the nearby speaker is on.
    5. The key to this phenomenon is time. As you move, it’s the minute arrival time differences that make the sound appear to move with you. This is one example of psychoacoustic at work.

C. Time-related volume changes – Take the exact same setup described above and add a very small amount of signal delay to one of the speakers. Let’s use 10 ms, which is ten, one-thousandths of a second. You won’t be able to do this at home, but it’s easily done with a professional sound system.

    1. During this experiment, the signal will appear to be coming from the speaker that is not delayed, and the non-delayed speaker will sound slightly louder, even though the volume of each speaker remains exactly the same. This is yet another example of time-domain psychoacoustics in action.

D. There are other examples that could be used to describe the role of psychoacoustics in AA and sound reinforcement, but these should suffice for now.

14.5 How do psychoacoustics and AA fit together?

A. In a theoretically-perfect venue all listeners would sit in exactly the same optimal seat, and be presented with a single, four-dimensional “sound stage” that represents all instruments, voices, loudspeakers, and complementary acoustical artifacts. There would be nothing to interfere with the sound as it travels through the air. All sound would arrive with perfectly-coherent timing, tonality, and fidelity.

    1. Being the conductor of a symphony orchestra is probably the closest anyone comes to experiencing this ideal scenario.

B. Unfortunately, this ideal listening experience is never going to happen for you, as a paying customer, in any venue.

    1. About the closest we can get to an ideal venue, and listening experience, is an unamplified music or vocal performance, presented in a modest-sized structure, with optimal acoustics. In this setting, it’s imperative that each instrument and/or voice has a direct, line-of-sight pathway to each individual listener.

a. There would be no perceptible “time smear” because there is a direct, one-to-one time relationship between each instrument or voice, and each listener.

b. Presuming you have a seat that is somewhat forward of the critical distance zone mentioned above, everything other than the direct sound would be secondary and complimentary; either integrated early reflections, or reverberation.

C. Once microphones, audio mixers, signal processors, and loudspeakers are brought into the equation, two or more time-dependent pathways are created between the individual instruments and/or voices, and each listener. As a result, there are many more time-domain variables, and potential psychoacoustic distortions. If left uncorrected, these cumulative timing errors translate into uneven tonality and sound coverage, which further diminishes quality, clarity, and intelligibility.

D. Reasonable solutions become possible when architects work with purpose to distribute chairs, or seating areas, evenly and symmetrically throughout a venue.

E. Audio system designers must also be allowed to place loudspeakers in very specific locations so they can present everyone with the best possible blending of direct and amplified sound; all while attempting to overcome the many timing, acoustical, structural, spatial, tonal, budgetary, and aesthetic challenges that are ever present.

    1. Today’s systems designers have many tools to minimize or eliminate the majority of these time-domain distortions, providing a room is built with good symmetry and acoustics. When asymmetrical room layouts come into play they complicate and/or exaggerate the timing variables, in some cases beyond what can be compensated for using today’s sophisticated digital technology.
Final Thoughts

That’s it for this installment. Hopefully, you’ve learned a few things you didn’t already know, and are looking forward to what’s upcoming in the final installment. Part VI has six Sections that should further broaden your basic understanding of applied acoustics: Section 15 provides insights related to pragmatism, opportunities, and compromises; Section 16 lists and shows some modern acoustic materials that may not be well know; Section 17 elaborates on new reverberation design goals for the 21st century; Section 18 introduces PMAT – the Parametric Method of Acoustic Treatment; Section 19 outlines this commentators list of acoustic priorities; and to wrap things up, Section 20 summarizes Applied Acoustics – the three feasibility questions. [42]

One of the essential goals of this commentary is to encourage architects to put as much time, effort, care, and awareness into optimizing the acoustics in rooms they design as they put into satisfying the customer’s needs for proper lighting, seating, heating, and air conditioning.

If you remember nothing else from this series, please remain focused on this: For better or worse, every architectural shape, feature, location, and finish influences sound behavior and quality.







[39] AVA consultant: Threshold Acoustics




Peer Review Team

A special thank you goes out to Neil Shade, FASA (Acoustical Design Collaborative), Charlie Hughes (Biamp and Excelsior Audio Labs), and Rob Miller (Threshold Acoustics) for assisting with notes, comments, and corrections that made this work presentable.

Study Hall Top Stories