Sound masking is the addition of natural or artificial sound (such as white noise or pink noise) into an environment to cover up unwanted sound by using auditory masking. This is in contrast to the technique of active noise control. Sound masking reduces or eliminates awareness of pre-existing sounds in a given area and can make a work environment more comfortable, while creating speech privacy so workers can better concentrate and be more productive. Sound masking can also be used in the outdoors to restore a more natural ambient environment.
Sound masking can be explained by analogy with light. Imagine a dark room where someone is turning a flashlight on and off. The light is very obvious and distracting. Now imagine that the room lights are turned on. The flashlight is still being turned on and off, but is no longer noticeable because it has been "masked". Sound masking is a similar process of covering a distracting sound with a more soothing or less intrusive sound.
Sound masking can be used anywhere to ensure speech privacy or reduce distractions. Sound masking is typically used in selected workspaces but it can also be helpful in residential environments. The most common sound masking installations are:
Open office plans - open offices can be either too quiet (where someone dropping a pen in the next cubicle is distracting) - or too noisy (where the conversations of others in the office make it impossible to concentrate). Open offices can benefit from sound masking because the added sound covers existing sounds in the area - making workers less distracted and more productive.
Private offices - private offices and other enclosed spaces often appear to provide privacy but actually do not. Many times, walls are lightweight and do not extend to the ceiling deck - only to the ceiling tile. In these cases, sound can easily travel through partitions or over the walls. Sound masking can be provided in adjacent private offices, or in hallways outside of private offices, to ensure that confidential conversations remain confidential.
Public spaces - sound masking is useful for reception areas, pharmacies, waiting rooms, and financial institutions. Sound masking is provided in the area where conversations should not be heard - not necessarily in the area where the conversation is taking place. For instance, a psychiatrist does not want those in the waiting room to overhear a private conversation with a patient, so sound masking is provided in the waiting area: not in the psychiatrist's office.
Sound masking may also be used to hide other unwanted noise, such as the intermittent sounds from machinery. In an office this could be sound of elevators and compressors. Sound masking may render conversations unintelligible by nearby listeners and may thus help compliance with HIPAA and GLBA regulations.
Sound masking is being used to protect confidential privacy in areas where sensitive or classified conversations are being held. The applications, among others, are in government, military, military contractors, corporate board rooms, and legal offices. The requirements for this type of masking are more stringent. The sound must be guaranteed to be continuous during room use, performance must be verified, and the equipment must be able to protect windows, doors, walls, and ducts with vibration maskers instead of loudspeakers.
A number of cases exist where sound masking has been successfully installed for exterior applications, the most common target of concern being roadway noise. In one example application, a large artificial waterfall was constructed as part of the garden exterior of an urban hotel in Santa Rosa, California. The waterfall cascades down an extensive wall approximately four meters in height and functions both for sound masking and as a physical barrier to road noise.
Sound masking systems
The plenum is the space between a dropped ceiling and the upper deck for the floor. In-Plenum sound masking systems, which employ a network of loud speakers located completely within the plenum, were the first such systems developed they have been in use since the 1960s. Plenum-based speakers typically range from in diameter. The industry standard spacing of plenum speakers is or less on centers. Speakers in the plenum generally face upwards, towards the upper deck. This is done to reflect sound from the speakers to broaden, as much as possible, the footprint from the speaker in the work area.
The actual pattern of the received acoustic energy in the workspace from speakers in the plenum is complicated by a number of factors, all of which cause spatial variability in the sound masking field in the workspace. First, because loud speakers actually radiate in all directions, some energy at low frequencies is radiated downwards. Thus some sound variability occurs directly below the speakers. Second, dropped ceilings have a wide range of acoustical "transparency" or transmission loss (their degree of sound penetration directly to the space below). Some common lightweight office ceilings tiles, particularly those made of fiberglass, have a high degree of transparency, which increases sound variability below the speakers. Third, the plenum is acoustically complicated by the presence of HVAC ducts, large beams and other structural members which act to compartmentalize the masking sound and cause scattering and reflections. This scattered sound can also cause spatial variability. Fourth, when less transparent acoustical tiles, e.g. mineral fiber tiles, are used, a reverberant acoustic build-up occurs in the plenum that can cause significant "overflow" from the intended treated space, e.g. an open plan office, into spaces where sound masking may not be needed or wanted, e.g. private offices or conference rooms. Fifth, when the ceiling has a very high CAC (ceiling attenuation class), e.g., plasterboard or foil-lined matrix over metal panel systems have a CAC exceeding 40, the transmission of masking through the ceiling plane will be significantly limited and the resulting masking signal problematic in terms of achieved level. Finally, when the plenum is used as a vehicle for return-air for the HVAC system, the ceiling necessarily has vents or open-air returns. If these returns are untreated, they will act as direct transmitters of the acoustic field from the plenum to the office area and create additional variability.
Treating the open-air returns is straightforward, but does add cost to the installation. Properly tuned and adjusted, plenum-based systems, when used in conjunction with treated open-air returns, have been shown to provide uniformity within many target sound masked spaces. Uniformity can be achieved by adjusting the acoustic output of individual or small groups of speakers. Adjustments routinely include changes in output volume and output spectra of individual speakers. To provide this adjustment capability, additional system electronics for individual speakers or for small groups of speakers are required.
Direct Field sound masking systems have been in use since the late 1990s. The masking acoustics is called direct field because the sound from any specific masking speaker travels directly to a listener without interacting with any other reflecting or transmitting feature. Initially used as an accessory for open office cubicles, direct field systems have been fully integrated into at least one open office furniture system and have been designed to be installed both in dropped ceilings and in offices without any absorptive ceiling systems. When installed in dropped ceilings, direct field systems use speakers that are mounted facing down, When a ceiling tile is not available, they are mounted facing down on any available structure, sending the masking noise directly into the intended space. Direct field masking requires speakers that are, in effect, omni-directional, meaning that they transmit energy equally in essentially all directions. The use of omni-directional speakers, spaced appropriately for the work area (typically on a grid equal to the ceiling height), provides sound masking that is evenly distributed. Using direct field speakers eliminates issues of spatial uniformity and overflow due to plenum conditions and open air returns. Because the plenum and ceiling materials are not acoustically involved, individual speakers do not have to be adjusted to counteract plenum conditions, so no tuning is required and electronic complexity is minimized. Also, because the sound from direct field speakers does not have to interact with the ceiling tile, the amount of energy required to produce a sound masked space is reduced.
In Open Ceilings
In many installations, particularly warehouses that have been converted to office space, the masking speakers are hung in a similar manner to those above suspended ceilings. Typically, the speakers are mounted higher and the spacing between them is often closer. The space generally has a very high structural ceiling, so the sound created is quite diffuse (occupants cannot locate the speakers easily by listening).
Under Raised Floors
In offices that utilize raised floors, masking speakers can be placed under them. Special speakers have been designed to accommodate even very shallow cavities, otherwise normal masking speakers can be used there. Listener acceptance is very high with this type of design as the sound is very diffuse.
Advances in Sound Masking
Sound masking has been in use for many years. It is likely that Roman villas with interior fountains had the benefit of masking the sound of chariots. Even today, fountains are used in malls to provide humidity, a pleasant environment, and sound masking. Due to increasing use of sound masking as a privacy tool, electronic methods have had a number of improvements that enhance the performance and acceptability of them: Equipment has been designed to protect sensitive conversations. Masking has been applied successfully in hospitals and other medical facilities. Masking speakers have been successfully located in open ceilings, above suspended ceilings, face-down in suspended ceilings and under raised floors. Most professionals have converged on a preferred spectrum and level of masking for both open and closed offices that balances the need for privacy with occupant acceptability. Most modern equipment is now capable of handling these requirements. Many systems have an initial ramp function. This function permits the user to increase the masking level automatically to the desired level over many days. Persons moving into a new office prefer slow changes and this function accommodates the acoustical aspect. Some systems have a power ramp function. When building power fails and then is turned on, the sound masking level would jump up quite noticeably, possibly creating a negative response. This function acts similar to the initial ramp function but acts in minutes rather than days. Some systems have a level scheduling function. This permits the user to have the sound masking level vary each day based on office use. If the office is fully occupied, the masking is at its highest level. As occupancy is low early in the workday, and persons are preparing to leave in the afternoon, the level can be reduced. During night hours the level is set low so security guards can hear any suspicious sound as they tour the office. Some systems have an adaptive function. In this system, a sound detector captures the sound, separates the activity sound (typically speech) from the background (sound masking) and adjusts the masking so the difference between the two is kept constant. Essentially, the system applies sound masking only when it is needed.
Attributes of Successful Sound Masking Systems
Sound masking level. For open offices, depending on design, the level can vary from 43 dB(A) to 48 dB(A) and for closed offices, the masking levels can vary from none to 44 dB(A). Adjust of levels must be done in small increments of level near 1 dB..
Sound masking spectrum contour. The frequency distribution of sound masking levels, at a minimum, cover the frequency range of speech which is 160 to 8000 Hz. Most experienced practitioners design a spectrum contour that decreases with frequency at about 4 dB per octave, requiring more low frequency sound than high. That particular shape of the contour has been found to be acceptable to occupants. The level of that contour is set to provide the desired privacy, so a balance between performance and acceptability can be achieved.
Spatial uniformity. Unlike most sound systems that call attention to themselves, sound masking is designed to be background (not recognized), just as the background sounds outdoors are not recognized. This is achieved by having sound masking uniform throughout a room. Uniformity is best achieved by having an intervening material (suspended ceiling, raised floor)between the masking speaker and the listener.
Invisibility. In keeping with the concept of being background, it is desirable to have the sound masking system invisible. The control equipment should be in a separate room. Putting speakers behind an intervening material hides them.
Sound diffusion. When sound is diffuse (arriving from all directions about equally), most people are not even aware of it. The quiet outdoor background is an example. The sound from an airport is just the opposite; the sound can be identified and disliked. Diffusion is greatly improved by use of an intervening material.
Applicability. Masking systems need to accommodate the many locations in which masking speakers are put. They include under raised floors, above suspended ceilings, in open ceiling plenums, above discontinuous suspended ceiling tiles, and on walls.
Portability. The system must be able to be moved to a new facility with minimum cost.
Phasing. When two masking speakers have the same signal, but are separated from each other, a person at the mid-point between them can sense a swishing sound. This is caused by level changes at various frequencies caused by phase shifts. When strongly noticeable, it is considered to be objectionable. This effect is most pronounced in a commercial facility with persons moving about. An intervening material can eliminate this effect. If there is no intervening material, special design is required to eliminate it.
The Relationship of Sound Masking to Speech Privacy
Many times an owner has an office noise problem and has heard that sound masking will solve it. This idea is true in many cases, but not always. Sound masking is only one of three factors that permit speech privacy to be achieved. If the other factors are insufficient, sound masking may not be a solution.
The Three Factors
1. Speech Level
The louder people speak the more they can be heard by others. Fortunately, people in offices tend to moderate their voices, although occasionally there are people with naturally loud voices. Much information about voice levels has been accumulated so that masking system designers have a good idea about speech levels.
2. Sound Attenuation
As the speech travels out it decays to lower levels; it is attenuated. There are three components that create the loss: blocking the speech, absorbing the speech, and natural spreading. The first component is composed of structures as walls or workstation panels that block the passage of speech through them. The second component is composed primarily of fibrous materials, such as ceiling tiles that reduce reflections from them. The third component is determined by distance from the speaker.
3. Background Level
The speech of a person is reduced in level when it finally reaches a listener. If that level is greater than the background level, it can be heard. The object of sound masking is to raise that level so that the amount of speech that is heard is reduced to the point that it causes no distraction.
The Privacy Index
Speech intelligibility has been the subject of much study for over many years and has resulted in a measure called Articulation Index (AI). It is a number from zero to one, with one representing full comprehension of speech, or 100%. The Privacy Index (PI) is inversely related to AI, PI is equal to 100(1-AI) (this is also the difference between the AI percentage and 100). It ranges from zero (no privacy) to 100 (complete privacy). Generally, for office settings, a PI below 80 is unsatisfactory, in that conversations and other noises are easily noticed or overheard. ASTM E 1130 Defines PI.
Privacy Type and Privacy Index
See ASTM E 1137 for officials standards
- SECRET Complete protection of conversations. PI=100+ Grade: A+
- CONFIDENTIAL Others cannot understand. PI=95-100 Grade: A
- NORMAL Very few distractions. PI=80-95 Grade: B
- TRANSITIONAL Many distraction. PI=60-80 Grade: C
- NONE Complete distraction. PI=0-59 Grade: F
Recommended Open Office Levels
- Panel Height: less than Level: 48 dB(A)
- Panel Height: near Level: 47 dB(A)
- Panel Height: near Level: 46 dB(A)
- Panel Height: near Level: 45 dB(A)
- Panel Height: near Level: 44 dB(A)
Recommended Open Office Spectrum when at 47 dB(A)
- 160 Hz Level: 46 dB
- 200 Hz Level: 45 dB
- 250 Hz Level: 44 dB
- 315 Hz Level: 43 dB
- 400 Hz Level: 41 dB
- 500 Hz Level: 40 dB
- 630 Hz Level: 39 dB
- 800 Hz Level: 37 dB
- 1000 Hz Level: 36 dB
- 1250 Hz Level: 35 dB
- 1600 Hz Level: 33 dB
- 2000 Hz Level: 32 dB
- 2500 Hz Level: 30 dB
- 3150 Hz Level: 28 dB
- 4000 Hz Level: 26 dB
- 5000 Hz Level: 23 dB
- 6300 Hz Level: 20 dB
- 8000 Hz Level: 18 dB
Recommended Closed Office Spectrum when at 44 dB(A)
- 160 Hz Level: 41 dB
- 200 Hz Level: 40 dB
- 250 Hz Level: 40 dB
- 315 Hz Level: 39 dB
- 400 Hz Level: 38 dB
- 500 Hz Level: 37 dB
- 630 Hz Level: 35 dB
- 800 Hz Level: 33 dB
- 1000 Hz Level: 31 dB
- 1250 Hz Level: 29 dB
- 1600 Hz Level: 26 dB
- 2000 Hz Level: 24 dB
- 2500 Hz Level: 22 dB
- 3150 Hz Level: 20 dB
- 4000 Hz Level: 17 dB
- 5000 Hz Level: 15 dB
- 6300 Hz Level: 12 dB
- 8000 Hz Level: 10 dB
Set above levels if above existing background level.
Sound Masking and the Privacy Index
How does sound masking fit into the Privacy graph? As an example, the design of an open office workstation starts with two people separated a few meters apart in sight of each other. Building a workstation means moving on the horizontal axis. The Privacy Index is near 10, so they have no privacy from each other. A furniture system with panels is added. If the panels block sound well and are high enough, the Privacy Index moves along the curve to near 50. A big, beneficial, improvement physically, but nothing much happens to the occupants privacy. Ceiling tiles are added, and the speech reflecting from the ceiling is reduced. If the sound absorbing qualities are good, the Privacy Index moves to 75. Occupants begin to notice an improvement, but as can be seen in the table the goal of Normal Privacy (little distraction) is still not met. Adding sound masking then can bring privacy to the goal. This process is known as the ABC s of speech privacy: Absorb speech (ceiling), Block speech (panels), Cover speech (masking). It is the steep part of the privacy curve that gives the most performance and getting there is merely setting up the conditions. This description tends to make sound masking the hero, but the additions can be done in any order; it is the balance of the three factors that is most important.
Most of the privacy dollars are spent getting to the bend in the curve; the rest is spent to achieve the desired privacy.
The failure of masking to provide privacy for nearby persons can also be seen with this graph. To be acceptable to occupants, the level must be restricted to that which is acceptable and this means an improvement of about 15 PI points. In adding sound masking to a poorly designed office, the PI goes from 10 to 25 for them, not worthwhile. This does not mean to imply masking cannot be effective for others in the room. In customer service areas, for example, the panels are very low so the above situation applies for nearby persons. But as the speech travels outward, the level is reduced an acceptable amount for more distant persons, and then sound masking can be effective. The distance beyond which sound masking is beneficial is called the Radius of Distraction. People beyond this radius have good privacy which they did not have before.
The possibility of a detrimental influence on occupants' health has always been a concern in connection with sound masking. There are three types of effects of sound: physical, physiological, and psychological. Physical effect are things that produce direct damage, for example, very high sound levels, at or above 130 dB. The physiological effects occur at about 70 dB and above. At lowest sound levels, these include a slight dilation of the pupils of the eye and slight galvanic skin response, but no permanent effect. Street noises, bus, train, or aircraft transportation noises are typically above 70 dB. Since sound masking is operated at 47 dB, well below 70 dB, physiological responses are not to be expected. Psychological response is the only remaining concern and can occur at any level. Dripping faucets at 20-30 dB for example can result in annoyance and then complaints. Annoyance in a commercial environment is caused by transient sounds intruding on the listener. It is the purpose of masking to reduce the degree of intrusion of these sounds. A quote from the classic book The Effects of Noise on Man by K. Kryter may illuminate the potential problem:
The general finding that the performance of the more anxious personality types is more affected by noise than that of non-anxious types would attest to the existence of a stimulus-contingency factor. A possible teaching of much of the data presented in this book is that other than as a damaging agent to the ear (physical), noise will not harm the organism or interfere with mental or motor performance.
- Kryter, K.D., The Effects of Noise on Man, Academic Press, 1970
- Chanaud, R.C., Sound Masking Done Right, Mitek Corporation, 2009
ko: pl:Maskowanie (psychoakustyka) ru: sv:Ljudmaskering