Noise Masking, Sound Masking

Noise Masking is often applied in indoor environments to hide unwanted noise, create a pleasant atmosphere, to provide a confidential work place and to create an acoustical privacy in open and partitioned office areas. Sound or Noise Masking can increase conversational privacy by as much as 100 percent.
Normally it should sound much like a gently blowing breeze. Sound masking will not make a noisy office quiet. It contributes to speech privacy, reducing the understandability of conversation between neighbors.
A masking sound system must assure uniform coverage and must be tuned to the appropriate level and frequency spectrum. HVAC noise and music need not be considered for masking situations because it is not steady and can be turned on and off. Normally the sound is created by a digital noise generator, processed through equalizers to adjust the frequency of the individual areas. The installation can be critical to reach the wanted even coverage in the room. After finishing an installation, the result could also very unsatisfying, that in some areas costumers want to turn the system off, in other areas it doesn't help to improve privacy ...
For a successful installation, it is important to be included in the design stage. There should be meetings that include the HVAC designer, the architect, the acoustic designer and the owner, all of whom need to clarify such issues as what kind of sound levels are required.
The term NC must be considered. NC (noise criterion curves): curves used to relate the intrusiveness of such noise as HVAC systems in buildings. NC 35 to 40 is rated optimum for speech privacy.

The noise signal should resemble the shape and range of normal human speech with a frequency range of about 200Hz to 5kHz. The signal should have an overall roll off of 5-6 dB per octave from 200Hz all the way to 5kHz.

preferred loudspeaker installation

There are two types of noise masking systems: "in-plenum systems" and "direct field systems". The in-plenum system is a loudspeaker system with all speakers installed in the space between the concrete ceiling of the floor and the suspended or dropped ceiling. With loudspeakers facing upwards a very even noise pattern can be achieved. For this system the material of the dropped ceiling much be somehow acoustically transparent. The combination with an acoustic ceiling material can be very useful. Because of the higher SPL reached within the space between the concrete ceiling and the suspended ceiling the system must be tuned carefully and eventually resonating pieces must be tightened. The system must be tuned for the appropriate frequency response in the room, eventually creating a very different frequency response in the ceiling space.
The direct field system is more critical because the individual loudspeakers are turned downwards into the room. Especially with low ceiling heights a sufficient number of speakers must be used. The tuning of these systems is not as critical as with in-plenum systems because no additional room is involved.

The loudspeakers must be placed so that they form a uniform pattern of coverage. Spacing of the individual loudspeakers depends on the room depth and ceiling height. Sometimes conventional box loudspeakers are used and fired sideways or through diffusers, depending on the room layout. Planar loudspeakers have been used successfully too in many situations because of their non-directional characteristics.

To create a random environment, three noise sources should be used. This way each 'listener' is positioned somehow in between thee different sources with slightly different signals for effective masking. More than three source signals are not necessary. , no two adjacent speakers emit the same sound. Each neighbor produces a randomly different sound.

creating an random environment with three noise signals

It is of extreme importance that levels do not vary more than 2dB through the areas. The proper EQing is also very important and can determine a good or bad result of the entire installation.

speech signalsnoise signal
masking close speechmasking far speech

One of the growing trends in masking is using dual noise generators or multi-channel generators for every other loudspeaker in the layout. The most positive aspect of this trend is the reduction of the overall coherency between the two channels and elimination of the potential phasing problems. Even looped recorded sounds (multi-channel and different loop lengths) can be used to make the sound masking system more convenient for office workers. Systems can be automatically controlled to provide the correct level of privacy dependant on the time of day and day of the week. Systems often have automatic attenuation to make the system less apparent after hours.

The sound/noise marking system should generally provide a level of 47 .. 50 dB(A), in smaller spaces 42 .. 45 dB(A).