T11M

In an effort to move all non-academic departments out of academic buildings the UO has opened up space in the first floor of Chapman Hall.  We are currently working on taking over that space and moving our professors into it – the goal being all Honors College profs having offices in Chapman.  During this time the subject of confidentiality/privacy has come up.  For some reason certain people believe that their new offices are not going to be confidential and that conversations will possibly be overheard by those outside of their office.  Obviously this could be of concern if very confidential meetings are taking place in these offices.  However, professors don’t generally hold super-secret meetings in their offices and if they do, those conversations aren’t usually loud enough to penetrate through walls.  But alas, paranoia has set in the minds of a few and now the Dean has taken up their cause.  Enter me.  I’ve been charged with finding a way to ensure the privacy of conversations conducted in these new offices.  Since this topic does come up on occasion and I haven’t found any good overviews of it, this post will outline my findings, reasoning, and recommendations.

Some Definitions:

• Decibel (dB) – in sound: a logarithmic unit used to express the magnitude of sound intensity.
• White Noise – noise with the same intensity at all frequencies.  Basically the noise is the same dB level and is produced at all frequencies.
• Ambient Noise – the pervasive noise associated with a given environment.  Basically the lowest dB sound you can hear and comprehend.
• Sound Masking – the addition of sound into an environment to cover up unwanted sounds.

How sound masking works:

Sound is a pressure wave that is received by the mechanisms of the ear and translated by our brain into comprehensible sounds.  When the ear receives these sound waves two things affect the way they are heard.  Without going too in-depth, these two things are the frequency of the sound and the power (also known as volume, magnitude, or pressure) of the wave.  Frequency is normally measured in cycles per second or Hertz (Hz) and is what determines a sound’s pitch.  The power of sound is measured in decibels (dB) and is commonly referred to as the volume of the sound.  The human voice can produce sounds with frequencies between 250-6,000 Hz and an ordinary conversation is generally conducted at about 60 dB. (About Decibels)  Because of the nature of pressure waves and the way the ear receives them, the highest power pressure waves are heard by the ear.  This means that the ear can hear multiple frequencies at the same time so long as they are at similar decibel ranges.  If for example a sound is created at 500 Hz and is 60 dB and another is produced at 1000 Hz and 60 dB, a person will be able to hear both and comprehend the difference between them.  However if one sound is produced at a much higher decibel level than the other it will be difficult, if not impossible, to hear the other sound.  One other principle that affects the way sound is heard is the way pressure waves of the same frequency will combine.  Thus if two sounds of the same frequency are produced they will combine together to become one wave and this new wave is what is heard.  It is these principles that make sound masking possible. 

According to Lencore, a prominent manufacturer of sound masking systems: “Sound masking works by subtlety raising the ambient background sound level, thereby reducing sound’s dynamic range. This effectively “masks” unwanted noise, makes speech unintelligible (creating privacy), and makes the work environment acoustically comfortable. (Lencore)“  Basically a sound masking system produces constant sound at a wide range of frequencies.  These sounds then combine with the sound waves you are trying to cover up causing the person to hear only a slight increase in magnitude of certain frequencies.  This increase in the magnitude of frequencies is incomprehensible because those frequencies are still surrounded by the constant production of sound at all other frequencies.  However, if an outside sound is produced that has a much higher decibel range than the sound masking system the sound will be heard and comprehended.  Again Lencore speaks to the appropriate adjustment of a sound masking system: “To be effective, the masking level should be 3 to 5 decibels louder than incoming speech from adjacent work stations. (Lencore)“  For example, if a person sitting in an office is able to hear through the wall another person speaking at 50 dB then the sound masking system must produce sound at 55 dB to adequately mask the sound. 

It should be noted that sound masking does not mask sound at the point of generation – it masks sound at the point of reception.  So in an office environment all people must be exposed to the sound masking

Keys to Effective Sound Masking:

Much of the following information was gleaned from Lencore’s website and the Wikipedia article concerning sound masking.  Some keys to effectively implementing a sound masking system are:

• Sound masking must produce sound at all frequencies that can be created by the human voice.
• Sound masking systems should be barely perceptible.  Generating too much sound makes them distracting and obvious.
• The sound generated should be non-directional.  If a person can tell where the sound is generated it is no longer subtle and may be distracting.
• The sound should be uniform throughout the space.  If it is not then the system does not provide complete sound masking.  It will also be noticeable that sound masking is being used because people will be exposed to different dynamic ranges of sound. (Example: If sound masking is being used in an office but not in the adjoining hallway then the sound masking will be noticed everytime a person enters the office from the hallway)

Options:

A well implemented sound masking system will use all of the keys to effective sound masking (listed above) and will be minimal in cost.  Unfortunately these two goals tend not to be compatible.  Sound masking systems appear to either provide all of the keys to be effective and are very expensive or do not provide the keys to be effective and are reasonably priced.  I have reviewed three products that I believe best represent the possibilities for sound masking in an office environment. 

1. Sonet Acoustic Privacy System (by Speech Privacy Systems) – This is a system designed specifically for home use.  It is advertised as producing a sound like the whoosh of a high quality ventilation system and is adjustable in volume.  However, there are no listed specifications regarding the frequency or decibel ranges that this device can cover.  It comes with a sound generator and two speakers with extensibility of up to four speakers.  It’s effective range is 50 – 500 square feet depending upon the number of speakers used.  The minimum $200 price tag makes it the cheapest solution considered.  The biggest benefit of this system is the lack of set up time needed along with the minimal (in comparison) cost.  The main downfall to this solution is it lacks the ability to produce non-directional, uniform sound.  It also uses residential grade, not professional grade hardware. 
2. Voice Arrest Speech Privacy System – This system is designed for business use and is composed of a central control module and a large number of sound emitters.  There are two different central control modules – one can handle 1-30 emitters and the other can handle 1-300 emitters.  This scalability allows for better customization to the office environment and the ability to better provide non-directional sound.  Again there are no specifications listed as to the frequency and decibel ranges capable by the system.  The central control module is priced at $679.95 for the smaller unit and $1,099.95 for the larger unit.  The emitters are sold separately and are $79.95 each.  While this solution is highly customizeable and will provide a very good sound masking environment it will prove to be very costly.
3. Custom Designed Implementation – There are a number of businesses that are willing to do custom implementations of speech privacy systems – Lencore and Speech Privacy Systems are two examples.  These companies will do full evaluations of your current operating environment and will design systems specifically for your facility.  These systems are best deployed in large, open offices (“cube farms”) and are going to be very costly.  Neither company gives any sort of estimate as to the cost because their solutions are custom built and vary widely in cost.

Recommendation:

It is imperative there be empirical evidence showing that there is a need for a sound masking system before any purchases are made.  Tests should be conducted using different levels of sound to find if people outside of each office could overhear a normal conversation.  Then if it is indisputably found that conversations cannot be conducted in private I suggest that a decision be made based upon three criteria – cost, the ability to install the system, and input by both the concerned people and the people who will be affected by the implementation of a sound masking system.  These three criteria have to be weighed and balanced against each other.

I personally would most likely choose the Voice Arrest Speech Privacy System because of its ability to be customized.  This allows for the system to meet most, if not all, of the key criteria for an effective sound masking system.  In my opinion, anything less than this will probably be more of annoyance than good solution.

Work Sound Masking