System Sensor Home
SEARCH IN
FOR


LifeSafety
LifeSafety Newsletter – Current Issue


Home

About Our Company

New Products

   SpectrAlert Advance
   Speakers and
   Speaker Strobes


   Carbon Monoxide
   Detector


   ExitPoint Directional
   Sounder with
   Voice Messaging


Product Support & Install/Specify

Online Training

Seminars new!

Contact Us

Careers

Privacy Statement
Conventional
Detection
Audible/Visible
Notification
HVAC Systems
Monitoring
Sprinkler Systems
Monitoring
Intelligent
Detection

BackVoice Evacuation Design Information

Tips for Voice Evacuation System Intelligibility

NFPA 72 - 2002, 3.3.208 Voice Intelligibility. Audible voice information that is distinguishable and understandable.

• The system should have a good enough signal to noise ratio, i.e., it must be loud enough. NFPA 72, in general, requires public mode evacuation signals to be 15 dB above ambient noise. Signal to noise ratios above this level produce diminishing returns.

• Speakers should have good frequency response between 150 Hz and 11 kHz, the frequencies that male and female voices fall into. The frequency response should be as flat as possible, that is, the response should not vary considerably at the low and high end. While UL only requires that speakers used in fire protection applications have a response from 400 Hz to 4000 Hz, studies have shown that up to 20% of intelligibility is contributed by frequencies above 8 kHz. Frequency response can be affected by speaker installation including the type of junction box and whether or not the speaker is flush or surface mounted.

• Total harmonic distortion should be minimized. Distortion can be caused by poor or damaged speakers, poor or overloaded amplifiers, speaker installation, incorrect wire size, poor message generators, poor message recordings, etc.

• Reverberation should be minimized. Reverberation is the persistence of sound in a room because of reflections of sound from walls and other surfaces after the source has been turned off. Reverberation can be minimized by decreasing the amount of sound energy introduced into a room. Generally, using more speakers at lower tap settings produces better intelligibility than fewer speakers at higher tap settings. Speakers should also be placed as to minimize the distance to the listener. For example, in applications with high ceilings, place speakers on the walls to get them closer to the listener.

• One often overlooked contributor to voice intelligibility is the degree of articulation of the speaker and the rate at which he or she speaks. This factor becomes more important as the area for the voice evacuation system become more and more reverberant. This factor has become more important as fire panel manufacturers have developed field recordable voice chips in their voice evacuation systems.

• System designers should also strive for uniform speaker coverage in order to provide the maximum potential for direct sound at the listener. Again, this factor becomes more important in reverberant spaces. Variation of between 3 and 6 dB should be the design goal, with the lower limit being used in reverberant areas.

Sources:

1. Seto, William W., Schaum’s Outline of Theory and Problems of Acoustics, McGraw-Hill Book Company, New York, 1971.

2. JBL Professional Sound System Design Reference Manual, Harmon International Company, Northridge, CA, 1999 edition.

3. Kamlet, Rick, “Designing better sounding in-ceiling business music systems,” Harmon International Company, Northridge, CA, 2004.

4. “Speech Intelligibility – A Technical Note,” Technical Notes, Volume 1, Number 26, Harmon International Company, Northridge, CA.