If you want to hear what various lengths of reverberation time sound like with the same speech source, just click on the samples below. These are now available in both "hi-fi" stereo 16 bit ".wav" files, for use with the standard 16 bit PC soundcard media players, MP3 files that can be played by a number of applications listed at www.mp3.com.

1. Dry speech sample, no reverberation
   • WAV File (110kB)
   • MP3 file (23kB)
     
     
2. Speech in a room with a 0.6 second RT
   • WAV File (150kB)
   • MP3 File (29kB)
     
     
3. Speech in a room with a 0.8 second RT
   • WAV File (158kB)
   • MP3 File (31kB)
     
     
4. Speech in a room with a 1.3 second RT
   • WAV File (180kB)
   • MP3 File (35kB)
     
     
5. Speech in a room with a 2.0 second RT
   • WAV File (251kB)
   • MP3 File (48kB)
     
     
6. Speech in a room with a 5.0 second RT
   • WAV File (294kB)
   • MP3 File (55kB)
     
     

Why does excess reverberation affect speech intelligibility?

Sound files constructed by Wade McGregor using Sound Forge software with a QSound plug in module for a more realistic listening experience.

Reverberation time refers to the amount of time required for the sound field in a space to decay 60dB, or to one millionth of the original power. In simple terms this refers to the amount of time it takes for sound energy to bounce around a room before being absorbed by the materials and air. Closed spaces that don't have materials that absorb sound have long reverb times (concrete arenas, big rooms with Gyproc walls, big water tanks etc.), and very absorbent rooms have short reverb times (like a movie theatre, or a carpet and drapery showroom).

Reverberation time is important because it can affect how well you understand speech, and it can change the way music sounds. The effect on speech intelligibility is noticeable in a gymnasium or arena, where you often can't understand someone who is only 10 or 15 feet away from you. Once you are an adequate distance away from a person speaking to you, the sound level of the room's reverberant field is as loud as their voice. The effect of this could be replicated by having a large group of people standing beside you repeating what you say at different levels and at different times. All this additional babble coming back from the walls, floor and ceiling mask and muddle what is being said. When the reverb time is long enough, it not only masks the next syllable, but it can mask the next word. The preferred reverberation time range for a space intended for speech is 1.0 second, or less. As the reverberation time (or RT60) becomes longer than that, it becomes increasingly difficult to understand unaided speech, and if it is much shorter the room sounds very dead. As the RT60 gets much longer, 3, 4, or even 5 seconds, speech becomes impossible to understand. This can become a life safety hazard in some environments like swimming pools, and industrial settings, where warnings cannot be understood.

Music is the other common victim of inappropriate RT60. A pipe organ sounds like a huge harmonica in a very dead room (say with a RT60 under 0.8 seconds), it may be loud but it has no character. A symphony orchestra in an anechoic chamber (a room with no reflected sound) would all but disappear if you looked the wrong way. On the other hand, a lone drummer playing in a space with a 6.0 second RT60 would disappear into a sound blizzard of reflections. For each type or style of music there is an optimum RT60, as well as a preferred set of early reflections.

Speech and music are not the only sounds that persist in a reverberant room. Every noise keeps going and going and going. All sound sources, including: impulsive noise from dropped objects; a hockey puck hitting the boards; lighting buzz, heating or cooling system noise; a shouting audience; a supercharged 500 cubic inch engine in a monster truck; or dozens of motocross motorcycles; all of these sounds keep rattling around through the highly reverberant room. As a result, noise levels are also adversely affected by excessive RT60.

There is an optimum RT60 for any space. In a church, for example, the optimum RT60 will depend upon whether the ministry is based on acoustic music, a large pipe organ and choir, or electric or amplified music. Because speech is a major component of all church services, some compromises are likley going to be necessary since the best conditions for music are often the opposite of what is required for speech intelligibility. Elegant compromise is the central component of good engineering, and a careful balance between the acoustical design and the sound system design can preserve the musical acoustics desired, and deliver the speech intelligibility required.

Excessive RT60 is the number one problem we see in spaces intended for assemblies of people or music performance. Somewhere along the line, the acoustics of the space were ignored, or the acoustical treatment was cut because it was too expensive. For any space where speech and music are central to the purpose of the room, the acoustics of the space are as critical as having enough air conditioning, enough light, enough heat or enough seats. If the space can't be used for the purpose it was built, then the room design isn't successful, no matter how nice it looks. What's frustrating for us is that it always costs more to retrofit acoustical treatment than to install it during construction.

JavaScript Quickie 2kHz Reverb Time Calculator (Requires a fully JavaScript capable browser)

Return to Common Acoustical Problems

Demo Details

The reverb samples included here are intended to demonstrate a sound source with a nominal Q of 2 at critical distance. What that techno-jargon means is; a sound source with a Q (or directivity factor) of 2 (approximately that of a person speaking) has a roughly hemispherical sound projection pattern; and critical distance is the distance between a sound source and the listener where the direct sound from the sound source is equal in sound level to the reverberant sound field. When the listener is at critical distance, the sound level will not drop off as the listener moves further away from the sound source. This is typically the greatest separation distance where face to face communication is possible between a talker and a listener in a reverberant space. You can increase the critical distance by increasing the directivity of the source. A talker can do this by cupping their hands around their mouth when they talk. This reduces the surfaces that contribute to the sound energy bouncing around in the room, reducing the reverberant sound level at the listener's position.

You can try this experiment yourself in a highly reverberant space like an arena or gymnasium. Compare the maximum effective communication distances between a talker and listener in a reverberant gymnasium, and outdoors in a quiet setting, with and without increased directivity.