Owen T. Heisler

Last updated 2017-Jul-02

On this page:

This page is also available in the following formats:

Questions or feedback? owenh000@gmail.com

1 Introduction

This document has been prepared as a resource to help you set up, improve, maintain, and operate sound systems for speech reinforcement.

The primary focus is on the theory, so you can apply it in your particular situation with the hardware you already have. Hardware suggestions are also offered. If you would like a ready-to-use complete system design instead, without the theory, please refer to the "Speech Reinforcement Design" documents available on owenh.net.

This is a work in progress. This document is provided with no warranty or guarantee of any kind. If you find any error here and can support an argument against it, please contact me. “Thank-you” to those who have answered my questions or otherwise offered input.

1.1 Signal path

A sound system is a collection of interconnected equipment, and the audio signal travels through each of those parts. The performance of the system as a whole is affected by each part of this signal path. Each chapter in this book discusses a particular part of the signal path. To keep your signal quality high, every component in the signal path needs to be working correctly.

Here is an example of a signal path:

Human speakermicrophonemixing/processing

  1. Room speakersroomears

  2. assistive/outside listeningears

  3. telephone networkears

1.2 Selecting hardware

When selecting hardware, look for quality first, then look for a feature set close to what you need. Often a well-designed device with a limited feature set will work better than a poorly-designed device with an extensive feature set. Do not make the mistake of looking for the cheapest product with the features you need: likely those features are poorly implemented.

If price is a concern, consider buying used equipment. Generally, used high-quality hardware will serve you better than new low-quality hardware. But remember to test used hardware thoroughly before using in an important event!

1.3 Other considerations

1.4 Todo

2 Human Speaker

The human behind the microphone is obviously the most important part of the signal path. A well designed sound system can easily accomodate most voices.

2.1 Suggestions for the person speaking

It is seldom practical to try altering the behavior of the person speaking, and generally impossible to affect a change in the actual sound produced. However, there are a few suggestions that could be shared with any speaker who is interested.

  1. It can be counter-productive for the lips to be closer than 3 inches from the microphone. This is always a compromise between gain-before-feedback and signal accuracy. ONLY when feedback is a problem should the speaker move closer to the microphone (and feedback should not be a problem with proper speaker placement and managed frequency response). For more information about the proximity effect, see the Microphones chapter.

  2. Speak clearly to establish a high initial SNR. TODO speaking loudly, Lombard effect

2.2 Frequency ranges

Having covered some suggestions for the speaker, we now are concerned with the technical characteristics of voice so we can understand what our "signal" actually is.


2.2.1 Speech critical bands

The following is a selection of narrow frequency bands which may be treated as the elementary signals in speech.

Speech critical bands
Center Width Range (Hz)
200 60 170-230
300 60 270-330
500 60 470-530
800 70 765-835
1000 80 960-1040
1500 100 1450-1550
2000 130 1935-2065
3000 200 2900-3100
5000 300 4850-5150
8000 600 7700-8300

Consequently, on a 1/3 octave equalizer, the following bands are especially important (though everything from 80 Hz to 10 kHz is important for high-fidelity reproduction):


2.3 Notes/Todo:

3 Microphone

The microphone is the first component in the signal path that you have full control of. You need a good one, one designed specifically for speech. Other traits include: type, pickup pattern, frequency response, and proximity effect.

3.1 Consider These Microphones

3.1.1 Dynamic

TODO: condenser phantom-powered, condenser battery-powered

3.2 Microphone types


3.3 Polar (pickup) patterns


3.4 Microphone stands

3.4.1 Basic

3.4.2 Portable

3.4.3 Remote motorized

3.5 Todo

4 Mixing/processing

4.1 Feedback suppressors

4.2 Equalizers

4.3 Todo:

  1. digital signal processor (DSP) configured by software, (3) processing by computer with audio interface

5 Room speakers

5.1 Selecting speakers

5.2 Placing speakers

5.3 Todo

6 Assistive/outside output

Consider the following requirements:

6.1 Options



6.2 Todo

7 Telephone

This chapter covers the use of a phone and a conference calling service provider to share the audio from a sound system to remote listeners.

Note that there are also many other ways to implement audio streaming, especially over the internet (such as with Asterisk and MusicOnHold streaming).

7.1 Connecting an output to a phone input

Use an XLR to TRRS adapter to connect a system output to a cell phone's microphone input. kV Connection sells two products for this, the KM-IPHONE-MICX and the KM-IPHONE-MICX-A22. These products are identical except that the second one attenuates the signal by 25 dB.

Most outputs will be line level rather than microphone level, and in that case the KM-IPHONE-MICX-A22 is the appropriate option. If the signal level is still too high, or to use the KM-IPHONE-MICX with a line level output, add the Audio-Technica AT8202 adjustable in-line attenuator (selectable 10, 20, or 30 dB attenuation).

Other options: JK Audio BlueDriver-F3, JK Audio Daptor Three

7.2 Using FreeConferenceCall.com

This service is no longer recommended because it provides free conferencing through the use of traffic pumping. This can result in unexpected blocked calls, dropped calls, and usage charges. See related FCC and T-Mobile pages.

Dial strings in this section use a p to indicate a pause, sometimes dial string pauses are indicated by commas (,). You should be able to save the entire dial string, including pauses, to your phone's contacts database. Always listen while your phone is dialing to verify successful connection and mode selection.

  1. Register at FreeConferenceCall.com.
  2. Record your dial-in number, access code, and host PIN.
  3. To provide program audio to the conference, use this dial string:
    <dial-in number> pp <access code> #pppp <host PIN> #ppp*5p*5p*8
    (The *5*5 switches to "presentation mode", with all guests muted; the *8 disables the tones indicating when guests enter or exit the conference.)
  4. Guests can be given the dial-in number and access code; they will be prompted to enter the access code and then press the pound/hash button (#). Here is a dial string:
    <dial-in number> pp <access code> #.

8 Room

The room greatly affects the audio signal. This can be fixed, in part, by installing acoustic treatment. It can be accomodated, in part, by applying to the signal an inverse of the room's frequency response.

Room acoustics, modes, and responses are perhaps the most complicated part of the signal path. This page is certainly not comprehensive, but aims to touch on some basics.

8.1 Acoustics


8.2 Todo

9 Ears

9.1 Frequency


9.2 Delay


9.3 Todo

10 Wiring and Cables

10.1 Mic/line level cable

Use a 2 conductor cable with a shield for any of the following:

Use eg. Belden 9145 for installed wiring (in a building or rack cabinet).

10.2 8 ohm speaker wire

As speaker wire, use 2 conductor wire. Shielding is not necessary and increases the risk of a short.

10.3 70V speaker wire

For 70V, use 2 conductor wire. Shielding is not necessary and increases the risk of a short.

18 AWG wire will yield less than 5% power loss up to about 100 m (330 ft).

For installed 70V audio, use 18 AWG or larger wire. Solid and stranded wire are both acceptable, up to 12 AWG. Stranded may be easier to work with when soldering connectors, especially with large gauge wire.

For portable 70V audio, use stranded 18 AWG wire or larger. Stranded wire is more durable for handling.

10.4 Attenuators & isolators

10.5 Todo

10.5.1 Compatible wiring scheme

Amplifier output is 70V, 2 channels, with male twist connector.

Supported outputs:

Standard cables/adapters required:

Special cables/adapters required:

Suggested connectors:

10.5.2 More speaker wiring

There are two options for speaker wiring: low impedance and high impedance. Low impedance is typical for basic systems with few speakers and short cable runs. High impedance or constant voltage wiring (typically 70 V in the United States) requires a transformer at each speaker but offers numerous advantages over low impedance:

See also:


11 Power

11.1 Uninterruptible power supplies (UPSes)

I recommend the CyberPower OR700LCDRM1U.

The following models are similar: OR500LCDRM1U, OR1000LCDRM1U, OR1500LCDRM1U, PR750LCDRM1U, and PR1000LCDRM1U.

12 Storage

12.1 Portable racks

I recommend the Gator Pro Series Molded Racks, available from 2U to 12U, eg. the G-PRO-6U-19.

Other options include Gator Standard Audio Racks, SKB Roto Racks, and Citronic LLDPE 19" Rack Cases.

13 Hardware Brands

This chapter lists some common brands with notes about their subjective reputations. You can learn more by testing devices or by reading reviews online.

14 Measurements

Being able to take measurements is very important, as it removes much of the guesswork involved in making a sound system effective.

You should be able to do the following measurements:

14.1 Hardware

Audio analyzers are available that can provide much of this functionality. However, the following hardware acts as additional audio inputs/outputs for a computer (via USB) and can be used not only for room measurement, but also for signal processing in general. These been selected for some combination of portability (size), versatility, and quality.

You will also need a measurement microphone. I am using a dbx RTA-M, but there are probably better ones. Dayton Audio offers 3 test measurement microphones, and provides a unique calibration file for each microphone that they sell (cross-referenced by serial number). In the case of the UMM-6, you would not need the Shure X2U input adapter, but you also would be unable to connect a regular XRL microphone as input if you wanted to eg. try analyzing a microphone.

14.2 Software

14.3 Other things of interest

14.4 Todo