WO1996036109A1

WO1996036109A1 - Distributed self-adjusting master-slave loudspeaker system

Info

Publication number: WO1996036109A1
Application number: PCT/US1996/006187
Authority: WO
Inventors: Harry L. Davis; Thomas R. Horrall
Original assignee: Bbn Corporation
Priority date: 1995-05-10
Filing date: 1996-05-02
Publication date: 1996-11-14
Also published as: AU5852596A

Abstract

A distributed self-adjusting loudspeaker system wherein each master loudspeaker unit includes self-contained circuitry (14) which senses ambient noise near each master loudspeaker (16) immediately prior to or during an announcement and automatically adjusts the amplified output level of the announcement at each individual master loudspeaker (16) and associated slave loudspeaker(s) (20), to compensate for the ambient noise levels. The system distributes a common audio signal to a plurality of locations subject to differing ambient noise levels and comprises a plurality of master loudspeaker units (16) each associated with a respective location, each of the units includes at least one loudspeaker and a microphone (18) for sensing noise at the respective location. The system includes single or multiple message inputs and a single central preamplification and conditioning circuit (10, 12) responsive to the signal input(s) for supplying the common signal input signal(s) to each of the master units. Each of the master units includes means, responsive to the respective noise sensing microphone (18), for adjusting relative gain applied to the message input signal(s) to a respective value corresponding to the local noise level and for applying the gain adjusted input signal to the respective loudspeaker.

Description

TITLE OF THE INVENTION

Distributed Self-A justing Master-Slave Loudspeaker System

FIELD OF THE INVENTION

The present invention relates to audio loudspeaker systems, and more particularly to a system which distributes a common audio signal (message) from one or more inputs to a plurality of locations subject to differing ambient noise.

BACKGROUND OF THE INVENTION

Audio loudspeaker systems are implemented in various environments which affect the intelligibility of the signal output, and the overall efficacy of the communication system. For instance, audio paging systems may be implemented in environments, such as a shipboard paging system implemented on a submarine or a public address system implemented in a stadium, wherein background noise varies both spatially and temporally as operating conditions change, e.g. mechanical equipment is cycled on and off, or crowd noise escalates and subsides. The undesirable background noise serves to mask amplified speech to a dramatic extent. Speech that is perfectly intelligible in a given location may be rendered completely unintelligible a few moments later when background noise levels increase. Increasing audio levels system wide, to compensate and maintain intelligibility during noisier periods in particular places, may result in system output being uncomfortably or unacceptably loud in other areas where background noise is invariant.

Some systems for adjusting the output level of loudspeakers in an audio paging system typically involve manual adjustment of loudspeakers on an individual basis. Such manually-adjusted loudspeakers generally remain at a fixed volume level regardless of varying ambient noise levels near each loudspeaker. Loudspeaker output levels are set and will change only when and if they are manually reset. Manually resetting the loudspeaker output levels to compensate for noise at a particular location during a particular time period usually results in the output levels being too great at other times when ambient noise levels are lesser. Consequently, loudspeaker output levels may be set unacceptably high so as to be uncomfortably loud during some announcements when ambient noise levels are relatively low, or loudspeaker output levels may be set too low rendering loudspeaker messages unintelligible during time periods when ambient noise levels are relatively high. Otherwise, frequent manual readjustment of the various speakers in numerous locations, or at some central location, must be undertaken to optimize outputs to accommodate temporal and spatial ambient noise level changes.

Some automated systems, such as one disclosed in U.S. Patent No. 3,410,958 to Cohen, are known to include a system-wide automatic output adjustment mechanism. Accordingly, the loudspeakers in the system operate in conjunction with a centrally located large power amplifier(s) serving a multiplicity of similar loudspeakers distributed among remote locations. Such syste s use a centrally located processor or circuitry near the centrally located power amplifier(s) to adjust gain, and a remote sensor in the area of loudspeaker coverage to send a signal to the centrally located gain circuitry to simultaneously control the level of all loudspeakers connected to the respective power amplifier(s) . During the design of such systems, it is necessary to be able to anticipate where background noise will vary in a fashion representative of all areas served by the amplifier(s) and loudspeakers attached to it so as to be able to appropriately locate the noise level sensor. There typically are not a plurality of master loudspeakers that are independently controllable. In terms of hardware, such systems disadvantageously require a processor and separate sensor with dedicated wiring typically several hundred or thousand feet in length for each power amplifier and loudspeaker or loudspeaker group controlled. Such systems typically require advance information regarding the nature and predictability of background noise level changes, and require elaborate calibration and setup procedures. More advanced versions of these systems are typically implemented including sophisticated digital electronics installed in the signal chain just before the power amplifier(s) , and adjust loudspeaker output levels en masse in accordance with the centralized noise compensation considerations.

SUMMARY OF THE INVENTION

The present invention provides a distributed self- adjusting loudspeaker system wherein each master speaker includes self-contained circuitry which senses nearby ambient noise and automatically adjusts the amplified output level of a common input signal to compensate in order to make the announcement audible or intelligible.

According to the invention, the system distributes a common audio signal (message) to a plurality of locations subject to differing ambient noise levels and comprises a plurality of master loudspeaker units each associated with a respective location or local area of coverage. Each of the units includes at least one loudspeaker and a microphone for sensing noise at the respective location. The system includes one or more signal inputs and may include signal compression, amplification and conditioning circuitry responsive to all of the signal inputs for supplying a common input signal to each of the distributed master units. Each of the master units includes means, responsive to the respective noise sensing microphone, for adjusting relative gain applied to the common input signal to a respective value corresponding to the local noise level and for applying the gain adjusted signal to the respective loudspeaker.

Each master unit incorporates circuitry to produce announcements with a fixed speech (or other input) to acoustic noise ratio over a wide range of noise levels, and a sensing microphone to sense ambient noise in proximity to the master loudspeaker. The sensing microphone works in conjunction with distributed noise controlled amplifier circuitry having automatic gain control to individually adjust each local loudspeaker's output volume as local ambient noise levels change.

An associated compressor/equalizer may be included to provide constant voltage control circuitry to adjust the audio signal input to the system to provide a uniform input level independent of its source level. The compressor/equalizer includes a gain control feature that allows the dynamic range of the system to be optimized to compensate for varying input source characteristics. Equalizer circuitry shapes the frequency spectrum of the input signal, to optimize system fidelity or clarity in accordance with important intelligibility or other frequency bands.

Individual loudspeaker outputs are automatically adjusted for the ambient noise levels in their immediate vicinity to optimize output of the common signal input signal for maximum intelligibility or clarity, regardless of differences in ambient noise levels at various respective locations. The common input signal is equalized by signal shaping and processing for accurate reproduction to provide maximum talker recognition and optimal signal intelligibility or to otherwise compensate for effects of background noise on the perception of the signal.

In an alternative embodiment, the noise sensing and level adjusting circuitry in the loudspeaker unit is powered by simplexing dc power onto audio lines to effect the use of common lines to simultaneously transmit ac audio signals and dc power. A primary feature of this alternative embodiment is that existing sound systems can be retrofitted with distributed automatic level adjustment capability, according to the invention.

BRIEF DESCRIPTION OF THE DRAWING

These and other features of the present invention will be better understood when reading the following detailed description, taken in conjunction with the drawing, wherein:

Fig. 1 is a diagrammatic illustration of a system having distributed automatic level control according to the invention;

Fig. 2 is a block diagram of noise control amplifier and control circuitry for a system having distributed automatic level control according to the invention;

Fig. 3 is a block diagram of a system having distributed automatic level control according to the invention; and

Fig. 4 is a diagrammatic representation of automatic level control implemented in a constant voltage (typically 70 volt) announcing system known in the art.

DETAILED DESCRIPTION

A system for distributing a common input signal to a plurality of locations subject to differing ambient noise levels, according to the invention, is generally illustrated in Fig. 1. The system involves an unconditioned input signal, or common signal input signal, which may be a voice broadcast signal, background music, or the like. Compressor 10 and equalizer circuitry 12 receive the unconditioned, common signal input signal to provide optimization. The optimized signal is passed to noise controlled amplification circuitry 14 located at a master speaker 16, which also receives a signal indicative of ambient noise from a sensor microphone 18 located proximate to the master speaker 16. The optimized common signal input signal is then adjusted in accordance with the signal indicative of ambient noise, to produce a volume corrected, amplified signal or output signal broadcast from the master loudspeaker 16 and any associated slave loudspeaker(s) 20.

The primary purpose of the signal compressor is to keep unconditioned, common signal input signal to the system constant in level. Other circuitry described hereinafter changes signal level by varying gain at loudspeakers as required to compensate for the local effects of variable ambient noise, but the common signal into the system must be maintained constant to assure that the final speech-to-noise ratio desired is achieved. The initial conditioning effected by the compressor (in conjunction with the equalization discussed hereinafter) contributes to the accurate reproduction of the input signal and to enhanced talker recognition.

Specifically, the compressor 10 provides constant voltage control to the input signal that compensates for signal variations, for example, in the sound pressure level at an input signal microphone diaphragm due to different talker effort and level, and due to variable microphone-to-talker distance. The latter effect can cause dramatically different microphone output levels with conventional microphones, and is exacerbated when noise-cancelling differential type microphones are used.

The signal compressor 10 operates by electronically monitoring its output level and automatically adjusting its gain as required to keep the level constant. If a talker speaks loud or holds the microphone close, the signal output from the compressor would tend to rise. However, the rise is kept to a fraction of a decibel, in accordance with a selectable gain as discussed hereinafter, because a high gain negative feedback control circuit quickly serves to reduce the overall gain of the compressor 10. If a talker speaks with low effort or holds the microphone further away, the compressor compensates by increasing its gain accordingly.

The signal compressor 10 may include a user- accessible switch 22 (Fig. 1) for changing the reference output, after compression, to the levels desired for various modes of operation. For instance, it may be desirable to be able to manually switch the compressor 10 to achieve selected gains for emergency (loud) announcements, normal announcements, and/or quiet announcements. Exemplary nominal compressed output voltages from the compressor implemented with selectable gains are 1 Vrms (0 dB) in an emergency mode, 1/2 volt (- 6 dB) in a normal mode, and 1/4 volt (-9 dB) in a quiet mode. A second, non-user-accessible switch (not shown) , may be provided to adjust the amount of gain to optimize use of the system in different environments or with different equipment.

The equalizer 12 is intended to adjust the overall frequency response of the system as required for optimum balance between intelligibility, naturalness and signal power. In a public address system such as in this illustrative embodiment, adjustments are provided for speech frequencies from 315 Hz to 5000 Hz in 1/3 octave bandwidths. The range of adjustability for each band level is 24 dB. The adjustment is accomplished by installing semi-permanent resistors soldered to a removable plug-in component carrier in the equalization circuitry of the illustrative embodiment. The level of each band is established by a single resistor in the range of 0-500,000 ohms. A plug-in potentiometer test panel permits convenient band level adjustability during initial system setup. Following optimum adjustment, the resistance value of potentiometers is determined at a test jack and the equivalent fixed resistors are soldered to the component carrier to replace the potentiometers. The accuracy of the resistors is non-critical: 10% is satisfactory. The equalizer 12 includes a calibrated precision stepped attenuator which permits gain adjustment with 1 decibel resolution over a range of 59 dB.

In the system according to the invention, the compressed/equalized signal is passed to a noise control amplifier (NCA) 14, as illustrated in Fig. 2, for correction in accordance with ambient noise levels sensed proximate to the loudspeaker(s) . The NCA circuitry is intended to be installed within an amplifier unit resident in the master loudspeaker 16 enclosure, as a representative embodiment. The NCA is comprised of two main elements.

The first NCA element is an ambient-noise sensing sound level meter, or SLM 24. The SLM 24 receives a signal representing ambient noise sensed with a dynamic omnidirectional sensing microphone 18. The noise signal, in this illustrative public address system embodiment, is filtered by a 1-4 kHz octave band filter 26 so that the SLM output is sensitive in the frequency bands which contribute most to speech intelligibility. The filtered noise signal is rectified 28 so that the SLM output is a rectified DC control voltage proportional to the sound pressure level in the 1-4 kHz octave bands. At very low ambient noise levels or when DC power appears at the control input indicating activation of the announcing system, the DC output voltage is held constant.

SLM control circuitry, in the form of a hold switch 30 and a hold capacitor 32 continuously sense the DC control voltage, as long as the switch 30 is in the closed position. With the switch 30 closed, the hold capacitor maintains a charge, or control voltage, that is proportional to the ambient noise level sensed by the sensing microphone 18. The charge changes according to the change in ambient noise levels while the switch 30 is closed. Upon opening of the switch, such as upon commencement of an announcement or delivery of the common signal input signal, the hold capacitor 32 contains a charge representative of the ambient noise level immediately prior thereto.

The second main NCA element is a variable gain audio amplifier or voltage-controlled-amplifier (VCA) 34. The VCA 34 accepts the fixed level, compressed signal from the compressor/equalizer at a VCA audio input. The VCA applies ambient noise controlled gain to the fixed level, compressed signal. Gain is applied proportional in decibels to the DC voltage supplied to a VCA control input by the hold capacitor 32 of the SLM 24. During an announcement the gain is kept constant because the SLM DC output is held constant due to the fact that the hold switch 30 is open and the charge on the hold capacitor 32 is fixed at the level it was when the switch 30 was opened. At very low levels of ambient noise the gain is also constant. The NCA audio output, in the form of a signal that has been compressed, equalized and amplified or adjusted in accordance with the ambient noise signal, is fed to other components resident at the master loudspeaker.

Referring now to Fig. 3, the NCA output signal is fed as input to a 20 watt audio power amplifier 36 resident in the loudspeaker enclosure. One general concept employed for the master loudspeaker enclosure layout is to include virtually all of the system electronics, except for the compressor/equalizer circuitry, so as to provide a self contained unit. Thus, the NCA 24 and the 20 watt power amplifier 36 are incorporated into the master loudspeaker enclosure along with a matching auto-transformer 38, and a loudspeaker transducer 40.

The autotransformer 38 provides impedance matching between the 20 watt amplifier 36 and the loudspeaker transducer 40, and is intended for mounting with each loudspeaker transducer in the system whether used as a master or slave. Only master loudspeakers include a resident 20 watt power amp, so the purpose of the autotransformer is to match the total loudspeaker load impedance seen by the master power amplifier for optimum power output and freedom from overload. Up to four slave loudspeaker units can be operated from the master, in this illustrative embodiment, each of which includes a transducer and autotransformer only.

A second important function of the autotransformer 38 is to permit adjustment of output power from each loudspeaker to the desired level. Power taps are provided in 3 decibel steps over a range of 15 decibels. The autotransformer is a single-winding tapped transformer. The advantages that accrue to use of a single-winding tapped transformer, compared to a dual winding transformer, include less insertion loss (internal power losses) and a more compact size.

The loudspeaker 40 is a directional re-entrant type exponential flare horn loaded reproducer, with a single voice coil and compression loaded driver (not multi-way) . The choice of transducer is primarily a function of the particular application and the dimensional requirements of the enclosure. The same transducer type may be used for both master and slave assemblies.

In an alternative embodiment, the concept of simplexing is implemented, integrating alternating current and direct current for transmission over a single wire pair. Accordingly, appropriate simplexing circuitry is effected to transmit to the master loudspeaker enclosure both the input audio signal and the d.c. power required to drive the NCA and associated loudspeaker level adjustment circuitry. In such a system, the NCA would include circuitry required to decouple the a.c. and d.c. signals. Such an implementation permits retrofitting of the distributed automatic level adjustment capability according to the invention into existing 70.7 volt or other constant voltage-type announcing systems.

Referring now to Fig. 4, the concepts of the technology described hereinbefore are applicable to 70 volt announcing systems presently used in commercial industry. Technology, according to the invention, can be implemented or retrofitted into current 70 Volt loudspeakers as either internal modifications (i.e., circuit card, microphone, power source) , or as an external module containing the required elements which could be located adjacent to each loudspeaker. The important design characteristics of the automatic level control circuit are the total adjustment range required and the resolution of the level control. The adjustment range is set by the maximum variability of ambient noise at the locations of the loudspeakers. The audio level resolution is set by acceptable variation in signal loudness or intelligibility. A typical 70 volt system such as known in the art may require a 40 dB adjustment range with a maximum of 3 db control increments. The circuitry for retrofitting automatic level control consists of a sound level meter circuit, an audio level control technique and a power source.

The sound level meter circuit is to measure and process the ambient noise to provide a control signal for the loudspeaker gain control. The Noise Control Amplifier described contains such a circuit which is applicable to the 70 volt system. This circuit consists of a dynamic microphone, a microphone preamplifier circuit, a three octave band filter which filters the ambient noise such that the loudspeaker gain is sensitive in only the 1 kHz to 4 kHz octave bands which are important for speech intelligibility, and a rectifier circuit which produces a DC control voltage proportional to the sensed ambient noise. This circuitry can be readily implemented into any 70 volt design, but would require continuous power. Power can be supplied to the circuitry locally from a 115 VAC source and with the use of a DC power supply. The required DC power 50 can be placed on the audio line 52 at the central amplifier(s) 54, which is part of the existing announcing system, and filtered from the audio signal prior to the loudspeaker. Simplexing techniques can be implemented as discussed hereinbefore, and the DC power used to power both the SLM 65 and the ambient noise sensing microphone 58. The SLM circuitry provides a DC output that is provided, according to logic circuitry 60 which may be required depending upon the implementation, to level control circuitry 62 that attenuates audio power so as to provide the necessary output signal to the loudspeaker 64.

The system described hereinbefore with respect to Figs. 1-3 provided variable gain to the audio signal prior to the local 20-watt amplifier contained in each high power loudspeaker. A 70 volt announcing system does not use a distributed amplified announcing implementation. The 70 volt system provides a high power audio signal to each loudspeaker and uses matching transformers to determine the fixed level of each loudspeaker. As a consequence of this different audio amplification philosophy, the 70 volt system requires an alternate level control or power attenuation technique.

In one embodiment, a power MOSFET circuit in the form of a voltage controlled power MOSFET circuit is placed in series with the 70 volt loudspeaker. The MOSFET circuit provides variable power dissipation controlled by the SLM DC control voltage. Power MOSFETs known in the art can be implemented without elaborate logic circuitry. Power dissipation requirements would be a function of system power considerations.

Alternatively, the audio level at the loudspeaker can be controlled by inserting a power resistor network in series with the loudspeaker and using a logic circuit to switch different resistors in or out of the circuit based on the SLM DC control voltage. During high ambient noise conditions, full available power would be delivered to the loudspeaker; as the ambient noise decreased, power resistors would be switched into the audio path to dissipate the unwanted audio power. This technique requires several high power resistors and a logic circuit to provide the required switching.

While MOSFET circuitry or a power resistor network can be implemented, it should be appreciated that other means can be implemented for dissipating or attenuating audio power to effect level control. For instance, a multi-tapped matching transformer can be configured to have the desired audio level adjustment range and resolution. Audio level control would be implemented through switching the transformer output taps based on the SLM DC control voltage level.

Similarly, the incoming audio signal could be digitally sampled at a high frequency and the audio signal gated off for a greater or lesser time as a function of the ambient noise level. The audio signal would be gated off for longer periods at lower ambient noise levels to reduce the power delivered to the loudspeaker. In the illustrative embodiment of the invention described herein, an appropriate ambient noise sensing microphone is a miniature, low impedance, dynamic, omnidirectional pressure sensing microphone. It could be recessed within the loudspeaker amplifier assembly cover. It should be appreciated that the microphone could alternatively be extended to another mounting location should the ambient noise at the amplifier be deemed not representative of noise levels near listeners.

The primary performance requirements of the sensing microphone in this illustrative embodiment are smooth frequency response throughout the speech frequencies and stable sensitivity from unit to unit so that the output DC voltage of the SLM is uniquely determined by the ambient noise level. However, it should be appreciated that other performance requirements may be dictated by the particular application. For instance, the microphone may be "hardened" to withstand large amounts of shock and vibration, or it may be required that it have a flat frequency response at other frequencies.

Although the illustrative embodiment described herein shows two slave loudspeakers it should be appreciated that a plurality of slave speakers lesser or greater than two, or none at all may be implemented in a system according to the invention.

While the signal output from the compressor in the illustrative embodiment discussed herein is passed to an equalizer in which adjustments are provided for speech frequencies from 315 Hz to 5000 Hz in 1/3 octave bandwidths and wherein the range of adjustability for each band level is 24 dB, it should be appreciated that for alternative implementations wherein speech is not the primary input, different adjustment bands may be implemented. For instance in an implementation wherein music is the primary input, the desired frequency bandwidth might be 20 Hz to 20000 Hz, or greater, with significantly different bands for adjustment and range(s) of adjustability.

Although output from the sensing microphone, as described in the illustrative embodiment, is filtered by a 1-4 kHz octave band filter so that the SLM output is sensitive in the frequency band which contributes most to speech intelligibility, it should be appreciated that other filter bands can be implemented to tailor the system to the types of ambient noise and audio signals processed.

It should be appreciated that while the electronics are described as being included in a self contained loudspeaker enclosure, excluding the compressor/equalizer, portions of the electronics can be implemented outside of the loudspeaker enclosure, and where it is possible to deliver adequate power to the power amplifier and compressor/equalizer circuitry, the compressor equalizer circuitry may be contained in the loudspeaker enclosure as well.

Further, while a 20 watt amplifier is described herein, other power amplification levels can be implemented. Likewise, the hold switch and hold capacitor functions described herein can be alternatively implemented, such as by sample and hold circuitry of analog or digital nature. It should be appreciated that the ambient noise sensing and gain control circuitry may be designed to operate either between message input signal intervals or continuously, including during message input signals. For example, the input voltage to the loudspeaker may be compared to the sensing microphone output voltage by a suitable comparator circuit. When there is a match implying uncontamination of the microphone pickup by local noise, the audio signal gain would be reduced until a given amount of mismatch occurs, implying a predetermined signal to acoustic noise ratio. Conversely, if a large mismatch occurs indicating an inadequate signal to noise ratio, the signal gain would be increased until the predetermined mismatch results. An obvious advantage to the continuous sensing method is that the signal to noise ratio is maintained even if the ambient noise changes during a message.

Although the invention has been shown and described with respect to illustrative embodiments thereof, it will be appreciated that various other changes, additions and omissions in form and detail thereof may be made therein without departing from the spirit and scope of the invention.

Claims

WHAT IS CLAIMED IS:

1. An audio loudspeaker system having distributed automatic level control for automatically adjusting output levels of a loudspeaker, comprising:

an input device providing an unconditioned input signal to said audio loudspeaker system;

compressor circuitry receiving said unconditioned input signal from said input device and providing constant voltage control to said unconditioned input signal to compensate for acoustical variations in said input signal and producing a constant level output signal;

equalization circuitry receiving said constant level output signal from said compressor circuitry and providing frequency adjustment of said constant level output signal and thereby producing a compressed/equalized signal;

a microphone sensing ambient noise proximate to said loudspeaker and generating a noise signal; and

noise controlled amplifier circuitry receiving said compressed/equalized signal from said equalization circuitry and said noise signal from said microphone, said noise controlled amplifier circuitry applying gain to said compressed/equalized signal in accordance with said noise signal to provide a constant signal level for broadcasting over said loudspeaker for a duration of a signal broadcast.

2. The system of claim 1 wherein said compressor circuitry includes a switch for selecting baseline gains.

3. The system of claim 1 wherein said equalizer is configured to provide adjustments in a speech frequency range of approximately 315 Hz to 5,000 Hz.

4. The system of claim 3 wherein said adjustments are provided in 1/3 octave bandwidths and each band level is adjustable 24 dB.

5. The system of claim 1 wherein said noise controlled amplifier circuitry includes a band pass filter for filtering said noise signal to provide sensitivity in a selected frequency band.

6. The system of claim 5 wherein said selected frequency band is 1 kHz to 4 kHz.

7. The system of claim 5 wherein said noise controlled amplifier circuitry includes a rectifier receiving a band pass filter signal from said band pass filter for rectifying and generating a D.C. control voltage proportional to sound pressure level in said selected frequency band.

8. The system of claim 1 wherein said system includes control circuitry controlling said noise controlled amplifier circuitry to sense ambient noise levels proximate to said loudspeaker and control signal gain correspondingly so that a signal to ambient noise ratio is maintained substantially constant.

9. The system of claim 8 wherein said control circuitry includes a switch and a capacitive storage element.

10. The system of claim 1 further comprising at least one slave loudspeaker that has its output adjusted in accordance with said constant signal level from said noise controlled amplifier circuitry.

11. The system of claim 1 further including a loudspeaker enclosure having disposed therein said loudspeaker, said noise controlled amplifier circuitry and an audio amplifier.

12. The system of claim 11 further including an auto- transformer disposed in said loudspeaker enclosure.

13. A distributed self-adjusting audio loudspeaker system having distributed automatic level control for automatically independently adjusting output levels at respective master loudspeakers, said system comprising:

at least one input device receiving a signal to be broadcast;

a plurality of master loudspeakers for broadcasting said signal received at said input device, each of said plurality of master loudspeakers having a respective loudspeaker enclosure and a respective loudspeaker transducer;

a plurality of noise sensing microphones, each of said plurality of noise sensing microphones being located proximate to one of said plurality of master loudspeakers and being associated with said one of said plurality of aster loudspeakers to sense ambient noise thereabout and to generate a noise signal indicative of said ambient noise;

noise controlled amplifier circuitry disposed proximate to each of said respective loudspeaker enclosure, and receiving said signal as an input signal, and receiving said noise signal from a respective one of said plurality of sensing microphones, said noise controlled amplifier circuitry applying a gain to said input signal in accordance with said noise signal to provide a constant signal to ambient noise ratio message signal for broadcasting over said loudspeaker;

wherein volume of a signal input at said at least one input device is adjusted at each of said plurality of master loudspeakers independently in accordance with said noise signal indicative of said noise signal from said respective one of said plurality of sensing microphones.

14. The system of claim 13 further including compressor circuitry receiving said input signal from said input device and providing compensation for acoustical variations in said input signal and producing a constant level output signal.

15. The system of claim 14 further including equalization circuitry receiving said constant level output signal from said compressor circuitry and providing level adjustment over at least one selected frequency band of said constant level output signal and producing a compressed/equalized signal to provide to said noise controlled amplifier circuitry as said input signal.

16. The system of claim 13 wherein said noise controlled amplifier circuitry receives DC power provided via a simplexed signal including said DC power and said input signal.

17. The system of claim 13 further including at least one slave loudspeaker, receiving said constant level message signal at a level determined by one of said plurality of master loudspeakers.

18. A system for distributing an audio signal to a plurality of locations subject to differing ambient noise levels, said system comprising:

a plurality of master loudspeaker units associated with respective ones of said locations, each of said units including at least one loudspeaker and a microphone for sensing noise at the respective location;

a signal input microphone; and

a signal preamplification and conditioning circuit responsive to said signal input microphone for supplying to said master units a common signal input signal, each of said master units including means, controlled by the respective noise sensing microphone, for adjusting the relative gain applied to the common signal input signal to a respective value corresponding to the local noise level and for applying the gain adjusted input signal to the respective loudspeaker.