DE3612708A1 - MICROPHONE SYSTEM - Google Patents

Description

The invention relates to a microphone system in the preamble of claim 1 mentioned type. It relates in particular to an automatic microphone control that is used for Sound amplification, recording, broadcasting, teleconferencing and other use cases can be used.

Because of the number of participants involved or the number of locations that require sound recording often used many microphones in, for example, churches, courtrooms, broadcast studios, plenary halls, and conference rooms and especially at distance conferencing. The outputs of these microphones are usually combined in a mixer and then into a sound amplification system, recording device, or transmission link fed to a remote location. If a conventional mixer with many microphones is used, the room noises become and amplifies reverb pickup compared to a single microphone; in addition, the tendency will to acoustic feedback, although typically only one microphone or several microphones provide an acoustic Record useful signal (speech) at any time.

This is detailed in the essay "Direction-Sensitive Gating: A New Approach to Automatic Mixing" by Stephen Julstrom and Thomas Tichy, JAES, Vol. 32, No. 7/8, 1984, July / August.

SL / N

-2-

Automatic mixers have recently been used to only switch signals from microphones through the mixer that pick up useful acoustic signals. The relative effectiveness of these mixers is primarily one Function of the means used to decide when a microphone should be switched through. A microphone should be quick and independent when using usable language input a wide dynamic range can be switched on. However, it shouldn't respond to room background noise and also not on interlocutors, who should better be picked up by another microphone. For one proper functioning of many remote conferencing systems including the system with the aforementioned application should also be the room speaker that carries the audio signal from the remote location, does not affect or cause the microphone to be switched on.

A switch-on procedure that depends on whether the microphone output level exceeds a fixed threshold, does not meet any of these criteria. Various other known references suggest various switching methods before.

For example, in U.S. Patent 3,814,856, the threshold level for switching each microphone follows an estimate of the clearing background noise based on a remote recording by means of a separate, noise-picking one Microphones or based on the average of all microphones in the system. This is done to increase the switch sensitivity to improve and at the same time avoid responding to room noise.

In US Pat. No. 3,751,602, a simple one is obtained as a result Microphone switched for use as an intercom microphone (a simple remote conferencing system). Here will the noise reference value obtained from the single microphone, its level via a circuit with a slow rise and rapid waste is processed. An additional circuit prevents the microphone from being switched on for loudspeaker sound

-3 -

in most room acoustic environments.

In US Pat. No. 3,755,625, only one and only one of several microphones is turned on at any one time. To switch on (and thus to switch off the microphone that has already been switched on), the level of a Microphone a predetermined threshold and also the level of the microphone already switched on by a preselected Exceed amount, for example 3 dB.

In U.S. Patent 4,099,025, to prevent multiple microphones from being turned on with a single source during the time in which the level of a microphone exceeds a threshold value, all other microphones are taken along for the duration plus a short, additional time essentially corresponding to the travel time of the sound to the most distant one Microphone in the system prevented.

In US Pat. No. 4,090,032, a preselectable, fixed threshold value is exceeded as soon as at least the level of one Microphone exceeds it, and it turns on. The threshold value then varies between a high maximum Level and approximately the level of the switched-on microphone with the highest level. Switching on more than a few Microphones at the same time, even with many sound sources, will be suppressed under all circumstances.

In U.S. Patent 4,489,442, assigned to the assignee of the present application, each "microphone" actually consists of one Arrangement of typically two directional microphones, arranged back to back in a common housing and their levels are compared with each other. If the level of the "front" microphone exceeds the level of the "rear" Microphone by a certain amount, typically 9.54 dB, (which indicates that the sound source is within an "acceptance angle"), the signal from the front microphone is switched through.

This arrangement (which is also part of the preferred embodiment of the above-mentioned article by Julstrom and Tichy) has an effective switching threshold, the

- 4 -

is about 5 dB above the level of the room noise at the location of the microphone. This arrangement results in a direction-sensitive circuit that limits the number of microphones that switch on for individual switching sources' during they do not cause any microphone to be switched on in order to achieve the desired activation of any one to suppress other microphones for other switching sources. This arrangement enables a loudspeaker to be placed in a teleconferencing system in such a way that it does not cause any microphone to be switched on and does not noticeably switch it on for local call participants prevented.

This known principle of operation requires some care in the arrangement of the microphone and loudspeaker. It enables a single switching source to switch on many microphones with overlapping acceptance angles. Two high quality matched transmitter elements are typically required for each "microphone", though only one of the couple is ever heard. It is of particular importance that a proper switching suppression is possible through acoustically reflective objects close to the back of the microphone or by placing the microphone too far from the switching source in relation to the echo field of the room, making the microphone is prevented from accurately assessing the direction of the switching source.

The sound that a microphone "hears" in a room can simply be described as having two parts be: A direct sound that decreases in level by 6 dB for each doubling of the distance from the source, and an echo field which comes from all directions and which is essentially uniform in level over the entire room, when it falls off.

The direction-sensitive switching technology works for so long good, until the microphone is far enough away from the switching source that the echo field is in the sound pick-up of the microphone

-5-

predominates. In larger rooms this will not occur until the microphone is 1.5 m or more from the Caller is away. At this distance, however, his recording would be hollow, "clayey" or perhaps be incomprehensible. In smaller rooms, such as offices or many conference rooms, the echo field predominate at distances of 0.6 m or less, whereby Proper circuitry using directional microphone technology with comfortable speaker-to-microphone distances would be prevented. In contrast to larger rooms, however, the sound recording quality is lower in many These smaller rooms, even if they are predominantly echoing, are still understandable and subjectively acceptable due to the rapidity Fall time of the echo. The microphones would be useful if switched properly.

None of the prior art patents cited above fully addresses the difficulty of achieving maximum switching sensitivity in the presence of itself In order to achieve a changing background room noise, the many microphones are switched on for a single speaker and, if there is little mutual hindrance, to allow many speakers to use several microphones at the same time to be switched on, and all of this, although they are in an almost completely reverberated acoustic environment (e.g. smaller Space) work. In addition, sound from the loudspeaker of a teleconferencing system must not switch on microphones, although desired switching for simultaneous local speech should not be appreciably hampered, again despite a very reverberant environment.

It is therefore the object of the invention to specify an automatic microphone switching method, its sensitivity in speech is large and in which responding to changing background room noise is prevented. aside from that only the single, best-fitting microphone in the system should be able to choose a single speaker to turn on. Many speakers should be able to switch on many microphones at the same time, whereby the mutual

-6-

hindrance of switching is as low as possible. Sound of a loudspeaker of a long-distance conference system is said to be microphones can not switch on with minimal obstruction of the desired microphone circuit for those originating from the location Language. All these properties should also be used in almost completely acoustically reverberant environments in small rooms be achieved. It should be possible to create "dead zones" in a room where sound sources are switched on a microphone does not work. The invention is also intended to enable a modification in which the information for switching the microphone to control other functions, such as switching on automatic video cameras, is usable. The switching method according to the invention should also be provided with a device for automatic gain adjustment be connectable to maintain a recording of a reverberation field when the number of by far Spaced, switched-on microphones varies above zero. In addition, a modification should be possible, by means of which a device for automatic gain adjustment maintains a constant recording of a Hall field, when the number of closely spaced, switched-on directional microphones varies above zero. The switching method according to the invention should also be used in the telephone conference system according to the device mentioned at the beginning by JuIstrom and Tichy may be used to their benefits to take advantage of. According to the invention, such a telephone conference system is intended with a combined loudspeaker-microphone arrangement be usable that optimally has the properties of the switching process, simple and error-free is to be built up, improved sound absorption and production through optimized acoustic and electrical design delivers, works in almost any acoustic environment and is easy to expand.

The object on which the invention is based is achieved and the advantages mentioned are achieved by the in the characterizing part of claim 1 given teaching.

-7-

AS

The mentioned and other advantages are achieved in a preferred embodiment of the invention, the processing of the circuit uses which reproduces accurate, very rapidly changing DC voltage levels of the frequency equalized output of each microphone is responsive and in a teleconferencing embodiment the electrical drive signal used for the loudspeaker, which are equalized in the same way.

Each microphone circuit block of the preferred type 0 form continues to process its DC voltage level value via a circuit with a very slow rise and rapid decay in order to create a threshold value (NAT) adapted to the noise. This threshold represents the background noise level that is present during short pauses in speech, but does not specifically speak up Voice vibrations. To meet the NAT criterion and to effectively initiate switching, the DC voltage level value of a microphone must exceed its NAT by an amount which may be 6 dB, just sufficient to protect against incorrect switching initiation of general background noise changes.

This enables a high switching sensitivity (lower Threshold value) for speech, even in the presence of room noise, which is further achieved through the use of directional microphones is improved. In a remote conferencing system of the preferred embodiment, the microphones comprise one Room noise recording of 5-7 dB compared to a direct voice recording. Direct speech thus becomes typical turn on a microphone at approximately the measured room noise level (measured with a standard all-round noise meter).

The NAT switching criterion also works with microphones well, which are arranged in the predominantly reverberant field of a speaker, at least over the range of the built-up halls and waste values that are found in acoustically poor, small to medium-sized conference rooms, although without the particular sensitivity tip that comes with it

-8th-

with directional microphones.

(The microphones were not reverberating in the predominantly Field can be used by large rooms with a very slow rise and fall of reverb, which lead to switching problems can, as the sound recording quality even of directional microphones at the assumed distance from the Sound source would be unusable in this environment.)

To prevent several microphones from being switched on when a single speaker is present, a second one is required Criterion (MAX bus) required, which must be met in addition to the NAT criterion before a Microphone is activated. The MAX rail is a single interconnection between each microphone switch block of the system, which has a varying

is maintaining DC voltage level equal to / the maximum the varying DC voltage level value of each of the switched-on microphones and de: s by a fixed amount,

it

which is 6 dB, reduced level values of each of the switched off Microphones. In order to meet the MAX rail criterion, one circuit block of a microphone must at least temporarily "keep the MAX rail upright" (the maximum tension for ... deliver).

The microphone will only be switched on if the MAX rail and NAT criteria are met for a microphone at the same time. This simultaneity leads to an entrainment signal which can be extended by a pulse that can be erased again over a "holding time ¹¹ " which is 0.4 seconds, sufficient to bridge gaps in entrainment and pauses between words. A circuit downstream controls the rise and decay speeds of the audio signal circuit to improve the subjective effect of the circuit These drag-lengthening and switching speed control circuits are essentially identical to those described in the aforesaid U.S. Patent 4,489,442.

Although not necessary in many cases, in the preferred embodiment all microphones are identical, and

-9-

/ nt

they are operated with the same relative gains, differing only in their direction and arrangement differ relative to the various switching sources. A speaker's language reaches him nearest microphone (the one most suitable for picking up his speech) in front of any other microphone and turn it on. (If the microphones were not identical, operated at different relative gains or would be either inconveniently or very close to one another, their would be relative Levels in addition to the relative arrival times become more important, but even then a microphone capable of Recording the speaker's voice is well suited to turn on.)

The switched-on microphone immediately receives a 6 dB advantage over the switched-off microphones on the MAX rail, which would effectively prevent the activation of other microphones for this speaker. The 6 dB advantage prevents in connection with the filter time constants and characteristics of the DC voltage level value circuits the secondary microphone activation (for the speaker) itself for impulsive sound or sound that ends abruptly (later for more distant microphones), and for reverberation of sound, which all microphones are approximately the same Levels reached. The 6 dB advantage for the switched on microphone ensures that the reverberation sound is switched on a switched off microphone does not work. The filter time con Stants, which may be around 11 ms, are hardly short enough to reliably prevent abruptly ending sound from turn on more distant microphones.

A single speaker thus only switches on one microphone, but several speakers can normally speak reliably switch on several microphones when what is required to optimally pick up the speaker's voices. The application of precise full-wave rectification following attenuation of low frequencies at the extraction of the DC voltage level values of the microphone

-10-

output enables the fast filter time constant of 11 ms. Normal speech courses often have peaks, dips, Failures and pauses that are precisely represented in these rapidly varying DC voltage levels. There If several speakers cannot synchronize these changes, there are often options for switching the the microphone closest to the speaker for the current "holding" of the MAX rail, which leads to a switch entrainment, which is expanded in the manner described above. The rapid DC level changes of each voice are usually much larger than the differences in mean level between individual voices. This offsets the switching of a microphone of a softer speaker in a position to maintain the MAX rail and turn on his microphone even though a louder speaker is turning on another microphone at the same time. this is not noticeably changed by the 6 dB advantage that already switched on microphones have on the MAX rail, which is small in relation to the changes in microphone level. The 6 dB advantage, however, means that a speaker tends to have an already switched on microphone switched on when it is only a little closer to a microphone that is switched off.

To hear sound from a long distance conference speaker attached to it To prevent a microphone from being turned on, a "speaker suppression signal" is generated. The speaker suppression signal is a very rapidly changing DC voltage level value (essentially identical in character to the already discussed) of the electrical driver signal for the loudspeaker. This is sized to be approximately represents the worst (highest) DC voltage level value received by the microphones due to the sound of the Speaker is expected.

A speaker suppression signal like this can be used in a directional microphone level comparison or to the MAX rail can be applied to the switching of the microphone

-11-

-11-

from the direct speaker to the microphone noise. This would have a minimal effect on a desired one Microphone circuit for simultaneous local speech (necessary to interrupt distant speakers) due to of the fast time constants, as discussed earlier in connection with the switching of multiple microphones has been. In most practical rooms, however, there would be sufficient energy in the reverb decay of the loudspeaker sound are available to effect the microphone switching after the speaker suppression signal has disappeared is when it is laid out in this way. The fall time of the loudspeaker suppression signal would be an extension require several tenths of a second or more (possibly much more in particularly bad rooms), to prevent wrong microphone switching. This would eliminate the rapid level change in the signal and make interrupting an incoming speaker by turning on a microphone much more difficult. Instead, the speaker suppression signal is used its rapid level changes are fed to the NAT of each microphone in a special way. The NAT of every microphone can be "sustained" (with the same meaning as used in connection with the MAX rail either through its own action or through the loudspeaker suppression signal, whichever is greater is. The speaker suppression signal is dimensioned appropriately, to switch on the microphone due to the worst case of the coupling between loudspeaker and microphone to prevent, and fast time constants are maintained in order to suppress the desired microphone circuit for excellent interruptibility.

However, the NAT level is not exactly the maximum of the loudspeaker suppression signal and the original NAT level. When a speaker cancel signal drops, the modified NAT level is not what it would have been before the loudspeaker suppression signal was applied.

-12-

-Zq-

The modified NAT follows the drop in speaker suppression level until it reaches the level of the microphone output, and from there it follows it the latter signal in normal NAT fashion. In the absence of local language, the acoustic input signal is the Microphone, which is immediately connected to the disappearance of the loudspeaker suppression signal connected, the reverberation decay of the sound from the loudspeaker. This does not cause any microphone switching, as the 6 dB microphone level to NAT level criterion is not fulfilled, even with unequal reverberation declines, due to the built-in dependency on the actual reverberation decay of a room, the speaker canceling function performed in this way eliminates the need for built-in safety conditions related to the worst and least predictable room acoustics encountered Conditions and enables a sensitive, local interruption possibility to be maintained for the entire range of usable room acoustics. The speaker suppression function can be more advantageous Can be used with several microphones in conjunction with the NAT and MAX rails or with a single one Microphone in connection with NAT without the MAX rail connection, as described below.

While the microphone switching scheme described here in conjunction with a large number of additional circuits can be used to complete the remote conferencing system, its properties come to their best advantage when used as part of a system as described in the aforementioned Julstrom article. This enables the advantages to be exploited of the system including total system gain and negative feedback control with more sensitive, faster, more conversation-oriented Send / receive direction switching in smaller, acoustically poorer rooms with less restrictive ones Construction requirements.

-13-

yr-

The preferred embodiment described here is a small, upward facing loudspeaker in an ordinary simple housing or module with three outward-facing, surface-mounted directional microphones combined, which are equally spaced around the circumference. When used in this way The acoustic design can be optimized and a smooth voice recording and playback with wide range can be optimized without the frequency response deviations that usually occur with acoustic transducers in the Near reflective surfaces are used. The loudspeaker radiates its sound uniformly to all conference participants around the module. In combination with the switch-on procedure described, the microphones take Speech is essentially uniform around the module with a noticeable reduction in room noise and the Reverb pick-up compared to a single microphone with an omnidirectional characteristic. In addition, the directional microphones are so arranged around the loudspeaker that they reject the loudspeaker sound as much as possible. Even though Loudspeaker sound strongly reflected by similarly arranged objects Must be taken into account, this allows a decrease in the level of the speaker canceling signal and the setting of the suppression controlling the negative feedback (described in the aforementioned Julstrom article) compared to what would otherwise be required with such a close transducer spacing.

The speaker-microphone modules can be used individually or in a "daisy arrangement" to achieve better acoustic coverage of longer tables or many tables. An interconnection between Modules is made by a single, multi-core cable, which, among other things, the switched microphone outputs, the MAX rail and connects the drive signals of the power amplifier for the loudspeaker.

-14-

A module can be muted at the discretion of the user which prevents audio signals from its microphones from being switched on, but allows that its circuit acts in the normal way in MAX-rail interactions, as if the microphones were switched would. This creates a dead zone around the muted module, where high pitched speech and other unwanted sounds (Paper crackling etc.) cannot cause the microphone to be switched on and thus not interrupt a conference call. This muting effect can of course also be used with individual microphones.

One variation that can be used in conference calls or in more general applications is to replace the arrangement of the microphones away from each other several directional microphones close together so that Essentially sound waves from different directions at the microphones over a large part of the audio frequency range arrive at the same stage. Under this condition, new polar directional patterns are formed, if different Turn on combinations of microphones in the array. As a result, the arrangement forms a single one Microphone, the polar directional diagram and orientation of which is automatically adapted to the recording of those around it to optimize the arranged speaker.

The invention is to be explained in more detail with the aid of the drawing.

Fig. 1 shows schematically and in perspective a remote conference system for use in the present invention,

Fig. 2 is a plan view of a preferred embodiment of an arrangement according to the invention of microphone and loudspeaker, Fig. 3 is a side view arrangement of loudspeaker and microphone of Fig. 2, Fig. 4 is a block diagram of the teleconferencing system

according to Fig. 2,

-15-

Figure 5 is a block diagram of the microphone speaker circuit the circuit according to FIG. 4,

FIG. 6 is a block diagram of the circuit part for the microphone circuit of the circuit according to FIG. 5; Fig. 7-13

schematically show the circuit for switching on the microphone according to FIGS. 6, and

14 shows microphone directional diagrams obtained with a Modification of the invention can be achieved.

Referring to Figure 1, there is one embodiment of a teleconferencing system shown. The system has a microphone / speaker unit 11 which is approximately 30 cm in diameter and which is arranged in the center of a conference table 30. The microphone / speaker unit 11 is provided with a control unit 15 of the system is connected by a cable 36 which runs through Holes that are drilled in the table or that may lie on top of the table.

The control unit 15 is connected to a wall socket 19 of a telephone line via a conventional telephone wire connected with the usual "two-wire" connections. A telephone 17 is connected to the control unit by means of a telephone wire 13 15 connected. The user is able to automatically select either telephone 17 or microphone / loudspeaker unit 11 and the associated circuit in control unit 15 to the Wall socket 19 of the telephone line.

Albeit shown as a two-wire connection (combined Send / receive path), the system can use a Four-wire connection (separate transmission and reception paths) can be used, as in the above-mentioned article by Julstrom is described. The four-wire connection can be a A wired connection or a radio or satellite connection that has a time delay. The radio link may be that of a mobile telephone system, although a more suitable physical arrangement of the microphone or the Microphones and the speaker would be used than the one described below.

-16-

As shown in Figures 2 and 3, the microphone / speaker unit 11 three outward-facing, unidirectional sensitive (with heart-shaped characteristics) Microphones 21, 23 and 25, equally spaced around the perimeter of the base 27 of the microphone / speaker unit 11 are arranged. The base 27 has a flat, generally conical shape as shown. A Loudspeaker 29 is arranged in the acoustic zero of the three microphones 21, 23 and 25 and directed upwards as well as in the mounted flat on the base 27 in the manner shown. Base 27 preferably rests on top of the conference table

With such a holder, the microphones 21 work, 23 and 2 5 very close to the surface with minimal acoustic interference from the base 27. The loudspeaker 29 is operating essentially as if it were aligned in a large, flat surface. Like that in the As described in the literature, microphones and loudspeakers can then produce smooth, uniform frequency responses over the speech range without the usual adverse effects of reflections the table surface. In addition, the directional indices of both microphones and of the speaker are up by an amount increased by almost 3.0 dB (compared to free space), whereby the clarity of the sound recording and reproduction is increased. Horizontal distribution of the loudspeaker sound is shown in Mainly achieved by its small size (about 5 cm in the cone diameter), which reduces the formation of lobes will.

Each of the microphones 21, 23 and 25 reduce unwanted ones Sound recording (room noise, reverberation and loudspeaker sound) by holding it in the normally off position until needed. A microphone will automatically switched on only if it is a speech-type sound source that is not the loudspeaker, densest microphone is. A microphone is not switched on for background noise in the room or for loudspeaker sound.

-17-

-rr-

In the absence of local speech, the microphones are switched off and the acoustic coupling path from the loudspeaker is switched off to Microphone is completely interrupted. No echo of the speaker sound gets to the far rear end returned to the transmission path. The microphones switch on quickly and reliably depending on the language, without interrupting syllables or words. Due to their directional effect in connection with the acoustic arrangement of loudspeakers and microphones in the manner previously described, the microphones minimize the acoustic coupling of the speaker and recording room noise and reverberation while they are on.

Several microphone / speaker units 11 can be along of the conference table 13 can be used. Two or more microphone / speaker units 11 can be electrically be connected in parallel, as has been mentioned above and is further described below.

4 houses microphone / speaker unit 11 Microphone / speaker circuit 41, while control unit houses control circuit 43. Microphone / speaker circuit 41 receives signals from microphones 21, 23 and 25 and generates Depending on this, switched and unswitched microphone signals on associated conductors 45 and 47. The Microphone signals are picked up by the control circuit 43, to generate a transmit signal for transmission to the wall socket. The audio information coming from the wall socket 19 is received by the control circuit 43 to generate a loudspeaker signal on conductor 49 used. The loudspeaker signal on conductor 49 is picked up by microphone / loudspeaker circuit 41 in order to depend on of which the loudspeaker 29 is to be fed.

Control circuit 43 can be constructed using the circuit of the preferred embodiment shown in FIG of the cited, associated application is described. In detail, a replacement can take place in Fig. 3a of the aforementioned Registration regarding his microphones 11, microphone

-18-

Interface / circuit 353, power amplifier 321 and Loudspeaker 13 through the microphone / loudspeaker unit 19 of the present application. The closed microphone signal can be sent to conductor 358 of the aforementioned application and the unswitched signal must be applied to its conductor 1003. The loudspeaker signal can be from the output of the Mixer / limiter stage 334 of the above application was removed will. It is clear that other types of control circuit than the control circuit 43 can be used, to on the telephone line or other long distance route Send microphone signals and receive speaker signals.

Control circuit 43 as in the preferred embodiment of the subject of the aforementioned application is described by Julstrom, causes a suppression (damping) the speaker level when local language interrupts received speech by switching on a microphone, but only in to the extent that is necessary to reliably maintain the negative feedback stability. When local language is replaced by the If the remote end is interrupted, the outgoing microphone signal is suppressed and the loudspeaker is switched on with normal Level heard. The circuit of the application mentioned also determines a necessary transmission / reception direction switching for suppression to occur in an unobstructed, conventionally oriented manner. When both ends speaking at the same time, priority is given to the interrupting part, thus creating the natural interaction is retained. Either side can always "get through without screaming."

The combined, switched microphone signal from everyone Switched on microphones reaches the control circuit 43 of control unit 15 via conductor 45 Microphone signal can, for example, through a current amplifier (as in the aforementioned application) before entering a conventional hybrid circuit can be carried out to the microphone signal via wall socket 19 on the telephone line

-19-

bring to. As in the aforementioned Julstrom application described, the control current input for the current amplifier is determined by a send / receive direction decision. A possible instability in the gain of the negative feedback loop is caused by the introduction of a suppression controlled in the transmission path during reception operation and in the reception path during transmission operation.

A number of circuits currently on the Market can also be used as control circuit 43 to control the switched microphone signal Wall socket 19 of the telephone line for transmission to the to transmit the distant end of the long-distance transmission link and the received signal from the wall socket 19 for transmission to the speaker 29 to derive. The microphone / speaker circuit 41 uses a different one Switching structure than that in the aforementioned Julstrom application. This becomes clear when the microphone / loudspeaker circuit 41 is described below.

In FIG. 5, the microphone / loudspeaker circuit 41 has a gate circuit 50 made up of three identical microphone gate circuits 51, 53, 55, which are assigned to the respective microphones 21, 23, 25. A conventional one Power amplifier 59 is responsive to the speaker signal from control circuit 43 appearing on conductor 49 and produces a like signal on conductor 58, which provides a sufficient level of power to power the speaker 29. A suppression signal generator 57 takes the loudspeaker feed signal from line 58 and in response generates a suppression signal for each microphone gate circuit 51, 53, 55. Each microphone gate circuit 51, 53, 55 is constructed in the same way, and one is described below with reference to Figures 6-12. A MAX rail conductor 56 connects the microphone gate circuits, but does not need to be connected to control circuit 43. be as described below with reference to FIGS. 6 and 11.

-20-

In Figure 6, microphone 21 is electrical with a preamplifier / intermediate circuit 61 connected. After pre-amplification, the microphone signal is the output signal as unswitched Microphone signal on conductor 47, and it is switched by an Optbkopplerschalter 69 to be switched as Microphone signal to go off on conductor 45.

The remainder of the circuit of Figure 6 controls the shift of the optocoupler switch 69 by generating an optocoupler control signal on conductor 71 to the optocoupler switch.

The signal on conductor 71 is formed in accordance with (1) the pre-amplified microphone signal of its associated microphone (appears on conductor 73), (2) a signal dependent on the other microphones (appears on MAX rail line 56), (3) the speaker driver signal (appears on conductor 58) and (4) a logical mute signal (appears on conductor 75).

The microphone signal on conductor 73 enters a circuit 81 for trimming and calibration. Circuit 81 is used to its associated microphone gate circuit 51 (Fig. 5) to be calibrated so that each of the three microphone gate circuits 51, 53, 55 are identical in its comparative analysis is working.

The microphone signal is then fed into a circuit 83 which equalizes the frequency of the audio signal and rectifies. Low frequencies and to a lesser extent high frequencies are relative to the level medium frequencies decreased. Circuit 83 is also used to accurately full-wave rectify the audio signal while filtering the result. The output signal of circuit 83 is a varying DC voltage level signal that contains information about amplitude and time the occurrence of speech as well as noise in the room, which is picked up by the associated microphone 21. The output of the circuit 83 is converted into a Threshold circuit 85 fed for noise adaptation, which forms a threshold voltage level that is for the Room noise in the vicinity of the microphone 21 is representative.

-21-

Threshold circuit 85 forms the threshold voltage level, by effectively following the DC microphone signal with very slow rise and immediate fall. When the DC microphone level signal increases, will a capacitor is slowly charged with a large rc time constant, and then when the dc microphone level signal disappears, the capacitor discharges quickly at the same rate as the DC microphone level signal sinks. Because of the diagram of the common language, the resulting one is on the capacitor apparent voltage representative of the noise in the room. The threshold voltage following the noise fits constantly adapts to the background room noise, but does not drop below a certain level. Normal language will does not noticeably charge the capacitor, which is continuously reduced to the background noise level while very much short pauses in the language.

The DC microphone output signal of the circuit 83 is also fed to an attenuation circuit 87, where the DC microphone signal is attenuated by 6 dB (factor 2). The output of the attenuator circuit 87 and the output of the threshold circuit 85 are shown in FIG a voltage comparator 89 is fed. Voltage comparator 89 produces an output signal indicating whether the Speech that changes very quickly exceeds the threshold level for the constant noise in the room by 6 dB is representative. The output signal of the voltage comparator 89 thus represents an independent decision on a microphone shows that speech occurs.

Each microphone switch gate circuit 51, 53, 55 (Fig. 5) makes a corresponding decision on its respective one Voltage comparator 89 when speech from a single speaker reaches each of the microphones 21,23,25. As desired is to limit the number of microphones switched on with a single switching source, the output signal arrives of the voltage comparator 89 into an AND switch

device 91 with a second decision signal to determine whether the associated microphone is switched on via optocoupler switch 69 should be.

The second decision-making process determines which of the microphones picked up the loudest speech first. Of the MAX rail conductor 56 receives input signals representative of the other microphone signals to be included in the to use the second decision-making process. The MAX rail conductor is connected to a decision circuit 97, in which the other microphone signals are compared with the signal of the associated microphone 21.

The DC voltage mxcrophone signal from circuit 83 is first through an attenuator circuit 93 before input attenuated in decision circuit 97. Attenuation circuit 93 can, however, be actuated electrically via a conductor 95, to cancel the 6 dB attenuation when the microphone is switched on. The output signal of the surmountable 6 dB attenuation s circuit 93 arrives in decision circuit 97, which it with the comparable signals in the other microfin circuits using the MAX rail link to determine whether it is currently the largest of all such comparable signals. The signal level on MAX rail conductor 56 is determined by decision circuit 97 and the comparable circuits assigned to the other microphones are controlled so that it is equal to such a maximum is.

If both the noise adaptation threshold criteria are met (i.e. speech occurs) and the MAX rail criteria is sufficient (i.e. the associated microphone 21 instantly receives the loudest speech, somewhat modified by the cancelable 6 dB attenuation of each microphone circuit), an output take-away signal on conductor 98 is used to take a monostable switch 99 generated. The output of the monostable switch 99 actuates an optocoupler driver 101, which in turn actuates optocoupler switch 69, which the microfine signal of the associated microphone switches to conductor 45.

-23-

S / l

Monostable switch 99 has a hold time of 0.4 s after each entrainment on its incoming conductor 98. The exit of the monostable switch 99 is to the control input of the cancelable 6 dB attenuation circuit 93 via conductor 95 returned. Attenuation circuit 93 responds to the HIGH signal from monostable switch 99 and lifts the 6 dB attenuation on.

The net result of the MAX rail interaction / how it has been described above, consists in a single speaker turning on only one microphone, but several speaking speakers can usually reliably turn on multiple microphones.

The loudspeaker 29 naturally reproduces speech signals for each of the microphones 21, 23, 25. To switch the A speaker suppression signal generator is used to prevent microphone channels from passing through speech from the speaker 57 used. Suppression signal generator 57 receives the loudspeaker drive signal via conductor 58 in order to be dependent on of providing a loudspeaker suppression signal on conductor 105. Suppression signal generator 57 equalized in FIG the frequency, rectifies and filters the loudspeaker driver signal to generate a DC voltage output signal on conductor 105. The speaker suppression signal appearing on conductor 105 feeds each of the microphone gate circuits 51, 53, 55.

The loudspeaker suppression signal is in the threshold circuit 85 fed in to influence the noise threshold level, like the one above and in connection below is described with FIG. 10. A microphone switch-on as with loudspeaker sound and its reverberation · is with minimal suppression of the desired microphone circuit for local speech.

It goes without saying that when two or more microphone / speaker units 11 in series along the Are connected to the top of the conference table, all MAX rail conductors are interconnected, all loudspeaker power amplifier inputs are interconnected to be driven by the loudspeaker signal from control unit 15

-24-

3612709

and all switched microphone outputs are interconnected on conductors, as is the case for all unswitched ones Outputs on conductor 47 applies.

Suppression input 75 can be through a suitable, logisehe Circuit controlled to prevent individual micro-switching without the MAX rail conductor interactions to change, which leads to the above results.

From FIGS. 7 to 13 it can be seen that all operational amplifiers and comparators are well regulated and filtered balanced voltage sources of - 15 volts are connected, as is the state of the art. In Fig. 7 is the microphone an electret condenser microphone. The converter has a preamplifier intermediate circuit 61 connected, which has a field effect transistor impedance converter 203 (Sanyo 25K156L), converter calibration resistor R1 and series resistors R2, R3, preamplifier components R4, R5, C2, C4 and operational amplifier 207 in the circuit connection shown. Resistor R1 is sized to have an equal sensitivity every arrangement of converter, impedance converter and R1. Resistor R5 represents the gain of the A preamplifier and is a precision resistor, so that identical acoustic sensitivities from each microphone to the audio mixing rails and the circuit 81 for calibration. Resistors R7 and R8 are also Precision resistors.

The microphone signal passes through capacitor C5, photoresistor R6 (part of optocoupler 69), resistor R7 and on to the switched conductor 45 for the microphone. The microphone signal also passes through capacitor C5 and resistor R8 to conductor 47 of the unswitched microphone. The microphone signal is also fed to the via conductor 73 Transfer circuit 81 for trimming and calibration, which is described in detail in connection with FIG. The switched and unswitched conductors 45, 47 are each terminated by resistors connected to ground. They have values of 5.6 k, 1.0 k and are preferably in

-25-

the control circuit 43, so that these values do not change, by connecting several microphone / speaker units 15 together are. These terminations ensure that background noise and reverberation pick-up are essentially constant remain if the number of microphone / speaker units 11 used varies and the number of microphones switched on varies above zero. This process is described in US Pat. No. 4,489,442 and in that mentioned at the beginning Essay discussed.

In FIG. 8, the microphone signal enters the trimming and calibration circuit 81 via conductor 73. Circuit 81 has an operational amplifier 211, trimming resistor R10 and resistors R11, R12 in the circuit shown. Circuit 8 is used to remove tolerance errors of the components in order to have identical microphone input signals to the MAX connecting rail. In combination with the Trimming of the resistor R1 and the use of high quality electret transducers can reduce the switching coverage area each microphone can be precisely adjusted and controlled.

In Fig. 9, the frequency equalization and rectification circuit 83 receives the microphone signal from the circuit 81 via conductor 213. Circuit 83 increases the speech components of the frequency spectrum and attenuates very high frequencies something and low frequencies that are out of the voice band, considerable. Since, moreover, in the high-frequency Parts of the voice band contain less energy, e.g. in "S" sounds, compared to the energy in low-frequency ones Dividing speech, circuit 83 is used to increase the high-frequency components within the frequency band of Language. Overall, this greatly reduces the disruptive effects of room noise when switching and enables Use of very fast filter time constants.

Circuit 83 has an operational amplifier 215, resistors R13-R18 and in the circuit connection shown Capacitors C7-C10. Circuit 83 further comprises a pair of operational amplifiers 217, 219, which in the general

-26-

showed way with diodes D1-D4, resistors R19-R22 and capacitors C11, C12 connected to the microphone signal in a precise way to direct and filter completely the same way. Rise and fall filter time constants are the same at 11 ms. Circuit 83 provides an accurate level sense over a wide dynamic range, especially for speech sound. An operational amplifier 221 buffers the output signal of circuit 83 appearing on conductor 223.

In FIG. 10, the threshold value circuit 85 for adapting the threshold value to the noise is shown in more detail. That on The signal appearing on conductor 223 is a linear amplitude reproduction of the frequency-equalized microphone signal. These Amplitude reproduction is compared to a voltage appearing on capacitor C13, which represents room noise. The signal appearing on conductor 223 arrives at the non-inverting input of a via resistor R68 operational amplifier equipped with a field effect transistor in the input 225. If the signal at the non-inverting input changes, then capacitor C13 changes accordingly charged or discharged. Resistor R 27 has a low value and is used to support stability of the operational amplifier 225 and can be neglected in the circuit analysis. Transistor Q102 is in between the output and inverting input of op amp 225 and serves as a low leakage diode.

If the non-inverting input of the op amp 225 in voltage is higher than the capacitor voltage, so op amp 225 keeps its inverting Input at the same voltage through transistor Q102 like its non-inverting input. Resistor R23 charges capacitor C13 slowly with a time constant of 10s. When the signal at the non-inverting input of the operational amplifier drops below the voltage on the capacitor tried, so the operational amplifier discharges 225 con-

-27-

-VT-

capacitor C13 via diode D5. The op amp pulls at a suitable speed depending on the negative feedback voltage across resistor R23 down, the has a negligible voltage drop due to the low input current of the operational amplifier 225 and the low leakage current from Q102. this leads to causes discharge diode D5 to appear as a precision diode.

The noise-adjusted threshold voltage on capacitor C13 follows the voltage on conductor 223 with a slight increase and immediate drop in characteristic towards to the lowest continuous background level.

The threshold voltage appearing on capacitor C13 is buffered by an operational amplifier 227 having a field effect transistor input with unity gain. The exit of operational amplifier 227 provides a signal representative of the room noise.

As shown in FIG. 10, the 6 dB attenuation circuit is 87 formed by resistors R24, R25 running along conductor 229 a 6 dB modification of the input signal appearing on conductor 223 cause. Comparator 89 through an open collector output LM339 is formed, receives the buffered output signal of the threshold voltage at its inverting Input and the 6 dB attenuated signal at its non-inverting input. A resistor R26 creates a low Hysteresis in comparator 89 for switching stability. The output of comparator 89 thus provides a logic high signal on conductor 98 indicating that speech is occurring on its associated microphone 21. The voltage level on conductors 98 is also dependent on decision circuit 97 (Fig. 6), which will be described in connection with FIG.

In Fig. 11, the 6 dB attenuation circuit 93 is on Resistors R28, R29 formed, which along conductor 233 a 6 dB attenuation of the input signal appearing on conductor 223 cause. A field effector transistor switch 235 (p-channel, Vp 3 volts) is located between resistor R29 and ground, to effectively cancel the 6 dB attenuation when the FET switch

-28-

235 is switched off. Turning off FET switch 235 results in removing resistor R29 from the circuit.

A signal appearing on conductor 95 passes through diode D6 to turn FET switch 235 off. A resistance R30 is between ground and cathode ther diode D6 as shown, and capacitor C14 connects conductor 95 with mass. Resistor R30 and diode D6 are used to supply the correct control voltages for switching the FET switch 23 5 create. Capacitor C14 somewhat reduces the voltage transfer on conductor 95 to avoid capacitively coupled noise spikes from the 6 dB cancelable attenuator circuit 93 and others To minimize parts of the circuit.

Decision circuit 97 has an operational amplifier 237, whose non-inverting input has Conductor 233 receives the output signal of attenuator circuit 93, the inverting input of operational amplifier 237 is with the MAX connection rail via resistors R31 and R32 tied together. Resistor R32 has a low value and is used for circuit stability. Tried the tension on that non-inverting input of op amp 237 to get larger than the amplitude of the MAX connecting rail, so diode D7 is forward biased so as to "hold up" the level on the MAX link rail, which is equal to the level of the non-inverting input. Is the non-inverting input of the operational amplifier 237 is lower than the voltage on the MAX-link, the diode D7 is in the reverse direction biased. Diode D8 prevents too large a negative voltage deviation at the output of the operational amplifier 237. The voltage level of the diode D7 is monitored by a comparator 239 (LM339), which has a logical "high" indicator on its Output on conductor 98 forms only when diode D7 is forward biased. Resistance R33 creates a hysteresis to stabilize the comparator 239. Resistors R34, R35 are connected as shown.

At least one resistor R36 should be present to underneath

-29-

all conditions for diode D7, which is the MAX connecting rail high to guarantee a load and thus reliable sampling of the forward bias by comparators 23 9 to enable.

The output of comparator 89 (Fig. 10) is connected to conductor 98 (Fig. 11). Comparator 89 and Comparator 239 thus serve to form a hard-wired AND circuit (symbolically given the reference number 91 in FIG. 6) along Head 98 as its collector is open. So only if both the noise adaptation threshold criteria and the MAX connecting rail criteria are met, the resettable monostable switch 99 is actuated.

The resettable, monostable switch 99 has a Comparator 241 (LM339) with an open collector output connected to resistors R37-R41 as shown is. Comparator 241 is responsive to a logic HIGH signal at its inverting input which is about to discharge a capacitor C15 is used. A comparator 243 (LM339), which is connected to resistors R42-R47, monitors the voltage on capacitor C15 to generate a logical Output signal on conductor 95. Capacitor C15, comparator 243 and associated switching elements create a hold time of 0.4 s for the output signal appearing on conductor 95. These 0.4 s time bridges gaps in the entrainment, as described above. If the entrainment stops, it begins Charge capacitor C15, which takes 0.4 s before capacitor C15 has reached a sufficient voltage to power the output of the comparator 243 to change. If the entrainment signal disappears before the 0.4 s has expired, it becomes a capacitor C15 discharges again and the signal on capacitor 95 does not change.

The output signal at a resettable, monostable Switch 99 appearing on conductor 95 enters optocoupling strut 101 as shown in FIG.

Optocoupler driver 101 is used to provide a controlled rise and fall time for the optocoupler LED 248 and so the change in resistance of the photoresistor R6 (Fig. 7).

-30-

2g

The optocoupler driver consists of operational amplifiers 247, 249, diodes D9, D10, capacitors C16-C18 and resistors R48-R55 connected as shown. The result is that the audio signal has a Rise time of 4 ms and a fall time of 0.3 s is switched, resulting in a click-free, undisturbed switching process is.

A muting circuit has a resistor R69 and a transistor QT13 and is used to keep the microphone switching signal away without influencing the switching control circuits 85, 87, 89, 93, 97, 99 (Fig. 6). Transistor Q 113 is with its collector and emitter at the input to LED driver 101 and ground to short the input to ground during muting. Resistor R48 isolates the short circuit signal from the rest of the switching control circuit. A fitting Logic circuitry (not shown) can be used to apply a positive voltage to conductor 75 and thus Turn on Q113 and activate mute. These special type of muting enables the creation of dead zones in the manner described above.

In Fig. 10, the loudspeaker suppression signal, that appears on conductor 105 (from Fig. 13) fed into the non-inverting input of an op amp 259, to change the threshold level appearing on capacitor C13. When the voltage level is on conductor 105 is at ground (0 volts) or lower, affects the voltage on capacitor C13, operational amplifier 259 and the associated switching elements do not affect the operation of the threshold circuit 85. Diode D14 prevents excessive negative Voltage drift at the output of operational amplifier 259, which would otherwise be caused by feeding back the positive voltage on capacitor C13 to the inverting input of operational amplifier 259 would occur through buffer op amp 227 and resistor R67. When the voltage level on conductor 105 tries to exceed the voltage on capacitor C13, so op amp 259 charges capacitor C13 through diode D15 and resistor R66 to keep the two voltages equal.

-31-

3812708

The signal at the output of operational amplifier 259 also goes to the non-inverting input of Operational amplifier 225 via diode D16 to the otherwise during the time to exceed the existing input voltage when the loudspeaker suppression signal increases the voltage across the capacitor C13 controls to influence the voltage level to prevent that appears on capacitor C13. The immediate positive change in voltage on capacitor C13, caused by the loudspeaker suppression signal ,, would normally cause op amp 225 to pull in opposite direction across diode D5. By creating the Suppression signal at the non-inverting input of Operational amplifier 225 at a slightly higher voltage level than on capacitor C13 avoids this problem. These slight voltage difference is caused by resistor R66 and the higher current level at diode D15 in relation to diode D16 ensured. If the loudspeaker suppression signal occurs Conductor 105 from its peak level becomes capacitor C13 exactly in one step by operational amplifier 225 via diode D5, controlled by operational amplifier 259 via diode D16, discharged.

The voltage on capacitor C13 and the output of the Buffer op amps 227 exactly follow the voltage level and rapid rise and fall times of the speaker cancellation signal on conductor 105 until it drops below the microphone level reading on conductor 223, at which point the normal NAT effect occurs again. At this time, the microphone signal will likely decrease the reverberation of the speaker sound represent which the voltage on capacitor C13 will follow as desired.

The values of the switching elements are as follows:

Resistors resistance

R2, R46 R3

35 R4, R27

R5

-32-

8th .2 K 9 .1 K 200 470 K

R6

Capacitors

C2 C4 05, C13 07,09,010,019,021,022 08, C20 011, 012, 015, 014 016 018 023,

Resistances , R48 resistance K R7 5.1 K R8, R11 11th K R31, R67, R50, R52 100 K R12, R55, R40 , R47, R58, R59 30th K R13, R56 R61 2.7 K R14, R57 , R62, R64 1 K R15, R16, R20, R22 R6 9 51 R17, R18, R60, 510 R19, R21, R32, R66 100 K R24, R25, R45, , R68 10 M. R23, R44 2.2 M. R26, R49, R42 1.5 K R28, R29, R34, R35 20th M. R30, R33, R36 1 K R37, R38 15th K R39 1.5 R41 750 K R43 27 R53 680 R54 390 K R63, R65 24 nominally
1.2 K R1 100 K
Trim potentiometer R1O

Photocell ON resistor

Photocell OFF resistor 10M

capacity

.33 P 4.7
.15 .068 .22 .01 .1

.047 .47

Microphones can be used in separate housings and those with directional characteristics cannot be used can only be used in one direction. A loudspeaker microphone (simple remote conference system) can be connected to the control circuit in accordance with the said application and other control circuits that only use a microphone if so is appropriate. The benefit of excellent shift quality, even in the presence of incoming speech, would be maintained will. For a remote conference system with only a microphone would be attenuated circuit 93, decision circuit 97, AND circuit 91 and MAX connection bar in FIG 56 are not required.

Sound recording can also be carried out with several microphones alone without the use of a remote conference loudspeaker. In Figures 5 and 10, of course, there would be power amplifiers 59, suppression signal generator 57 and loudspeaker 29 are not required. The same would be op amp 259, diodes 014,15,16 and resistors R66, R67, R apply, with R68 replaced by a compound.

It will also be understood that in Fig. 11 the switching of the logic signal on conductor 95 is used for this purpose, for example can be used to control the other related functions, such as muting an overhead speaker the switching of an automatic video camera or speaker indicator lights. An output port 94 is used to recover the logic signal. There can be precise microphone switch coverage areas can be created without overlap.

Another modification is to use directional microphones, which are arranged very close to one another are. This would typically be an arrangement of two, three or four directional microphones (cardiac characteristic), which with operated with identical sensitivities and directed outwards in the same plane at the same angular distances from one another are. In this case, the MAX tie bar decisions are based primarily on individual sound sources Line on relative microphone amplitudes and not on relative arrival times of speech.

-34-

The directional characteristic in the horizontal plane is for an arrangement of three directional microphones in polar coordinates 14 shown. Diagrams 401, 403 and 405 clarify the relative sensitivities as the distance from the center point 411 of the three microphones (on a linear basis, not dB) with respect to sound sources at any angle around the Arrangement.

Diagram 407 shows the directional characteristic of the combination or sum of the characteristics 401 and 403. Corresponding Charts could be drawn for the sums of the other two pairs. The relative sensitivity of Diagram 407 is sized to have the same overall sensitivity how one of the individual heart microphones for background room noise and reverberation is maintained that is the same comes in all directions. The directional characteristic can be used as a wide-angle heart characteristic are designated.

Diagram 409 represents the sum of diagrams 401, 403 and 405 and is again dimensioned so that the same sensitivity for room noise and reverberation is maintained. This directional diagram is the same in all directions. In symmetrical Arrangements with two or four microphones form combinations of opposing pairs with cardiac characteristics or combinations all microphones also have an all-round characteristic. In the case of the four-microphone arrangement, forms a combination of two or three neighboring microphones with heart characteristics with different opening angles. The variety of directional characteristics and orientations as just described are achieved by using the switching method according to the present invention is applied when different numbers of speakers in different locations around the Speak arrangement in different combinations. One or more of the microphones in the arrangement can be muted to effect a directional sensitive circuit similar to that in U.S. Patent 4,489,442, but comparatively less affected by close reflections and reverberation.

It is generally desirable in a system for the automatic switching of microphones, the recording of the room

-35-

Noise and reverberation the same if the number of activated microphones varies above zero. This prevents "Pump" or "breathing" effects on the audible background noise and keep the negative feedback loop amplified constant under the assumption that the microphones are in the reverb field of the loudspeaker. Usually, and this also applies to the preferred embodiment of the The remote conferencing system of the present invention are those Microphones far enough apart so that they have random phase relationships when recording the room noise and accept the hall. In order to keep the recording of these noises constant, the gain or sensitivity should be be attenuated according to the following formula, which has become a standard of practice:

Attenuation = 10 log _{1 ()} NQM where the attenuation is given in dB and NOM is the number of activated microphones. This is equal to an additional 3 dB attenuation for each doubling of the NOM.

The required damping law is different for the closely adjacent arrangements. To the design characteristics according to Fig. 14 for the three microphone arrangement, the following relative attenuations are required:

1. Microphone ON 0.00 dB

2. Microphone ON -5.12 dB 3. Microphone ON -8.29 dB

In order to keep the recording of the reverberation noise constant in the four-microphone arrangement, of which the two-microphone arrangement is one Subgroup, the following relative attenuations are required:

1 . Microphone ON 0.00 dB 2. opposite
Microphones ON -4.77 dB 2. neighboring
Microphones ON -5.44 dB 3. Microphones ON -8.45 dB 4th Microphones ON -10.79 dB

Referring to Fig. 7, these damping characteristics can be approximated very closely by using the same mixing bar arrangement as already used in the Mixing bar assembly 45 is used for switched microphone. If the rail is completed with a resistance equal to 4.0 times the sum of the ON resistance of the photoresistor R6 and resistor R7, the result is the relative damping characteristics:

1. Microphone ON -0.00 dB

2. Microphones ON -5.11 dB

3. Microphones ON -8.30 dB

4. Microphones ON -10.63 dB

In this case the terminating resistor (not shown) of the connecting bar are 22K ohms and not 5.6K ohms.

An array can be used alone or in combination with other remote microphones or arrays. In In these cases the connection bar terminated with the 22K Ohm resistor is local within an arrangement by a Buffer isolates the mixer connection rail for the entire system of switched microphones via another Optocoupler feeds. Associated resistance values are those as originally specified for this connecting rail are. The additional optocoupler is fed whenever any microphone is in its associated arrangement is switched on.

The possibility described above allows the creation of microphones with essentially automatically variable Directional characteristics and automatically variable orientation and their inclusion in an automatic microphone switching system.

The option is not to use equipment microphones (heart-shaped) or microphones oriented in the same plane limited.

Claims (1)

Claims: 1. microphone system, with a plurality of microphones, each of which is responsive to speech, in order to record one containing the speech information, generate an electrical microphone signal, with switching means for receiving each of said microphone signals and for switching one of them Microphone signal to switch on its associated microphone depending on a switching signal, characterized in that the switching means have Monitoring means for continuously monitoring the maximum of the microphone signal amplitudes of the aforesaid Microphone signal at a level on a scale for microphone signals from activated microphones and at a level on a different scale for microphone signals from switched-off microphones, where the Monitoring means generate an electrical MAX signal representative of said maximum and each Compare said microphone signal with said MAX signal and, depending on this, a take-away signal generate that is assigned to a named microphone to switch on the assigned microphone, -2- Means for generating a switching signal as a function of an entrainment signal for generating a Switching signal of at least a predetermined, minimum, non-zero time period for switching on the Microphone that is assigned to the aforementioned entrainment signal, and Switching means which respond to a said switching signal in order to switch on an associated microphone. 2. Microphone system according to claim 1, characterized in that the loudspeaker means for Generation of sound in the area of the microphones are provided that the loudspeaker means through a loudspeaker signal are fed, and that suppression signal generator means for generating an electrical suppression signal are provided which is related to the loudspeaker signal and prevents the switching means from a to switch said microphone depending on the sound from the loudspeaker means. 3. Microphone system according to claim 1, characterized in that indicates that the monitoring means further include a threshold device for forming a threshold value and in that the monitoring means increases the amplitude of each said microphone signal with a said Compare threshold value, and that a said trigger signal is generated when a said microphone signal a predetermined relationship with said threshold and a predetermined relationship with said threshold MAX signal reached. 4. Microphone system according to claim 3, characterized in that indicates that the threshold device has means for generating a threshold signal level, which is representative of the room noise. 5. microphone system according to claim 1, characterized g e - - 3— indicates that said monitoring device furthermore has a threshold value device for generating a plurality of threshold value signal levels which are each representative of room noise, each said threshold signal level as a function of a microphone signal generated by said microphone is generated, and that the monitoring means the amplitude of each said microphone signal with a said Compare threshold value signal, and wherein said entrainment signal is generated when said Microphone signal has a predetermined relationship to a said threshold signal and a predetermined relationship reached to the aforementioned MAX signal. 6. Microphone system according to claim 2, characterized in that - indicates that the monitoring device further threshold value means for forming a threshold value have that the monitoring device the amplitude compares each said microphone signal with said threshold value, and that said trigger signal is generated when a said microphone signal has a predetermined relationship to said threshold value and a reached a predetermined relationship to said MAX signal, and that the threshold value means also means for Have setting of said threshold value as a function of said suppression signal. 7. microphone system according to claim 2, characterized in that said monitoring means threshold value means for generating a number of Have threshold signal levels each representative of room noise that each of said number microphone signal generated by threshold signal levels as a function of / a named microphone, and that the monitoring means match the amplitude of each said ai microphone signal with a said threshold signal compare, and that a named entrainment signal -4- is generated when a said microphone signal has a predetermined relationship to a said threshold value signal and generates a predetermined relationship with said MAX signal, and that the threshold means means for Have adjustment of said threshold signal level as a function of said suppression signal. 8. Microphone system according to claim 4, characterized in that said threshold value means record a microphone signal and adjust its amplitude by generating a said threshold value signal level address slow rise and rapid fall. 9. Microphone system according to claim 2, characterized in that that each of said number of microphones is in a fixed relationship with respect to one another are held and that said loudspeaker is held in a fixed relationship with said microphones is. 10. Microphone system according to claim 9, characterized that the said microphones are directed outwards from one point and are the same Be spaced on a circle of which the said point is the center, and that the said loudspeaker is directed upwards from said point. 11. Microphone system according to claim 1, characterized in that - indicates that exit gate means to provide of said switching signal are provided as an output signal. 12. Microphone system according to claim 1, characterized that the said switching signal generating means for muting serve control means which respond to an electrical signal for muting a microphone that is switched on. -5- 13. Microphone system according to claim 2, characterized that several loudspeakers are provided in a daisy arrangement. 14. Microphone system according to claim 1, characterized in that - indicates that automatic gain adjustment means to adjust the gain or the sensitivity of the mentioned, switched-on microphones depending on the number of switched on microphones are provided to record the total of the reverberation field to keep the switched-on microphones constant, and that the said microphones are arranged at a large distance from one another are. 15. Microphone system according to claim 1, characterized by automatic gain adjustment means for Adjustment of the gain of the mentioned, switched-on microphone depending on the number of switched-on Microphones to keep the recording of the entire reverberation field constant by the mentioned, switched-on microphones, and that said microphones are arranged in close proximity to one another. 16. Microphone system, characterized by a number of microphones, each of which is microphone is responsive to speech to generate an electrical microphone signal containing the speech information, Switching means which pick up each said microphone signal and switch a said microphone signal to switch on its associated microphone depending on a switching signal, Threshold means for generating a threshold signal level depending on room noise, Comparator means for comparing the amplitude of each said microphone signal with said threshold value, -6- said switching means for generating a switching signal when said microphone signal has a predetermined one Relationship with respect to the threshold signal level reached, Loudspeaker means for generating sound in the area of said microphones, the loudspeaker means are fed by a loudspeaker signal, Suppression signal generator means for generating a electrical suppression signal in relation to said loudspeaker signal to said switching means to prevent a named microphone from being switched depending on the sound from the named loudspeaker means, wherein the suppression signal modifies the threshold signal level to a microphone circuit for direct To suppress loudspeaker noise and wherein said threshold signal level follows said microphone level, to suppress microphone switching for loudspeaker reverberation. 17. Microphone system, characterized by a microphone that responds to speech and a speech information generated containing electrical signal, Switching means that pick up said microphone signal and switch said microphone signal to its associated Switch on the microphone depending on a switching signal, Threshold means for generating a threshold signal level depending on room noise, Comparison means for comparing the amplitude of said microphone signal with said threshold value, said switching means for generating a switching signal, when said microphone signal reaches a predetermined relationship with the threshold signal level, - 7 - by loudspeaker means for generating sound in the area of said microphone, the loudspeaker means are fed by a loudspeaker signal, by suppression signal generator means for generating an electrical suppression signal in relation to the said loudspeaker signal to prevent the said Switching means for switching said microphone as a function of sound from the loudspeaker means, wherein the suppression signal is said threshold signal level suppressed to prevent a microphone circuit for direct loudspeaker sound, and where the called threshold signal level follows the microphone signal level to a microphone circuit for loudspeaker reverberation to prevent. 1518. Microphone system, characterized by a noise adjustment threshold circuit comprising (i) Means for generating an amplitude signal representative of the amplitude of a microphone input signal is, (ii) capacitive means of storing charge to generate a threshold voltage level, (iii) amplification means responsive to said amplitude signal and said capacitive means charge, (iv) feedback means for feeding back said threshold value to said amplifier for control the charging of the capacitive means, and (v) means for developing a signal representative of loudspeaker sound, the developing means said signal to said capacitive means and to said input of said amplification means invest. -8th - 19. Noise adaptation threshold circuit, characterized by (i) Means for generating an amplitude signal representative of the amplitude of a microphone input signal is, (ii) capacitive means of storing charge to generate a threshold voltage level, (iii) amplifier means which, as a function of said amplitude signal, said capacitive means charge, (iv) discharge means for discharging the capacitive means at a rate substantially equal to that The rate at which the said amplitude signal decays, (v) Means of measurement for generating a representation of the said amplitude signal with a different scale, and (vi) comparison means for comparing said threshold voltage level with the said, scaled reproduction of the said amplitude signal.