Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.

Patents

  1. Advanced Patent Search
Publication numberUS3660603 A
Publication typeGrant
Publication dateMay 2, 1972
Filing dateAug 28, 1969
Priority dateAug 29, 1968
Also published asDE1942814A1, DE1942814B2
Publication numberUS 3660603 A, US 3660603A, US-A-3660603, US3660603 A, US3660603A
InventorsAndersen Bjorn
Original AssigneePhilips Corp
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Speech-controlled bilateral amplifier
US 3660603 A
Abstract
A speech-controlled bidirectional amplifier that includes first and second transmission channels each with a control member controlled by speech signals derived at control points preceding the control member. First and second comparison devices responsive to the levels of the speech signals control a third comparison device that in turn controls the attenuation level of the control members. The amplifier is arranged so that in the rest condition both control members provide a high signal attenuation. When speech signals are transmitted through one channel, the other channel is cut-off. However, an interrupt facility is provided whereby a speech signal above a given level in the cut-off channel will cause the apparatus to switch over so that the cut-off channel is opened and the previously open channel is then cut-off.
Images(1)
Previous page
Next page
Claims  available in
Description  (OCR text may contain errors)

I United States Patent [151 3,660,603

Andersen 1 1 May 2, 1972 [54] SPEECH-CONTROLLED BILATERAL FOREIGN PATENTS OR APPLICATIONS AMPLIFIER 668,858 4/1949 ore-d1 1111121111 ..179/1 vc [72} Inventor: Bjorn Andersen Oslo Norway 1,281,499 3/1966 Germany ..179/170.6

[73] Assignee: U.S. Philips Corporation, New York, NY. p i E \,aminer Kathleen Cl ff [22] Filed: Aug. 28, 1969 Assistant Examiner-Jon Bradford Leaheey Attorney-Frank R. Trifari [21] Appl. No.: 853,881

[57] ABSTRACT [30] Foreign Application Priority Data A speech-controlled bidirectional amplifier that includes first Aug. 29, 1968 Norway ..3369/68 an econd transmission channels each with a control member 7 controlled by speech signals derived at control points preced- [52] US. Cl. ..179/1 VC, 179/1 H ing the control member. First and second comparison devices [51] Int. Cl. ..H03g 3/24 responsive to the levels of the speech signals control a third Field of Search 1 VC, 1 F5, 170 R, comparison device that in turn controls the attenuation level 179/170 B, 1 31 170-6 of the control members. The amplifier is arranged so that in the rest condition both control members provide a high signal [5 6] References and attenuation. When speech signals are transmitted through one UNITED STATES PATENTS channel, the other channel is cut-off. However, an interrupt facility is provided whereby a speech signal above a given level 10751045 1/1963 Clemency 179/1 HF in the cut-off channel will cause the apparatus to switch over 3,113,181 12/1963 Soderbaum ..179/l70 so that the cut-off channel is opened and the previously open channel is then cut-off.

9 Claims, 1 Drawing Figure DEVICE FR CONTROL AWL DEVICE AMPL. SPEAKER M 1, R 1 1 1 1 r 1 1 I 1 1 1 AMPL. 1 r PR2 1 1 RECTIFIER, l 1 ND 1 i 1 a $2 2 1 1 1 1 1 GATES D I l a 115 l N0 0 g COMPARISON I 1 l 1 1 1 1 1 l l- PATENTEIIIIII 2 I972 3. 660,603

CONTROL AMP! DEVICE AMPL. SPEAKER I I R I 2 HI r'b I I I I I I i AM L. AMP I 5 1 my I I RE TIFIER 4 RECTIFIER I DELAY I M I DB9 I I H Q I I I 0A1 51 g GATE K I I D2 I I I I B E3 m I 2 I I C2 S2 0A2 I COMPARISON I DEVICE I I l PR1 FKZ I I l I I I l I I l INVENTOR. BJORN ANDERSE N SPEECH-CONTROLLED. BILATERAL AMPLIFIER The present invention relates to a speech-controlled bilateral amplifier in which each amplifying channel includes a control member governed by a control circuit to which speech signals are applied that are taken from the two amplifying channels at points preceding the control member of each channel. I

Such amplifiers may be employed with loud-speaking telephones or so-called intercom systems which allow a conversation in two directions. The inputs of the amplifiers are connected to microphones and the output to loud speakers, the microphone of one channel being located in the same room as the loudspeaker of the other channel and conversely. In general it is required to arrange the microphone and the loudspeaker in the same cabinet so that they are positioned relatively close together. A very strong acoustic coupling between the two elements is then unavoidable.

This implies the risk .of acoustic feedback. In order to avoid same it is common practice to have the amplifier controlled by the speech signals so that the channel in use has a high amplification and the other channel is cut off or at least has a sufficiently low amplification.

Of course, this method of control has to be performed rapidly and smoothly and apart therefrom such systems have to satisfy various other requirements.

When one of the parties starts speaking the associated amplifying channel has to open rapidly, even for a low speech level, so that even the beginning of the first word will not be lost. During short intervalsbetween words or sentences of the conversation it should not be possible for the direction of transmission of the amplifier to be automatically reversed due to reverberation or cross-talk between the loudspeaker and microphone in the other'room.

It is desirable for theloudspeaker to reproduce the sound with adequate volume, for example, at the normal speech level, whereas at any instant the other-party should be able to interrupt .the first one andto reverse the direction of transmission of the amplifier without theneed for raising his voice markedly. The reversal of the channelsshould be performed rapidly so that each party is able'to make a remark during the continuous flow of words of the other party without the beginning ofthe remark being suppressed due to inertia of the interchange. In addition, the interchange of the channels should not give rise to instability.

lt isparticularly important thatthe persons speakingshould be free to move about inthe roomsso that they need not always be near the microphone-loudspeaker combination.

Known devicesdomot satisfy all of said requirements or satisfy them only to alimited extent because great difficulties arise from the fact that various requirements are to some extent in conflict with each other.

In general a loudspeaker is located nearer to the microphone than the person talking into it. When the loudspeaker reproduces the sound at speech level,-the associated microphone which, picks up the cross-talk signal can obviously provide a much stronger signalthan the microphone to which the direct speech signal is. applied.

In systems in whichspeech signals for controlling the amplification are derived in each channel at a point preceding the control members and in which .the channel to which the strongermicrophone signal is applied dominates and is driven to full amplification whereas the other channel is cut off, it is not possible, dueto the strong cross-talk between the loudspeaker and the microphone, to obtain a high acoustic amplification, that is to say, a strong sound reproduction, and moreover the parties have to remain near the microphones.

in known devicesin which the speech control signals are derived at a point beyond the control members, a strong sound reproduction can be obtainedybut it is very difficult to interrupt the conversation except in-the case of long pauses in the other partys speech.

The invention provides a; particularly efficacious solution of these problems.

The invention is characterized in that the speech signals from each channel are applied each through a first signal circuit to a first input of two comparison devices and the speech signals from each channel are furthermore each supplied through a second signal circuit to a second input of the two comparison devices. The amplification of the first two signal circuits exceeds that of the second signal circuits and the comparison devices are designed so that they supply an output work signal when the signal level at the first input is higher than that at the second input. The output signals are fed to an input of a first and a second gate circuit, respectively, in a third comparison device, while the outputs of the gate circuits are coupled with a second input of the other gate circuit in a manner such that, when a work signal is fed to one of the firstmentioned inputs of one of the gate circuits the other gate circuit is cut ofi". The control voltages for the two control members are taken from the outputs of the gate circuits of the third comparison device.

The invention will be described more fully with reference to a single preferred embodiment shown schematically in the sole FIGURE of the accompanying drawing.

The amplifier comprises two channels. In one channel the speech signal of the microphone M, is applied through the microphone preamplifier T,, auxiliary control member DR,, the control member R, and the output amplifier U, to the loudspeaker H,. The other channel is arranged in an identical manner and comprises the preamplifier T,, auxiliary control member DR,, the control member R, and the output amplifier U The auxiliary control members DR, and DR, may be omitted since they are not essential to the invention. The purpose and function thereof will be described in detail below. The microphone M, and the loudspeaker H, are located in the same room, as are the microphone M, and the loudspeaker H,. The speech signals for the control circuit are derived in the channels at points P, and P, preceding the control members R, and R,, respectively, so that these signals are not affected by the random states of the control members R, and R,

The signal at point P, is applied through a first signal circuit including an amplifier FK,, a rectifier DA, and a delay device S, to the input A, of a voltage comparison device K,. The signal at point P, is applied similarly through the amplifier FK,, the rectifier DA, and the delay device S, to the input A, of a second comparison device K Moreover, the signals at points P, and P, are each applied through a second signal circuit including the amplifier FR,-and the rectifier D8,, and the amplifier PR, and the rectifier DB,, respectively, to a second input B and B,, respectively, of the comparison devices K, and K,, respectively.

The comparison devices K, and K, are arranged in known manner so that they supply an output work signal, for example, a high output voltage, when the signal level at their first input A, add A, respectively, exceeds by a given threshold value the level at the second inputs B, and B,, respectively.

The outputs of the comparison devices K, and K, are connected to the inputs C, and C, of the gates ND, and ND, of a third comparison device. The outputs of the gates ND, and ND, are each connected to a second input D, and D,, respectively, of the other gate and are furthermore connected via the junction points P and P, to the control members R, and R of the amplifying channels.

The gate circuits ND, and ND, may be formed by Nand-circuits so that the output voltage is low only when the voltage at thetwoinputs is high. Thus, the voltage at the output is high whenthe voltage at just one of the inputs is low. The control members R, and R, are arranged so that when the voltage at the points P and P.,, respectively, is high, the control members are cut ofi or at least the amplification in the amplifying channels is so low that acoustic feedback cannot occur. In principle, the operation of the arrangement is of course not dependent upon a given polarity of the various signal voltages, neither individually nor in relationship, so that it will be obvious that the gates may be arranged differently and, if desired, voltages may be inverted by means of inverters, for example,

the control voltages applied to the control members R, and R,, provided that the arrangement as a whole continues performing the same function.

Apart from a given degree of frequency dependence, if any, the signal levels at the various inputs of the comparison devices K, and K, are proportional to the signals delivered by the corresponding microphones, independently of the state of the control members R, and R,.

In the rest position of the arrangement the microphones do not deliver a voltage and the signal levels at the inputs of the comparison devices K, and K, are low (zero level) so that the voltage of A, is not higher than that of B, and that of A, is not higher than that of B,. The voltages at the inputs C, and C, of the gates ND, and ND, will be low and the voltages at point P and P will be high. The control members R, and R, are then cut off.

When someone talks into the microphone M, the signal produced by the microphone is amplified by the amplifiers T,, FK, and FR, and rectified by the rectifiers DA, and DB, so that the signal level at the inputs A, and B, increases. In known manner, the delay device S, is designed so that the rise in the voltage level at A, is substantially not delayed. The voltage at the input C, remains low, but that at the input A, exceeds that at the input B, so that a high voltage appears at both inputs C, and D, of the gate ND, and the voltage at point P, becomes low. As a result, the amplifying channel between the microphone M, and the loudspeaker H, is opened through the control member R,.

The loudspeaker H, is located in the same room as the microphone M, and preferably is mounted in the same cabinet as the microphone so that the distance between these two elements is small. It is desirable for the sound signal delivered by the loudspeaker H, to be as high as possible, preferably at the normal speech level. Therefore, the microphone M, also receives a very strong sound signal which, owing to the nearness of the loudspeaker H,, may be many times stronger than the sound received by the microphone M,. In a practical case, as a result of the direct cross-talk between the loudspeaker H, and the microphone M,, the strength of the signal at point P, may be times that of the signal atrpoint P,. This cross-talk signal is amplified via the amplifiers T,, FR, and FK, and subsequent to rectification by the rectifiers DB, and DA, it is applied to the inputs B, and A, so that the signal level at these inputs rises. However, owing to the acoustic delay between the loudspeaker H, and the microphone M, the rise of the signal voltage at the inputs B, and A, will take place somewhat later than at the inputs A, and B, and, more particularly, not before the voltage at point P, has become low.

The ratio between the amplification by the amplifiers PK, and F K, on the one hand and that by the amplifiers FR, and FR, on the other hand is chosen so that it slightly exceeds the ratio between the signals appearing at the points P, and P, when the microphone M, is talked into. When the signal strength at point P, is 10 times that at point P,, the signal amplification by the amplifiers FK, and FK, may be 1 1 times that of the amplifiers FR, and FR,.

As a consequence the signal level at the input B, remains slightly lower than that at A, so that the low voltage at point P, is maintained.

On the other hand, since the amplification by the amplifier FK, is much larger than that by the amplifier FR,, the signal at the input A, will be much stronger (for example, more than 100 times in the present case) than the signal at point 8,. However, as a result of the aforesaid acoustic delay the voltage at the input C, cannot rise to a high value before the voltage at the input D, of the gate ND, has assumed a low value. As a result, the gate ND, is closed and hence the signal at the input A, cannot have any effect or, in other terms, the amplifying channel between the microphone M, and the loudspeaker H, remains open and that between the microphone M, and the loudspeaker H, remains cutoff.

During the short intervals between the words this situation must not vary, at least not as long as the other party at the microphone M, does not want to interrupt. It should be noted that as a result of the acoustic delay the signal at point P, not only appears somewhat later than at point P,, but also disappears somewhat later. Owing to the effect of the delay device 8,, which discharges comparatively slowly, the high signal level at the input A, is maintained for a short time after the signal at point P, has disappeared, i.e., for a time sufficient to insure that the signal at point P, also has decayed. The party in the other room may interrupt the first one by supplying an adequate sound signal to the microphone M,. Owing to the combined effect of this sound signal and. of the cross-talk signal from the loudspeaker H,, the signal level at the input B, can rise above that at the input A,. As a consequence the voltage at the input C, of the gate ND, drops so that a high voltage appears at point P resulting in the cut-off of the control member R,. The voltages at the two inputs of the gate ND, are now both high. The point P, then has a low voltage so that on the one hand the gate ND, is closed and on the other hand the control member R, opens the amplifying channel between the microphone M, and the loudspeaker H,,. This change-over is performed very smoothly.

As stated above, the signal level at the input B, initially was slightly lower than the signal level at the input A,. It is desirable for this difference to be as small as possible. This means that the microphone M, requires only a little additional signal for causing the signal level at B, to exceed that of A, or, in other words, the speaker at the microphone M, need raise his voice only to a slight extent to gain access to the device by reversing the direction of transmission of the amplifier.

Cross-talk from the loudspeaker H, to the microphone M, is in general comparatively intimately dependent upon frequency. When conventional loudspeakers and microphones are employed, cross-talk will be strongest at frequencies appearing at the central region of the speech spectrum, for example, between 1,000 Hz and 1,500 Hz. The ratio between the amplifications of PK, and FR, can then be adjusted in the optimum case so that at frequencies at which cross-talk is a maximum the signal level at the input B, still is slightly lower than that at the input A,. At higher and lower frequencies the level difference may then be considerably greater, which means that at these frequencies a comparatively strong signal has to be supplied to the microphone M, in order to cause the level at input B, to exceed that at A,. This is not desirable because it is quite possible that frequencies at which maximum cross-talk occurs are only weakly represented in the speech spectrum of the speaker.

In order to mitigate this disadvantage frequency correction may be carried out in the amplifiers so that the transmission through the amplifier FK, exhibits the same frequency depen dence as the transmission of the signal through the control member R,, the amplifier U,, the loudspeaker H,, the microphone M,, the amplifier T, and the amplifier F R,.

A correction may be carried out, for example, only in the amplifiers FR, and FR, so that the amplification of the high and low frequencies exceeds that of the medium frequencies. As an alternative, a correction may be carried out only in the amplifiers FK, and FK, so that the low and high frequencies are amplified to a lesser extend than the medium frequencies. Even a mixed correction in the four amplifiers is possible. By means of this frequency correction it may be achieved that despite cross-talk the speech level of the interrupting party at the microphone M, need exceed the level at the microphone by only a small amount, for example, by 5 db.

When the loudspeakers and the microphones are arranged in the same cabinets, the cross-talk level will be high. This is only an imaginary disadvantage. There is also an advantage in that direct cross-talk exceeds by far indirect cross-talk due to reflections from nearby objects or echos in the room. The overall cross-talk is therefore a practically constant factor and exhibits very little dependence upon the incidental disposition of the loudspeaker microphone combination in the room. This permits an optimum adjustment of the amplifiers FK,, FR,, etc.

As stated above, in principle, in the rest condition the control members R and R may be adjusted so that they either completely cut off the amplifying channel or so that a small steady amplification is left. It has been found that the latter alternative is to be preferred because in that case the distance between the rest condition and the completely open condition of the channel is smaller so that the opening of a channel from the rest condition is performed more smoothly or, in other words, even if the strength of the first syllable spoken is comparatively low, the relevant amplifying channel is immediately opened so that no part of this syllable is clipped.

However, some degree of transmission will then take place also through the channel not conveying speech which is not desirable in view of the fact that due to cross-talk the sound level at the microphone of said channel is much higher than at the microphone of the open channel. This may give rise to acoustic feedback in the system.

In order to avoid this phenomenon, auxiliary damping or control members DR and DR may be included in the amplifying channels. These auxiliary control members are controlled by the voltages at points P and P that is to say by the control voltages which govern the control members in the other channel. The damping members DR and DR are arranged so that they are conducting in the rest condition of the arrangement, that is to say, in the example described above, when there is a high voltage at the points P and P When speech is transmitted through the amplifying channel between the microphone M and the loudspeaker 1-1,, the voltage at point P, remains high so that the damping member DR continues to conduct. However, under the control of the low voltage at point P the damping member DR completely cuts off the other amplifying channel.

What is claimed is:

l. A speech-controlled bidirectional amplifier comprising first and second amplifying channels each of which includes a transmission control member governed by a control voltage derived from the speech signals appearing in the two amplifying channels at control points preceding the control member of each channel, first and second comparison devices each having first and second inputs and designed so that they supply an output work signal when the signal level at the first input exceeds the signal level at the second input, means for applying the speech signals of each channel individually through a pair of first signal circuits to the first input of the two comparison devices and individually through a pair of second signal circuits to the second input of the two comparison devices, the amplification of the first two signal circuits being arranged to exceed that of the second two signal circuits, a third comparison device including first and second gate circuits each having first and second inputs, means for individually applying said output work signals to the first input of said first and second gate circuits respectively, means for coupling the output of each gate circuit with the second input of the other gate circuit in a manner such that a work signal applied to one of the first inputs of one of the gate circuits will cut off the other gate circuit, the control voltages for the two control members being derived from the outputs of the gate circuits of the third comparison member.

2. An amplifier as claimed in claim 1, characterized in that the first and/or second signal circuits include frequency-correction circuits arranged to produce a uniform frequency response over a wide range of audio frequencies.

3. An amplifier as claimed in claim 1 characterized in that the two first signal circuits each include a charging circuit for maintaining the signal voltages appearing at the first inputs of the first and second comparison members for a given time after the speech signal applied to the respective signal circuit has terminated.

4. An amplifier as claimed in claim 1, characterized in that the control members are adjusted to a control level such that in the rest condition of the amplifier each channel is partially open and exhibits a relatively low amplification, each amplifying channel further including a second control member governed by the control voltage developed at the output of the gate circuit of the other channel so that when one channel is open in response to a speech signal the other channel is completely cut ofi.

5. A voice-operated bidirectional amplifier comprising first and second transmission channels including first and second transmission control members, respectively, each controlled by speech signals appearing at a control point in each channel that precedes the respective channel control member, first and second comparison devices each having first and second inputs and designed to compare the signal levels applied thereto so as to produce a first output signal when the signal level at the first input exceeds the signal level at the second input, a first pair of signal coupling circuits for individually coupling the speech signals at the control points of said first and second channels to the first input of said first and second comparison devices, respectively, a second pair of signal coupling circuits for individually coupling the speech signals at the control points of said first and second channels to the second input of said first and second comparison devices, respectively, a third comparison device including first and second gate circuits each having first and second inputs, means for coupling the output of said first and second comparison devices to the first input of said first and second gate circuits, respectively, means for coupling the output of the first gate circuit to the second input of the second gate circuit and to a control terminal of said first control member, and means for coupling the output of the second gate circuit to the second input of the first gate circuit and to a control terminal of said second control member, said first, second and third comparison devices being arranged so that in the rest condition of the amplifier with no speech signals present each gate circuit supplies a first control voltage to its respective control member that conditions same to provide high signal attenuation and simultaneously primes the second input of the other gate circuit so that the first channel to receive a speech signal will switch its respective gate circuit to provide a second control voltage that conditions its respective control member to provide low signal attenuation and simultaneously blocks the operation of the other gate circuit.

6. An amplifier as claimed in claim 5 wherein said first and second pairs of coupling circuits are arranged to transfer signals to the inputs of said first and second comparison devices that are proportional to the signal levels appearing at the respective channel control points so that the comparison devices can be switched in the middle of a speech transmission by an increase in the level of the signal coupled to the second input of the operative one of said first and second comparison devices such that the signal level at the second input thereof exceeds the signal level at the first input thereof.

7. An amplifier as claimed in claim 6 wherein said signal coupling circuits are adjusted so that the amplification of the first pair of signal coupling circuits exceeds the amplification of the second pair of signal coupling circuits.

8. An amplifier as claimed in claim 5 further comprising signal delay means connected in series with the first inputs of said first and second comparison devices.

9. An amplifier as claimed in claim 8 wherein said first and second pairs of coupling circuits are arranged to transfer signals to the inputs of said first and second comparison devices that are proportional to the signal levels appearing at the respective channel control points and independently of the attenuation condition of said first and second control members.

Patent Citations
Cited PatentFiling datePublication dateApplicantTitle
US3075045 *Aug 29, 1960Jan 22, 1963Bell Telephone Labor IncSpeakerphone
US3113181 *Nov 10, 1961Dec 3, 1963Ericsson Telefon Ab L MAmplifier arrangement
DE1281499B *Mar 23, 1966Oct 31, 1968Siemens AgSchaltungsanordnung fuer sprachgesteuerte Freisprechgeraete mit veraenderbarem Daempfungshub
GB668858A * Title not available
Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US3925618 *Feb 12, 1975Dec 9, 1975Nippon Telegraph & TelephoneVoice switch circuits for use in loudspeaking telephone circuits
US3952166 *Oct 3, 1974Apr 20, 1976Nippon Telegraph And Telephone Public CorporationLoudspeaking telephone circuit
US3953676 *Dec 12, 1974Apr 27, 1976Northern Electric Company, LimitedDigital control of a loudspeaking telephone system
US3963868 *Jun 27, 1974Jun 15, 1976Stromberg-Carlson CorporationLoudspeaking telephone hysteresis and ambient noise control
US3970786 *Oct 21, 1974Jul 20, 1976Stromberg-Carlson CorporationLoudspeaking telephone with improved receive sensitivity
US4025728 *Jan 29, 1976May 24, 1977Sava JacobsonSpeaker telephone
US4028506 *Dec 9, 1974Jun 7, 1977Nippon Electric Company, Ltd.Maximum value tracing circuit for digitized voice signals
US4465902 *Oct 8, 1982Aug 14, 1984Zenith Electronics CorporationDigital space phone system
US4542263 *Mar 7, 1984Sep 17, 1985Oki Electric Industry Co., Ltd.For a telephone set
US4980908 *May 30, 1989Dec 25, 1990Voicetek CorporationVoice-switched gain control for voice communication equipment connected to telephone lines
US5544242 *Dec 6, 1994Aug 6, 1996Exar CorporationSpeakerphone with event driven control circuit
Classifications
U.S. Classification379/344, 379/388.4, 379/395, 379/347
International ClassificationH04M9/10, H04M9/08
Cooperative ClassificationH04M9/10
European ClassificationH04M9/10