CA1244168A - Speakerphone for radio and landline telephones - Google Patents

Abstract

IMPROVED SPEAKERPHONE FOR RADIO AND LANDLINE TELEPHONES

Abstract Improved speakerphones (120 and 130 in Fig. 1) for both radio (112) and landline (138) telephones are described. The improved speakerphones (120 and 130) each include a microphone (102 and 132), a speaker (104 and 134) and unique control circuitry (106 and 136). The control circuitry of the improved speakerphone (200 in Fig. 2) interfaces a microphone (250) to a transmit signal (220) and speaker (260) to a receive signal (222) of a duplex communication path, such as a radio channel or telephone line. Variable gain stages (212 and 202) amplify and attenuate the speaker and microphone audio paths, respectively, in response to a gain control signal (223) from control logic (230). Transmit and receive signal detectors (206 and 207), each include gain adjust circuitry (208) a logarithmic amplifier (240), an envelope detector (241), a smoothing filter (245), a valley detector (242), a summer (243) and a comparator (244) for detecting the presence of audio signals in environments that may be subject to high background noise. Binary output signals from the transmit and receive signal detectors (206 and 207) are applied to control logic (230) which generates the gain control signal (223) and detector control signal (224). The control logic (230) includes delay circuitry (316 and 318) and gain control circuitry (392) for generating the gain control signal (223), the magnitude of which varies the amount of gain stages (202 and 212). Delay circuitry (316 and 318) and logic circuitry (304, 306, 308, 310, 312 and 314) in control logic (230) set and reset a flip-flop (302) for generating the detector control signals (224 and 225). Detector control signals (224 and 225) are coupled to gain adjust circuitry (208) and smoothing filter (245) for adjusting gain and response time of transmit and receive signal detectors (206 and 207).

Description

Ii~lPROVE~ SPEAKE~PHONE F~R I~DIO AND LA~DLIN~ TELEP~N~S

~ack~round Ar~

The present invention relates generally to speakerphones, and more particularly to an i~proved speakerphone that can advantageously be utillzed in both radio and landline telephone systems.
In both radio and landline telephone systems, a user communicates by means of a handset that includes a speaker at one end which is placed close to the user's ear and a microphone at the other end which is held close to the user's mouth. Thus, the user has only one free hand since the other must be used to hold the tel.ephone handset. In order to provide a greater degree of freeaom to the user, speakerphones have been developed for use in landline telephone systems. A conventional speakerphone is typically coupled to the telephone line and when enabled allows the hands-free party user to freely move about while still communicating with another party. Such speakerphones typically compare the volume of the hands-free party and the other party and select the party speakin~ the loudest. Furthermore, conventional speakerphones, such as that described in U.S. patent no.
3,7Sl,602, always revert to the listen rnode a short ti~e interval after the hands-free party stops talking.
However, conventional speakerphones suffer from a number of problems including inability to accormnodate high ambient noise environments and breakup due to feedback from the speaker to the microphone when the speaker amplification has been turned up. These conventional speakerphones work reasonably well when utilized with landline telephones in relatively quiet surroundin~s, but S may not operate correctly when used in noisy environments and especially when used with a radio due to the high degree of ambient noise encountered in vehicles. ~'or example, a hi~h level of ambient noise may cause such speakerphones to always select the hands-free party.
The problem of accomodating a high degree of aMbient noise has been addressed in the speakerphones described in U.S. Patent nos. 4,37~,603 and 4,40~,584 by selectin~
only one of the parties while totally muting the other party, and switching between parties usin~ the output of audio signal detectors that detect the presence of audio signals only when the audio si~nals exceed the background noise by a predetermined amount. Another prior art speakerphone totally ignores the hands-free party and always selects the other party whenever audio signals from the other party are detected.
However, none of these prior art speakerphones accomodates both a noisy environtnent and allows the hands-free party to break in relatively easily on the other party. Accordin~ly, there is a need for an improved speakerphone that can accomodate noisy environments and allow the hands-free party to break in relatively easily on the other party.

Summary of the Invention Accordingly, it is a ~eneral object of the present invention to provide an improved speakerphone that allows hands-free voice communications in environments havin~
high levels of ambient noise.
It is yet another object of the present invention to provide an improved speakerphone for radiotelephones where voice path gain switchin~ between the hands-free party and the other party is primarily dependent upon the presence or absence of audio signals from the other party and voice detector sensitivity switching is dependent upon the last party to speak.
Briefly described, the improved speakerphone of the present invention is coupled to a microphone, speaker and communications path for applying to said path microphone audio si~nals and applying incoming audio signals from said path to the speaker. The novel speakerphone includes first and second amplifiers having variable gains for amplifying the incoming audio signals and the rnicrophone audio si~nals by substantially equal but opposite amounts of gain, respectively, in response to a gain control signal; first and second detectors each having a sensitivity varyin~
oppositely from each other by pre-selected arnounts in response to a detector control si~l~al for detecting the presence of the incoming audio si~nals and microphone audio si~nals, respectively, and producing an output si~nal; and

2~ control circuitry for generating the gain control signal in response to the first detector output si~nal and generating and storing the detector control si~nal in response to the first and second detector output signals and the previously stored detector control signal.

Brief Description of the Drawin~s Fig. 1 is a block diagram of a cornmunication system including radio and landline telephones that rnay advan-tageously utilize the improved speakerphone embodyin~ the present invention.
Fi~. 2 is a detailed block diagrarn of the improved ~0 speakerphone of the present invention.
Fig. 3 is a detailed circuit diagram of the control logic 230 in Fig. 2.
Fig. 4 is a detailed circuit diagram for the smooth-in~ filter 245 in Fig. 20 Fig. S is a detailed circuit dia~ram for the ~ain adjust circuitry 208 in Fi~. 2.
Fi~. 6 is a detailed circuit dia~ram for the variable ~ain stages 202 and 2l2 in Fi~. 2.
Fig. 7 is a detailed circuit dia~ram for the ~ain control circuitry 392 in Fig. ~

Detailed Description of the Preferred Embodinnent In Fi~. l, there is illustrated a communications systeln including radio telephones l08 and landline telephones l38 that may advanta~eously utilize the improved speakerphone l20 and l30 embodyin~ the present invention. In conventional radiotelephone systems, such as IMTS (Improved ~lobile Telephone Service) and cellular radiotelephone systems, a party in a vehicle communicates by way of rnobile telephone l08 and mobile radio ll2 with a base station l22 which is interconnected by a control terminal l24 to a landline telephone central office l26.
The control terminal l24 includes control circuitry that provides a communication path frorn the base station l22 to hybrid circuitry (not shown) which is interconnected to a telephone line or trunk of telephone central office l26. A number dialed by the mobile party by means o~
mobile telephone 108, is received by the control terminal 124 and forwarded to the telephone central o~fice 12~ on the telephone line vr trtmk. The telephone central office 126 recei-ves the dialed telephone number and routes the mobile party to the desired land party telephone, e.~., to landline telephone 138. A similar process is employed when a land party aials the number of a ulobile party by means of conventional telephone 13~.
Furthermore, such communications systems may also include portable radios of the type described in U.S. ~atent ~os.

3,906,16~, 3,962,553 and 4,486,624 and in an article by Albert J. Leitich and Donald C. Linder, entitled "Portable Radiotelephone for ~ellular Systems", published in the Proceedings of the 30th ~nnual Conference of the IEEE Vehicular Technology Society, September 15-17, 1980, Dear'born, Michigan. Such portable radios can be carried by the user and operated wherever he may be, for example, from his office or his vehicle, and can likewise include the improved speakerphone of the present invention.
Further details of the si~nalling process and the nature of the circuitry for control terminal 124 is provided in Motorola Service ilanual 6~P~102~E65; for the base station 122 in l~lotorola Service Manual 68P~1033E10; for portable radio 112 in Motorola Service ~lanual 68P81046E60; and for mobile radio 112 and mobile telephone 10~ in Motorola Service Manuals 68~81039~25, 6~P~1029E65 and ~8P~1037~85, all of which are published by and available from Motorola Service Publications, Schaumbur~, Illinois.
In order to provide a mobile party with the same type of telephone service that a land party receives, it is necessary that each radio channel 114 be a duplex radio channel, having a separate transmit frequency and a separate receive frequency. Thus, mobile radio 112 may be simultaneously transmittin~ on one fr~quency and receiving on the other, such that both the mobile and land parties may simultaneously talk and listen. Mobile telephone lU~ may include a handset that is simil~r to handsets typically found in conventional land-line telephone instruments. During a telephone conversation, it-is necessary that the mobile party hold the handset up to his ear just as a land party does. ~owever, it would 3() be much safer and convenient for the mobile party if the conversation durin~ a telephone call could be achievea without the necessity of any physical actions. By uti'Lizing the improved speakerphone 12~ of the present invention, a mobile party can converse with a land party for the duration of a call with no required physical action, thus freeing the mobile party of tasks which would interfere with the safe operation of his vehicle.
The improved speakerphone 120 of the present invention, together with a mobile telephone 108 such as that described in U.~. Patent Nos. 4,122,203, 4,22U,820 and

4,486,624 which provides for on-hook automatic dialin~ of telephone numbers stored in a memory, allows a mobile party to safely make telephone calls without the necessity of rernoving one of his hands from the steerin~
wheel for the duration of the call. In Fig. 1, speakerphone 120, includin~ microphone 102, speaker 104 and control circuitry 106, is switchably coupled to rnobile radio 112 by switch 110. Switch 11~ selectively couples a transmit signal and a receive signal from mobile radio 112 to either speakerphone 120 or mobile lS telephone 108.
In landline applications, the improved spea~erphone 130 of the present invention can likewise be used to allow total freedom of movement durin~ conversations and multi-party conference calls. In Fig. 1, speakerphone 130, includin~ microphone 132, speaker 134 and control circuitry 136, can be switchably coupled to hybrid circuitry 142 and thereafter to telephone line 144 by means of switch 140. Switch 140 selectively couples a transmit signal and a receive signal from hybrid circuitry 140 to either speakerphone 130 or a conventional telephone 138. ~iybrid circuitry 142 may be any suitable commercially available device for interfacing a transmit signal an~ receive si~nal to telephone line 144. ~urthermore, there are commercially available telephones 138 that provide for on-hook automatic dialing of telephone numbers stored in memory.
In order to provide for hands-free operation, speakerphones 120 and 130 include a separate microphone 102 and 132, a separate speaker 104 and 134, and control circuitry 106 and 136. Alternatively, the microphone and speaker in the handset of telephones 108 and 13~ could be utilized if properly designed and oriented with respect to the hands-free party. In most radio and landline applications, it is advanta~eous to utilize a separate microphone 102 and 132 ~or receiving the hands-free party's speech. By experimentation, it has been found that the best mounting location for microphone lO2 in a vehicle is on the interior roof lining of the automobile either cn the visor or just before the edge of the visor.
This location has been found to rrlaximize voice pick-up while minimizing background noise pick-up. In addition this microphone mounting location also provides some degree of inherent vibration damping. ~xperimentation has also been determined that an omni-directional electret rnicrophone cartridge provides good performance while at the same time being cost effective. Also, ~or vehicular applications, speaker 104 can be located on the dashboard of the vehicle or on the deck near the rear window of the vehicle. In landli~e applications, microphone 132 and speaker 134 rnay also be spacially separated and/or differently oriented to reauce the coupling therebetween.
Referring to Fig. 2, there is illustrated a detailed block diagram of an improved speakerphone 20~ embodying the present invention, which interfaces microphone 250 and speaker 260 to a transmit signal 220 and a receive signal 222, respectively, of a duplex communications path. According to a feature of speakerphone 200, two variable gain stages 202 and 212 are included which dynamically vary the amplification of audio signals from the rnicrophone 250 or from receive signal 222, respectively, in response to a gain control signal 223 from control logic 230. The rnagnitude of gain control signal 223 linearly varies the gains of sta~es 202 and 212 in substarltially equal but opposite amounts as measured in decibels. That is, as the gain of stage 212 increases, the gain of stage 202 decreases) and vice versa. Since gain sta~es 202 and 212 produce opposite but substantially equal changes in the amount of ~ain, stages 202 and 212 could be interchanged as long as the sense of the gain control signal 223 is chan~ed. The magnitude of ~ain control signal 2~3 is primarily dependent on the output si~nal from receive si~nal detector ~07. In the preferred embodiment, the gain control signal 223 is a current generated by control logic 230. Also, each of the audio paths includes fixed gain stages, 201 and 203 in the microphone audio path and 211 and 213 in the speaker audio path. Amplifier 213 in the speaker audio path preferrably has an adjustable ~ain which may be manually adjusted by the hands-free party for controlling the volume of speaker 112.
The audio si~nals in the microphone àudio path are coupled from amplifier 203 to transmit signal 220 for application to the communication path, which may be a duplex radio channel 114 or a telephone line 144 in Fi~.
1. Audio signals received from t~e communication path are selectively coupled by receive signal 222 to amplifier 213 in the speaker audio path. In radio applications, the transmit signal 220 is coupled to the transmitter, and the receive signal 2~2 is coupled to the receiver, of a rnobile or portable radio, such as radio 112 in Fig. 1. In landline applications, the transrnit signal 220 is coupled to the transmit input, and the receive signal 222 is coupled to the receive output, of conventional hybrid circuitry 142 in Fig. 1, which inter-faces telephone 13~ to its associate~ telephone line 144.
The novel speakerphone 2U0 in Fig. 2 also includes transmit and receive si~nal detectors 2~ and 2U7 having sensitivity switching for detecting the presence of the hands-free party's voice si~nals and the other party's voice signals, respectively. If microphone 250 does not provide a high-pass response, a high-pass filter 2~5 may be interposed between microphone 250 and transmit signaL
detector 206. The microphone sigrlal is high-pass _ 9 _ '~f ~ $~

filtered by filter 205 in order to remove low frequency background noise typically present in the vehicular environment. A high-pass filter 205 havin~ a cut off frequency of a~proximately 300 ~Iz will filter out most of S this low frequency back~round noise. Similar hi~h-pass filtering Inay also be added to the receive si~nal path 222. Furtherrnore, in applications that are not character-ized by such low frequency back~round noise, a high-pass filter 205 may not be required in either audio path.
The transmit and receive signal detectors 206 and 207 include substantially identical circuit blocks 20~-245. Both detectors 2U6 and 207 are comprised of conventional circuit blocks including gain adjust circuitry 20~, a lo~arithmic amplifier 240 (commonly referred to as a "soft" limiter), an envelope detector 241, a smoothin~ filter 245, a valley detector 242, a sutnmer 243 and a comparator 244. The gain adjust circuitry 2U8 varies the ~ain between two pre-selected levels depending on the binary state of detector control si~nal 224 (245 for detector 2~7). The logarithmic amplifier 24~ extends the dynamic range of the detectors 20b and 207 due to its amplification characteristic. The envelope detector 241 provides an output si~nal which follows the maxima and minima, or envelope, o~ the audio si~nals. The smoothing filter 245 provides variable bandwidth low-pass ~iltering depending on the binary state of detector control signal 224 (225 for detector 2~7). Gain adjust circuitry 2~8 and smoothing filter 245 may be bypassed when not utilized by Jurnpers 27b and 278, 3U respectively. The valley detector 242 operates as a rectifier which follows the minima of the smoothing filter output. The valley detector output essentially correspon~s to the steady state background noise present on the transmit si~nal 22~ in the case of detector 2~b or on the receive si~nal 222 in the case of detector 2~7.
Summer 243 adds a reference volta~e VR to the valley detector output. ~y addinO reference volta~e VR to - l o -the valley detector output, comparator 244 will not ~enerate spurious output signals due to low ~nplitude variations of back~round noise. Thus, the comparator 244 will only provide an output si~nal if the smoothin~
filter/envelope detector output exceeds the steady state back~round noise by the ma~nitude of reference volta~e VR. Reference volta~e VR is chosen to~ether with the response characteristics of the envelope detector Z41 and smoothing filter 245 so as to avoid generation of spurious comparator output si~nals due to low amplitude variations of the background noise. The output signal from comparator 244 has a hi~h voltage level, or binary one state, when the smoothin~ filter/ envelope detector output exceeds the valley detector output by reference volta~e VR and otherwise has a low volta~e level, or binary zero state. A binary one state of the output signal from comparator 244 in detector 206 into control lo~ic 230 indicates that audio si~nals from microphone 250 have been detected, and a binary one state frorn the comparator 244 in detector 207 into the control logic 2~() indicates that audio signals from the receive si~nal 222 have been detected.
Summer 2l4 in It'i~. 2 applies to receive si~nal detector 207 a signal that is the sum of both the audio signals from the receive si~nal 222 and the audio si~nals from ~ain sta~e 2l~. Amplifier 2l3 has an adjustable ~ain control to allow the hands-free party to manually adjust the volume of speaker 2~0. Thus, the si~nal applied by s~nmer 2l4 to detector 207 varies as ~he volume of speaker 260 is varied. In other embodiments, the inputs of summer 2l4 may be coupled to receive si~nal 222 and the output of amplifier 2l3. Allowin~ the si~nal level applied to detector 207 to increase as the volume of speaker 2~ increases, and vice versa, helps to main-tain a balance between the si~nal levels applied to thereceive si~nal detector 207 and transmit si~nal detector 2~6. As a result, the voice detect sensitivities of detectors 20~ and 207 are substantially equalized over the volume ran~e of the speaker 2~0. This feature of speakerphone 200 avoids unnecessary sensitivity switching between the transmit audio path and receive audio path.
In addition, this feature insures that an increase in the volume of speaker 260 does not make it more difficult for the hands-free party to keep the transmit audio path in the higher gain state.
The control logic 230 in Fi~. 2 is responsive to the output signals from the transmit and receive signal detectors 206 and 207 for switching the gain in the transmit and receive voice paths and the sensitivity of voice detectors 206 and 207. The control circuitry 230 provides a gain control si~nal 223 for controlling the amount of ~ain in ~ain stages 202 and 212, and detector control signals 224 and ~S for controlling the amount of sensitivity in the translnit an~ receive signal detectors 206 and 207, respectively.
~'he control lo~ic 230 in ~'ig. 2 is illustrated in more detail in the circuit diagram in ~'ig. 3. In Fig.
3, ~ain control signal 223 for variable gain stages 202 and 212 is provided by ~ain control circuit 392. ~hen the magnitude of the gain control si~nal 223 from ~ain control circuitry 3~2 is low, the audio si~nals from microphone 250 are passed by variable gain stage 202 with a low amount of gain and the audio signals from receive si~nal 222 are yassed by variable ~ain stage 212 with a high amount of gain. ~hen the rr,agnitude of ~ain control signal 22~ is high, the opposite is true. In the preferred embodiment the gain control si~nal 223 cornes froln a current source output, so low current represents a low magnitude for gain control signal 223.
In Fig. 3, detector control signal 224 for decector 20~ and detector control si~nal 225 for detector 2U7 in Fig. ~ is provided by the Q output of fliy-flop 3U2 and the output of gate 204, respectively. When detector control signal 224 has a binary ~ero state (225 has a binary one state), gain adjust circuitry 20~ and smoothing filter 245 have a higher gain and quicker response time which results in a ~reater amount of sensitivity in transmit si~snal detector 2U6 than in

5 receive signal detector 207. When detector control si~nal 224 has a binary one state, 8ain adjust circuitry 208 and smoothing filter 245 have a higher ~ain and quicker response time in receive si~nal detector 207 than in transmit signal detector 206. As previously 1() indicated, the preferred embodiment does not include gain adjust circuitry 208 in transmit signal detector 206 and does not include smoothiny, filter 245 in receive signal detecto~ 207.
~eferrin~ to Fi8. 3, the Q output (detector 15 control signal 224) of the flip-flop 3Q2 will be set to a binary zero state (low voltage level) when ~IO~ gate 304 has a binary one state (hi~sh voltage level)O NOR y,ate 304 provides a binary one state at its output when the output signal ~rom detector 2()6 changes to a binary one state indicating that the hands-free party's speech has been detected, the held receive signal detector output signal 394 has reached a binary zero state after a time interval defined by delay circuitry 316, and the output of flip-flop 312 has a binary zero state. The Q
output (detector control si~nal 224) of flip-flop 3()2 will be set to a binary one state when NOK y,ate 306 has a binary one state. NOR gate 3()6 provides a binary one state at its output when the held transmit si~,nal detector output si~nal 3~6 reaches a binary zero state, which means that the output si~nal from detector 206 has had a binary zero state for a time interval defined by delay circuitry 318, the ~ output of flip-flop 314 tlas a binary zero state, and the output signal from the receive signal detector 21J7 has ~ one state.
Delay circuitry 316 and 318 in ~'ig. 3 both have fast charge paths provided by dioaes 342 and 362, respective-ly. Thus, when the output si~,nal from transmit si~nal detector 206 changes to a binary one state, diode 362 bypasses components 363, 365 and 3~ so that capacitor 360 quickly char~es to a hi~h voltage level. Likewise, when the output signal from receive signal detector 207 chan~es to a binary one state, diode 34~ bypasses components 343, 345 and 346 so that capacitor 340 quickly charges to a high voltage level. Both delay circuitry 316 and 318 provide delay time intervals for output signals 394 and 396 when dischar~in~ from a hi~h voltage level to a low voltage level in response to chan~es of the output signals from detectors 207 and 206, respectively, from a binary one state to a binary zero state.
The tirne intervals for the delay circuitry 316 and 31~ is a compromise which reflects conflicting requirements for fast sensitivity switchin~ ana minimum degradation of operational characteristics in the presence of high background noise. The time intervals for delay circuitry 316 an~ 31~ include a fixed time interval and a variable time int~rval havin~ a ma~nitude dependent on the duration of the output si~nal frotn detectors 207 and 206, respectively~ In the preferred embodiments, delay circuitry 31~ ana ~1~ provide a short tilne interval of approximately 10 nlilliseconds for output si~nals from detectors 207 and 206 havin~ a short dura-tion; and provide a variable time interval ran~in~ frorn zero to approximately 130 milliseconds for output si~nals from detectors 207 and ~06 havin~ a lon~ duration. The time interval provided by delay circuitry 31 b and 318 will, therefore, vary between a minimum of 10 ~nilli-seconds and maximum of 140 millisecon~s. ~urthermore, by proper design of delay circuitry 31~ and 318, the time interval can be chosen to be proportional to the duration of the output si~nal from detectors 207 and 2~ over the ran~e from 10 milliseconds to 140 milliseconds.
In ~eneral, the time interval of delay circuitry 316 and 31~ may vary from S to 400 milliseconds depending upon the requirements of a particular application of the speakerphone of the present invention. A predecermined minimun1 amount of time delay, such as 5 to 10 milliseconds, is always necessary to avoid spurious sensitivity switchin~ caused by signal delay throu~h the communications network, i.e. the control terminal 124 and telephone central office 126, at the cessation of either party's speech.
In the preferred embodiment of a landline speaker-phone, capacitor 340 and resistor 341 in delay circuitry316, and capacitor 360 and resistor 361 in delay circuitry 318, have a time constant of approximately 1 milliseconds, while capacitor 344 and resis~or ~41 in delay circuitry 316, and capacitor 364 and resistor 361 in delay circuitry 318, have a time constant of approximately 130 milliseconds. This selection of time constants provides a fixed time delay of approximately 10 milliseconds and a variable time delay of from 0-13 milliseconds in the preferred embodilnent of ~elay Z0 circuitry 31~ and 318. 'l'he proporcionality between the variable time delay of capacitor 344 and resistor 341 of delay circuitry 316 and the duration of the receive detector output signal 207 is controlled Dy the time constant of capacitor 344 and resistor 345 which can ran~e from 3~ to 15~ milliseconds dependin~ on the environment of the application and is ~ milliseconas in the preferred embodiment of a landline speakerphone.
Likewise capacitor 364 and resistor 3~5 of delay circuitry 31~ control the proportionality of the variable 3~ time delay with respect to the duration of the transmit detector output si~nal 206.
The foregoing features of the speakerphone 200 facilitate switchin~ the state of detector control si~nal 224 when short bursts of audio si~nals are occurrin~ on ~5 the audio path presently having hi~her sensitivity. If these bursts are relatively short, sensitivity switchin~
can occur during pauses between speech bursts, since the time interval of the delay circuitry 31~ or 31~ is also rather short under such conditions.
Accordin~s to another feature of speakerphone 200, detector control si~nal 224 may change state only when the presently speaking party has not provided sufficient audio activity to maintain a detector output signal 2()6 or 207. Flip-flops 312 and 314 are used to provide this feature. Assuming all other inputs to N()l~ gate 304 or 306 have binary zero state, break-in takes place if the (~
output of flip-flop 312 or 314, respectively, momentarily has a binary zero state. For example, the t~ output of flip-flop 312 will have a binary zero state when the receive si~;nal detector output si~nal 207 has a binary zero state (no audio signals bein~ detected on the receive signal 222) and the transmit signal detector output signal 206 thereafter changes state from a binary zero state to a binary one state (indicating that the hands-free party is startiny, to speak). ~'lip-flop 314 operates in a similar manner. This operation of flip-flops 312 and 314 allows cleteccor control si,~,nal 224 to chan~,e only if the detector outljut signal of the presently speal;ing party chan~es state an-l the detector output si~,nal of the other party thereafter changes state. The major reason for ttliS feature is to prevent spurious post syllabic sensitivity switchin~ caused by near simultaneous detector outputs which are due to sidetone in control terminal 124 or telephone central office 126 in Fig. 1 and sidetone aue to speaker to rllicrophone couplin~ in the speakerphone.
According to yet another feature of speakerphoné
200, smoothing filter 245 in ~i~. 4 May be included in detectors 2()~ and 207 in Fi~. 2 for increasin~, the smoothing in the microphone audio path when the speaker audio path has the hi~,her ~ain, ana vice versa. This ~5 feature of the present invention minimizes syllabic attack time choppin~ due to sidetone in control terminal 124 or telephone central office 12/~ in i!i~. 1 and sidetone due to speaker to microphone coupling in the speakerphone. In the preferred embodiment, slooothin~
~ilter 245 is bypassed by jumper 27~ in receive si~nal detector 207. Referring to Fig. 4 smoothin~ filter 245 includes capacitors 404 and 40~, resistors 402, 406 and 412 and transistor 410. Transistor 410 is turned on in response to a binary zero state of the detector control signal 224 in Fig. 2 in the case of detector 206 and a binary zero state of the detector control signal 225 in the case of detector 207. When transistor 410 turns on, resistor 406 is coupled in parallel with resistor 4~2, lowering effective magnitude of the resistance in the path between envelope detector 240 and both valley detector 242 and comparator 244. As a result, the sensitivity of detector 206 or 207 is increased by a pre-selected amount since the signal gain is increased and the response time is lowered when transistor 410 turns on. The effect of this feature of speakerphone 200 is that the detect gain and response time of one of 2U detectors 20~ or 207 is ~reater ~han the other even though the audio signal level applied to the other may be hi~her. Thus, if detector control signal 225 has a binary zero state, receive signal detector 207 has the hi~her ~ain and faster response time; and if detector control signal 224 has a binary zero state, transmit si~nal detector 206 has the hi~her gain and faster response time.
According to a further feature of speakerphone 200, gain adjust circuitry 20~ in Fig. 5 may be interposed between high pass filter 205 and the logarithmic amplifier 2~0 of the detectors 206 and 207 in Fig. 2 in order to further mir1imize syllabic attack tirne chopping.
In the preferred embodiment, gain adjust circuitry 20~ is bypassed by j ~lper 27~ in the transQit signal detector 206. R~Eerrin~ to Fig. 5, 8ain adjust circuitry 20~
includes a potentiometer 502, transistor 504 and resistor 506. Transistor 504 is turned on in response to a binary zero state of the detector control signal 224 for detector 206 or the detector control signal 225 for detector 207 in Fig. 2, and shorts out potentiometer 50 to increase the amount of gain. As a result of the higher signal gain, the sensitivity of one of the detectors 206 or 207 is increased by a pre-selected amount.
Referring next to Fig. 6, there is illustrated a detailed circuit diagram for the variable gain stages 202 and 212 in Fig. 2. In the preferred embodiment, a Signetics ~ type NE570 compandor 600 is used for variable gain stages 202 and 212. The gain control signal 223 is coupled via resistors to rectifier cells 602 and 612 of compandor 600. The audio signals from the microphone are coupled to variable gain cell 604 in the input of operational amplifier 606 in stage 202. The audio signals from the receive signal are coupled to the input of operational amplifier 616 which has variable gain cell 614 in its feedback loop. Since both variable gain cells 604 and 614 are affected equally by changes in gain control signàl 223, substantially equal but opposite changes in gain in deciBels (plus or minus 14 deciBels in the preferred embodiment) are produced in variable gain 25 stages 202 and 212 of Fig. 2. That is, as the magnitude of the gain control signal 223 increases, the gain of the microphone signal is increased and the gain of the receive signal is decreased. In the preferred embodiment, changes in the magnitude of the gain control signal 223 are changes in current.
Referring to Fig. 7, there is illustrated a detailed circuit diagram of the gain control circuitry 392 in Fig. 3.
Operational amplifiers 702 and 704 act as comparators for the held output signals 394 and 396 from blocks 316 and 318 in Fig. 3, respectively. Operational amplifier 706 is arranged as an integrator. The dual time constants of gain control signal 223 are produced by capacitor 722 and resistors 724 and 726. When che output o~ amplifier 7~4 switches hi~h (audio signals frorm receive signal 222 detected by detector 2~7), the time constant is determined by capacitor 7~2 and the parallel combination of resistors 724 and 726, typically 30 milliseconds. ~Jhen the output of amplifier 704 switches low, the time constant is determined by capacitor 72~ and resistor 726 since diode 732 is reverse biased, ~ypically 500 milliseconds. When the hands-free party begins to talk and is detecteà by detector 206, the output of amplifier 702 will ~o high turning on transistor 734 and shorting out diode 732. Shorting out diode 732 changes the cime constant by placing resistors 724 and 72~ in parallel and reducin~ the time required for changing the gain control signal 223 to allow the han~s-free party to be heard (ie., tilne constant chan~es from typically 500 to 30 milliseconds). Amplifier 710 and transistors 712 and 714 convert the output of integrator 706 to chan~es in current of the gain control signal 223. The chan~es in current of the ~ain control signal 223 cause linear changes of substantially equal magnitude in the impedance of gain cells 604 and 614 in Fi~. 6 which as previously described produces substantially equal but opposite ~ain changes in the variable gain stages 202 and 212 of Fig.
~.
'l'he logic circuit elen~ents 302, 304, 306, 30~, 310, 312 and 314 in Fig. 3 may be mechanized by ucilizing the circuit elelnents in conventional integrated circuits, such as the C~IOS integrated circuits described in the "C~IOS Lnte~rated Circuit Book", published by ~iotorola Semiconductor Products, Inc., Austin, Texas, 197~.
~loreover, the logic circuit elements 302, 304, 306, 3~, 310, 312 and 314 in Fig. 3 to~ether with many of the resistors an~ capacitors of blocks 316, 31~ and 392 Tnay be readily inte~rated into a semiconductive substrate and packaged as a sin~le inte~rated circuit device.

- l 9 -The improved speakerphone of the present invention may be advanta~eously utilized in a wide variety of applications requiring complete freedom of movement or multiple-party conference calls. For exaMple, the improved speakerphone of the present invention can be utilized to provide hands-free communicatiotls to a mobile party in radiotelephon~ sy~ten~s or to a land party in landline telephone systems. The speakerphone of the present invention can be adapted for wide variety of 1~ applications since the level of ambient b~ck~round noise is taken into consi~eration.

Claims (6)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A speakerphone including a microphone and a speaker and being coupled to a communications path for applying to said communications path microphone audio signals and applying incoming audio signals from said communications path to said speaker, said speakerphone further comprising:
first amplifying means having a gain varying in predetermined proportion with a gain control signal for amplifying the incoming audio signals;
second amplifying means having a gain varying in opposite proportion to said predetermined proportion with the gain control signal for amplifying the microphone audio signals;
first detecting means having a predetermined response characteristic varying by a pre-selected amount in response to a detector control signal for detecting the presence of the incoming audio signals and producing an output signal;
second detecting means having a predetermined response characteristic varying by an amount opposite to said pre-selected amount in response to the detector control signal for detecting the presence of the microphone audio signals and producing an output signal;
first control means coupled to the output signal of said first detecting means for generating the gain control signal having a magnitude related to the output signal of said first detecting means, said first control means further including means coupled to the first detecting means output signal for integrating the first detecting means output signal to produce an integrated output signal, and means coupled to said integrating means for generating a gain control signal current having a magnitude related to the integrated output signal of said integrating means;
and second control means coupled to the output signal of said first detecting means and the output signal of said second detecting means for generating the detector control signal.

2. The speakerphone according to claim 1, wherein said first amplifying means includes an operational amplifier having a feedback impedance and input impedance, and said second amplifying means includes an operational amplifier having a feedback impedance and input impedance, said input impedance of one of the first and second ampli-fying means varying with the gain control signal for varying said gain of said one of the first and second amplifying means, and said feedback impedance of the other of the first and second amplifying means varying with the gain control signal for varying said gain of said other of the first and second amplifying means.

3. The speakerphone according to claim 1, wherein said speaker audio signals include incoming background noise, said microphone audio signals include microphone background noise, said first detecting means includes means for detecting the incoming background noise and means for comparing the detected incoming background noise to the incoming audio signals for detecting the presence of the incoming audio signals, and said second detecting means includes means for detecting the microphone background noise and means for comparing the detected microphone background noise to the microphone audio signals for detecting the presence of the microphone audio signals.

4. The speakerphone according to claim 1, wherein said first detecting means includes attenuating means having a variable attenuation for attenuating the incoming audio signals, and said second detecting means includes low pass filter means having a variable passband for filtering the microphone audio signals, said attenuating means of the first detecting means having an attenuation varying by a pre-selected amount in response to the detector control signal, and said low pass filter means of the second detecting means having a passband varying by an amount opposite to said pre-selected amount of said attenuating means in response to the detector control signal.

5. The speakerphone according to claim 1, further including, interposed between the communications path and the first detecting means, third amplifying means for amplifying the incoming audio signals by a pre-selected gain and summing the incoming audio signals from the communications path and the amplified audio signals from the first amplifying means.

6. The speakerphone according to claim 5, wherein said third amplifying means has switchable first and second pre-selected gains, said speakerphone further including means for switching from said first pre-selected gain to said second pre-selected gain in response to the output signal of said second detecting means and switching from said second pre-selected gain to said first pre-selected gain in response to the output signal of said first detecting means.