United States Patent (19)
`Alcivar
`
`3,746,789
`[11]
`(45) July 17, 1973
`
`54). TISSUE CONDUCTION MICROPHONE
`UTILIZED TO ACTIVATE A VOICE
`OPERATED SWITCH
`75 Inventor: Ernesto A. Alcivar, Guayaquil,
`Ecuador
`: Dennis J.
`73) Assignees ennis J Johnson, Swampscott
`Brian N. McCarthy, Cambridge,
`Mass.
`Oct. 20, 1971
`22 Filed:
`(21) Appl. No.: 191,006
`
`OTHER PUBLICATIONS
`Gazey and Morris, An Underwater Acoustic Telephone
`for Free-Swimming Divers, Electronic Engineering,
`Vol. 36, No. 436, 6/1964 p. 364-368.
`
`Primary Examiner-Kathleen H. Claffy
`o
`Assistant Examiner-Jon Bradford Leaheey
`Attorney-C. Yardley Chittick, Richard J. Birch et al.
`
`ABSTRACT
`57
`52 U.S. Cl............................................... 17911 VC A voice-activated transmit switch (VOX) for high
`51) int. Cl. .............................................. G10) 1/04
`noise environment voice communication systems
`58 Field of Search............ 17911 VC, 1 SW, 1 SA,
`which employ a speech microphone, a transmitter and
`179/121 C, 167, 168, 164, 187, 188, 1 P, 1
`a receiver. A separate tissue-conduction microphone is
`VW, 157, 1 ST; 340/8R, 5 R, 5 T
`employed to generate a signal which activates a trans
`mitter enabling and receiver disabling circuit. The tis
`References Cited
`sue-conduction microphone is positioned in contact
`56
`with the user's neck tissue in the vicinity of the larynx.
`UNITED STATES PATENTS
`is
`2/1916 Forest...................................
`Abandpass filter eliminates the unwanted signal from
`l, 170,882
`the tissue-conduction microphone output and passes
`41 1945.
`French....
`2,374,090
`17971 AL
`... 34015 T
`the desired speech signals to an amplifier. The ampli
`12, 1966 Davis.......
`3,292,618
`fier output actuates a Schmitt trigger which in turn op
`325/22
`1935,744 1 1/1933 Holden .......
`3, l89,691
`6/1965 Simpson.....
`- - 17911 VE
`erates a transmitter enable-receiver disable switching
`2,424,216
`7|1947 Atkins.....
`..., 17911 VC
`circuit. Delay means are provided so that the transmit
`2/1972 Griggs............................... 17911 SA
`ter does not turn off during brief intersyllabic pauses.
`3,646,576
`FOREIGN PATENTS OR APPLICATIONS
`13 Claims, 2 Drawing Fi
`734,732
`3/1943 Germany......................... 1791121 C
`aims, Z Urawing Figures
`
`BreATHNG fa
`APPARATUS
`SA 2d
`N
`
`-Y's
`
`SPEECH
`PREAMP
`
`
`
`ACTWE
`HGH - PASS
`FLTER
`
`speech
`AMPLFer
`CLPPER
`
`f
`
`go
`
`32
`
`UNEO
`WOTAGE
`POWER
`CONROUEO
`OSCAR AMPLFER
`so
`
`24
`
`ld W - PASS
`ACTWE
`FitER
`
`
`
`34
`
`3a
`
`AUC
`
`sch Mitt
`
`detector
`78
`
`RiGGR
`2 8O
`
`NOISE
`TMNG
`GAE
`32.
`
`SENCE TRESHOld
`TMNG
`GATE
`
`de TECOR
`
`ALARM
`AUDO
`OSCAOR
`
`TRANSMIt
`RECEWE
`GAE
`
`34
`
`ACTWE
`bANPASS
`
`HGH-GAN
`AMPFER
`
`AUDO
`DeTecTOR
`
`schMitt
`RGGER
`
`ses
`
`aso
`
`B4) 62. 5
`
`8
`
`Receiver
`POWER
`SWITCH
`72
`(R
`
`A
`
`TRANSMtter
`9e
`P
`WeR
`function
`SWitch NE
`7.
`SWitch
`()
`B2-e
`Nico
`o
`
`AUDO
`AMPFER
`
`r
`0.
`
`ANDPASS
`FTER
`
`R
`MTE
`
`80 OB
`AMPFER
`
`NFU
`PREAMPLIFER
`
`so
`
`4.
`
`44
`38 f
`
`AyA
`
`42
`
`Ao
`
`9A
`
`NC
`o
`2
`S
`4
`
`
`
`t
`SPEECH
`t
`MC D
`f 2
`
`Ao
`N-
`
`
`
`
`
`
`
`
`
`98
`
`se
`
`HROAT
`MC
`sy
`
`BONE
`conduction
`HEADPONE
`R
`32
`
`- 1 -
`
`Amazon v. Jawbone
`U.S. Patent 8,019,091
`Amazon Ex. 1005
`
`

`

`PATENTED Jut 1 7 1873
`
`3,746,789
`
`SHEET 1 OF 2
`
`weor
`
`43a1aIndWV
`¥YOIVTTIOSO
`¥3L1I4
`
`4Y3MOd
`G3TIOULNOS
`SAILOV
`
`Q3nnt
`
`SSVd-MOT
`
`a°6
`
`
`
`HOLIMSHOLIMS
`
`,
`
`
`NOILONNS“SS93130ataiqdwy
`
`
`LINSNVYLLWUvIV“ZONaTIS3SIONLLIWHOS;olgny
`
`tLCh8?2?0?‘BYOLOS13S
`
`
`
` BALLOVrpJ—95*(v)|PweggaivoHOWWTNOSO|“|yoLroaLaaaLvs“ws99INL4010313036a1v9
`
`
`3AIZO3MotanyONIWILONIWISL
`
`
`92Woka)A“o/
`
`UALUNSNVYLU3AIZ03"
`
`
`yalngnvaed)=|walayaLidAYqIndlaao9SSVdONvaotanyWaLINITyas
`Y3MOduaModHS0oruLyOLOaLaualaiidssvd0nvaGqo1n
`
`
`
`on2hthfos
`
`BE
`)Ry)(3)cogamos
`
`
`
`LIINHOSolanyNIV9O—HOIHLIVOMHL
`
`
`B3Id1dAV
`
`
`
`Yaddi1oyan
`
`HOaadSs
`
`SAILSV
`
`H933dSCHOaadSu
`dWVaud\OIN
`
`eAl
`
`of
`
`9!
`
`
`
`SNLVUVddV
`
`ONIHLV3yN6
`
`NOILONGNOD
`
`SNOHdQV3H
`
`zs
`
`o/
`
`- 2 -
`
`
`
`
`
`
`
`

`

`PATENTED Ju 1 7 ta2s
`
`3,746,789
`
`SHEET 2 OF 2
`
`YSLLINSNVYOL+
`
`e6SiL
`
`16s
`
`VNHOLIMS
`YBLLINSNVYL
`
`AL
`
`Y3AIZ034
`
`ZL
`
`,eneHOLeAOQi\/
`
`©
`
`HOLIMS
`/\/\
`
`
`1asOnanfo]7WANvW
`
`L6SiLe6SiL
`y3A1393a+fo|
`fo}AZI+
`Z?LG
`cmt
`feo
`SeOeWO
`
`woud
`
`Azgilt+
`
`yaoolddLtsMOLVINDaySutonBIoF
`Ww-aw399!WHO:
`893OVLIOA
`eeMBLSSVd
`ONVS1NIML
`
`BF
`
`¢Dl
`
`—~+®
`
`“|
`
`100°
`
`[2]
`
`SNOHdONOIW
`
`LVOYHL
`
`- 3 -
`
`
`

`

`5
`
`10
`
`15
`
`3,746,789
`2
`TISSUE CONDUCTION MICROPHONE UTILIZED
`speech cavity 12 which surrounds at least the diver's
`TO ACTIVATE A VOICE OPERATED SWITCH
`mouth in order to provide a gas space into which the
`diver can articulate speech in a more or less normal
`BACKGROUND OF THE INVENTION
`fashion. The speech cavity is also connected to the div
`The present invention relates to communication sys
`er's breathing apparatus 16 so that the acoustic input
`tems in general and, more particularly, to a voice
`to the cavity is a mixture of speech occurring at random
`activated transmit switch for high noise environment
`intervals and gas flow and breathing noises occurring at
`voice communication systems.
`more or less regular intervals.
`The use of voice activated transmit switches (VOX)
`The speech cavity 12 has sharp, well-defined reso
`in communication systems is well known. In such sys
`nances typically below 1000 ha, which cause the acous
`tems, the output from a speech microphone is used to
`tic output of the diver's vocal track to rise at a rate of
`actuate a circuit which enables the transmitter while at
`roughly 12 decibals per octave below that frequency.
`the same disabling the receiver. However, this type of
`This abnormal frequency response in the speech signal
`system is highly susceptible to false keying of the trans
`seriously impairs intelligibility. A speech processor, in
`mitter by non-speech noises. The problem of false key
`dicated generally by the reference numeral 18, is em
`ing of the transmitter is particularly acute in underwa
`ployed to electrically modify the speech signal from the
`ter communications systems where the diver's breath
`speech microphone 14 in such a way that resonances
`ing noises and gas flow noises may actuate a standard
`are removed, the power content of the signal is in
`VOX circuit. Similar problems also occur in other high
`creased and the intelligibility in noise is improved. The
`noise environments.
`speech processor 18 comprises a speech pre-amplifier
`20
`It is, therefore, a general object of the present inven
`20, a highpass filter 22, and a speech amplifier/clipper
`tion to provide a voice actuated transmitter switch or
`24.
`VOX that is substantially insensitive to non-speech
`The implementation of the circuitry of the speech
`generated sounds.
`pre-amp, highpass filter and speech amplifier/clipper
`It is a specific object of the present invention to pro
`25
`are well-known in the art and need not be described in
`vide a voice actuated transmit switch for use in a high
`detail. However, preferably the speech pre-amplifier 20
`noise environment voice communication system.
`comprises a junction field-effect transistor connected
`It is another object of the invention to provide a voice
`as a common-source amplifier having again of approxi
`actuated transmit switch which is especially suited for
`mately 20 decibels from 500 to 10,000 hz., with an
`underwater communication systems.
`input impedance of 1 megohm. The function of the
`It is still another object of the invention to provide a
`speech preamplifier 20 is to boost the output of the
`voice actuated transmit switching circuit which is in
`speech microphone 14 to approximately 500 mV peak
`sensitive to breathing noise.
`to-peak for further processing.
`It is a feature of the invention that the circuitry
`The active highpass filter 22 comprises a fourth
`thereof can be easily incorporated in existing commu
`35
`order filter having a Butterworth response with a cor
`nication systems.
`ner frequency of 1 100 ha. It comprises two second
`It is still another feature of the invention that opti
`order sections in series, synthesized by means of RC el
`mum speech intelligibility is maintained while at the
`ements and unity-gain source-follower JFET amplifier.
`same time providing for voice actuation of the commu
`Since the speech information below 1000 ha are em
`40
`nication system transmitter in a high noise environ
`phasized by the speech cavity 12 at a rate of 12 decibels
`per octave, the output of the highpass filter is a speech
`ment.
`These objects and other objects and features of the
`signal in which components below the corner fre
`invention will best be understood from a detailed de
`quency roll off at a rate of -12 decibels per octave.
`scription of a preferred embodiment thereof, selected
`This type of frequency response results in virtually
`45
`for purposes of illustration, and shown in the accompa
`complete elimination of the first formant of the speech
`nying drawings, in which:
`signal, which is not essential to good intelligibility, and
`FIG. 1 is a block diagram of a communication system
`of low-frequency noises arising either in the water or in
`utilizing the voice actuated transmit switch of the pres
`the speech cavity.
`ent invention; and,
`The speech amplifier/clipper 24 preferably.com
`50
`FIG. 2 is a schematic diagram of the VOX circuit.
`prises a bipolar transistor connected as a common
`Turning now to the drawings, the voice actuated
`emitter amplifier having a gain of approximately 40
`switch or VOX of the present invention will be de
`decibels. Two silicon diodes in parallel, but in opposite
`scribed in connection with an underwater communica
`directions, are capacitively coupled between the base
`tions system shown in block diagram form in FIG. 1 and
`and collective terminals of the transistor. The resulting
`55
`indicated generally by the reference numeral 10. It
`non-linear negative feedback allows signals below ap
`should be understood that the description of the inven
`proximately 10mv peak-to-peak to be amplified with
`tion in connection with the underwater communica
`little distortion. Signals above this level are heavily
`tions system is by way of illustration only and that the
`clipped. The maximum output from this stage is ap
`voice-activated transmit switch can be used in other
`proximately 1.2 V p-p, regardless of input level.
`communication systems and that it is particularly suited
`The net effect of filtering the speech signal as de
`for high noise environment communication systems.
`scribed above and clipping it is to increase its average
`Looking at FIG. 1 which depicts the illustrative un
`power content and its intelligibility in noise by reinforc
`derwater communications system, the diver interface
`ing essential portions of the frequency spectrum. The
`comprises a speech cavity 12 and a speech microphone
`process of clipping generates harmonics which are re
`65
`14. Preferably, the speech microphone 12 is a high
`moved by an active low-pass filter 26 which is a second
`impedance, piezoelectric unit which is suitable for the
`order Butterworth filter having a corner frequency of
`underwater use. The microphone is enclosed in the
`3.5 khz. This filter is also synthesized with RC elements
`
`30
`
`60
`
`- 4 -
`
`

`

`10
`
`15
`
`3,746,789
`3
`4.
`and a unity-gain JFET source-follower. The corner fre
`duction headphone 52. The input pre-amplifier 40
`quency of 3500 ha is high enough to preserve frequen
`preferably is a low-noise JFET connected as a com
`cies which are essential to good intelligibility.
`mon-source amplifier with a gain of approximately 20
`The transmitter portion of the underwater communi
`db at the carrier frequency. The amplifier 42 provides
`cations system indicated generally by the reference nu
`60 db of linear gain and the limiter 44 provides approx
`meral 28, comprises a voltage controlled oscillator 30
`imately 40 db of the gain before limiting. The gain and
`and a tuned power amplifier 32. The linear VCO 30 is
`limiting have been chosen to provide a useful limited
`set to operate a center frequency of approximately 40
`signal of at least 100 mv rms with a minimum input of
`khz. The function of the VCO is to provide a frequen
`1 microvolt rms from the transducer 34.
`The bandpass filter 46 is a simple parallel-tuned LC
`cy-modulated carrier. Its transfer characteristic is such
`that changes in control voltage of plus or minus 1000
`resonant circuit placed at the output of the limiter 44
`mV cause the center frequency to change by plus or
`in order to restore the sinusoidal character of the sig
`minus 3.5 khz. Thus, if the maximum speech frequency
`nal. The output from the bandpass filter 46 is applied
`to be transmitted is 3.5 khz the modulation index is 1.0
`to the phase-locked loop 48 which is used as a fre
`and the speech signal can be transmitted using a total
`quency demodulator. The VCO of the phase-locked
`bandwidth of 7 khz.
`loop is set to operate at approximately 40 khz, and its
`The tune power amplifier 32 preferably comprises a
`low-pass filter elements are chosen to provide a capture
`single-transistor class-C tuned power amplifier. The
`range of at least plus or minus 4 khz. This allows the de
`amplifier delivers 500 mW of electrical power into a
`modulator to lock onto any signal whose frequency is
`1000-ohm resistive load. Since the output of the volt
`within 10 percent of the nominal carrier frequency,
`20
`age controlled oscillator 30 is a square wave the input
`This relatively large capture range makes it possible to
`of the tuned amplifier contains a series resonant circuit
`utilize simple RC oscillators in the transmitter, since
`frequency stability is unimportant.
`to eliminate harmonics of the desired carrier fre
`quency. A tuned transformer matches the collector of
`The output of the phase-locked loop 48 drives an
`the transistor to a transducer 34. The secondary of the
`25
`audio amplifier 50 having a maximum power output of
`transformer is designed to resonate at the carrier fre
`150 mW and a sensitivity of 45 mV rms for pull-power
`quency of 40-khz with a 3-db bandwidth 7 khz, corre
`output. The audio amplifier is transformer coupled to
`sponding to an effective Q of 5.72. The output of the
`a high-impedance piezo-electric bone -conduction
`tuned power amplifier 32 is connected through a trans
`headphone 52 which couples the audio signal into the
`mit receive gate 36 to the previously mentioned trans
`30
`diver's ear.
`ducer 34. The transducer is a resonant, air-filled tubu
`Having briefly described the speech microphone,
`lar ceramic transducer whose dimensions and material
`transmitter, and receiver portions of the underwater
`are chosen to resonate at the carrier frequency in the
`communications system 10, I will now describe in detail
`radial mode of vibration. The length-to-diameter ratio
`the voice-activated transmit switch circuitry. It is im
`is approximately 1, so as to provide some directionality
`practical to maintain the transmitter portion of a com
`in planes parellel to the longitudinal axis of the trans
`munication system continuously activated since it con
`ducer. The transducer is capped to increase its receiv
`sumes much more power than any other portion of the
`ing sensitivity and it is encapsulated in a thin sheet of
`system. On the other hand, manual activation of a
`sound-transparent material so as to render it water
`switch for transmission would require a working diver
`40
`proof and capable of operating in at least 250 feet of
`to maintain one hand available for this purpose. This
`sea water. The static capacitance of the transducer 34
`may not be possible at all times and thus may prevent
`is used as part of the capacitance necessary to tune the
`the diver from being able to communicate during a crit
`secondary winding of the tuned power amplifier output
`ical situation.
`transformer so that a separate tuning coil is unneces
`A voice-operated transmit switch (VOX) is desirable
`45
`sary.
`in such a communications system. However, it is not
`Since the power amplifier operates in the class-C
`possible to use the output of the speech microphone 14
`mode, it is unnecessary to disconnect it from the tuned
`for VOX purposes, since it contains breathing noises
`transformer during reception, because the reverse
`which would activate the transmitter 28 every time the
`biased transistor becomes a resistor having a resistance
`diver inhales or exhales. Furthermore, it is not possible
`50
`which is several orders of magnitude larger than the
`to discriminate against these noises on the basis of
`resonant resistance of the transducer 34.
`acoustic levels alone, since these noises are of at least
`The transducer 34 is also used for the detection of
`the same level as speech signals. If a VOX system is set
`acoustic signals for reasons of economy and simplicity.
`so that it will not be activated by these noises, a diver
`The receiver input, therefore, must be effectively dis
`would have to shout into the microphone to make the
`connected from the transducer during transmission.
`system operate. It is not possible to discriminate against
`This is accomplished by the previously mentioned
`these breathing noises on the basis of frequency alone
`transmit/receive gate 36 which comprises a carrier
`either since their frequency spectrum is rather wide
`activated series/shut analog gate having an inductor,
`and erratic.
`two pairs of switching diodes, a series tuning compaci
`Various solutions have been proposed to this prob
`60
`tor and a parallel tuning capacitor connected so as to
`lem. Some form of frequency analysis is usually em
`block the signal from the input of the receiver during
`ployed to distinguish noise from speech, but the com
`transmission.
`plexity required to make this solution rather unenomi
`The receiver section of the underwater communica
`cal for use in a self-contained miniature communica
`tions system, indicated generally by the reference nu
`tions system of the type shown in FIG. 1.
`65
`meral 38, comprises an input pre-amplifier 40, an am
`The present invention utilizes a separate tissue
`plifier 42, a limiter 44, a band-pass filter 46 a phase
`conduction microphone 54 to generate a transmitter
`locked loop 48, an audio amplifier 50 and a bone con
`keying signal. Preferably, the tissue conduction micro
`
`35
`
`55
`
`- 5 -
`
`

`

`5
`
`25
`
`3,746,789
`S
`6
`phone 54 is a throat microphone which is located in the
`A resistor 76 is used to couple the output of the audio
`vicinity of the larynx of the diver, in such a position that
`detector 62 to the input of the Schmitt trigger 68. The
`the tissue vibrations due to modulation of the vocal
`value of this resistor is chosen to provide a delay of ap
`chords (speech) have a much higher amplitude than
`proximately 0.2 seconds after the rectified audio signal
`those due to gas flow or breathing and swallowing
`is removed, so that the transmitter does not turn off
`noises. However, the tissue conduction microphone
`during brief intersyllabic pauses. The total time elapsed
`also can be placed against the chest cavity to detect the
`between the start of the word and full activation of the
`speech sounds. The detection of the speech sounds at
`transmitter is approximately 1 millisecond given the
`very favorable signal-to-noise ratios makes it possible
`circuitry shown in FIG, 2. Since the average syllabic
`to discriminate against unwanted noises on the basis of
`length is approximately 24 milliseconds, there will be
`O
`amplitude alone, without having to resort to more elab
`no loss of syllables during normal conversation.
`orate signal processing techniques. At the same time,
`The primary advantage of using an underwater
`optimum speech intelligibility is assured by the use of
`speech communication system is the enhancement of
`the separate speech microphone 14 positioned directly
`diver safety, resulting from the capability of the diver
`in front of the mouth of the diver.
`to inform others of his own situation, and of others to
`The tissue-conduction throat microphone 54 should
`recognize potential dangerous situations from informa
`be in good contact with the diver's neck tissue. The op
`tion supplied by the diver. In a case of portable commu
`timum location varies from individual to individual and
`nication devices, however, it is not practical to main
`it must be determined by each individual by trial and
`tain the diver's transmitter activated continuously. Cer
`error. This is usually a one-time operation which is eas
`20
`tain types of diving accidents may take place during
`ily performed in the water. The throat microphone 54
`diver silence periods. A specific example is the loss of
`is isolated from waterborne noises by a layer of sound
`consciousness due to low oxygen concentration in the
`absorbing material 56. Preferably the sound absorbing
`breathing medium which takes place in a very gradual
`material 56 comprises a layer of 3/16 inch mark cellu
`manner and usually goes unnoticed by the victim. If an
`lar rubber or neoprene of the type normally used to
`unconscious diver is not within visual range and imme
`construct wet diving suits. If the diver wears a diving
`diate reach of the other divers, his problem may go en
`hood of this material, the best location for the throat
`tirely unnoticed until it is too late for rescue.
`microphone is under the neck portion of the hood. Oth
`One solution to this problem is to use telemetry tech
`erwise, a simple holder (not shown) is used to maintain
`niques for monitoring diver physiological functions by
`the microphone in place and to provide acoustical iso
`sensing cardiac rhythm, blood pressure, and body tem
`lation from the water.
`perature or environmental conditions, such as breath
`The output from the throat microphone 54 is applied
`ing medium pressure or partial pressure of critical
`to a bandpass filter 58 having a center frequency
`gases. The technology for accomplishing this type of
`slightly below 1 khz and a passband of 40 to 100 ha.
`monitoring exists at the present time, but the equip
`The purpose of the bandpass filter 58 is to eliminate
`35
`ment required is usually so expensive as to be justifiable
`high or low-frequency noises, either waterborne or gen
`only when the objective of monitoring is the acquisition
`erated by the diver which may accidentally trigger the
`of bio-medical data.
`VOX system. This filter is particularly important when
`In my co-pending application, filed of even data here
`the diver is using a single-hose demand regulator hav
`with, and entitled PHYSIOLOGICAL ALARM SYS
`ing an exhaust port only a few inches away from the
`40
`TEM, there is described an alarm system which is espe
`throat microphone, since the noise generated by the ex
`cially suited for an underwater communication system.
`haust bubbles would otherwise cause the VOX system
`The alarm system monitors the diver by means of facili
`to operate every time the diver exhaled.
`ties already present in the underwater communications
`The output from bandpass filter 58 is applied to a
`system and evaluates the data within the system itself
`high gain amplifier 60 which provides sufficient gain to
`45
`and triggers the transmission of the alarm signal only
`amplify the signal from the bandpass filter to a level
`when anomalous conditions are detected. The alarm
`sufficient to drive an audio detector 62. The output
`system is shown in block diagram form in FIG. and
`from the amplifier 60 is rectified by the combination of
`will be briefly described in order to show its interrela
`diode 64 and capacitor 66 (FIG. 2) to produce a DC
`tionship with the VOX system of the present invention.
`50
`level at the input of Schmitt trigger 68. The value for
`The speech microphone 14 is exposed to speech sig
`resistor 70 and the smoothing capacitor 66 are chosen
`nals which are generated at randon intervals, and to
`so that the resulting time constant is less than 500 mi
`breathing and gas flow noises which take place at regu
`croseconds, to assure fast actuation of the VOX sys
`lar intervals. The time distribution of diver-generated
`te.
`signals which are conveniently present at any point in
`Looking at FIG. 2, the Schmitt trigger circuit 68 has
`the speech processor 8, can be used as a means of de
`a high input impedance, an upper trip point of approxi
`tecting anomalous respiratory conditions.
`mately 1.2 V, and a hysteresis of approximately 0.2 V.
`it is obvious that the diver must breathe at least once
`The function of the Schmitt trigger is to provide sharply
`during a certain period of time from which a definite
`defined ON/OFF levels and some degree of immunity
`upper bound T. exists regardless of exertion level and
`to low-level randon noise inputs. The output of the
`individual breathing habits. If a signal is not produced
`Schmitt trigger 53 is used to activate a receiver power
`within this period of time, the diver may be assumed at
`switch 72 and a transmitter power switch 74). These
`best to be breathing too slowly, or at worst, to have
`switches comprise PfP transistors which are con
`stopped breathing altogether.
`nected to the positive power supply and are activated
`Similarly, the diver or his breathing equipment must
`65
`by the combination of inverters shown in F.G. 2 so that
`be silent at least once during the breathing cycle or at
`the receiver power is OFF when transmitter power is
`the end of sentences during the generation of speech.
`Civ and vice versa.
`Thus, another upper bound T, exists for the maximum
`
`30
`
`55
`
`60
`
`- 6 -
`
`

`

`3,746,789
`7
`8
`duration of a diver-generated signal. If this upper band
`26 and 30 is deactivated while the input section of the
`is exceeded, the diver may be assumed to be breathing
`VOX system comprising bandpass filter 58, amplifier
`abnormally. This condition may come about, for exam
`60 and audio detector 62 are activated. Thus, when the
`ple, if a demand regulator fails and delivers a continual
`diver speaks, the throat microphone 54 detects the
`stream of gas into the diver's speech cavity 12 or if the 5
`speech signal which after amplification and detection
`speech cavity is accidentally removed from the diver's
`turns OFF the receiver power switch 72 and then turns
`face and gas flow into the water generates a continuous
`ON the transmitter power switch 74. This activates the
`noise.
`low pass filter 26 and the voltage controlled oscillator
`The detection of these abnormal conditions is per
`30, which causes a modulated carrier to be delivered to
`formed by the circuitry shown in block form in FIG. 1. 10
`the tuned power amplifier 32 which begins to draw cur
`Specifically, an audio input signal is obtained from the
`rent. The output of the power amplifier causes the
`output of the speech amplifier/clipper 24. The audio
`transmit/receive gate 36 to block the signal from the
`signal is detected by audio detector 78 which provides
`input of the receiver as soon as the transmitter output
`a DC voltage to another Schmitt trigger 80 whenever
`voltage reaches a predetermined level. The transmitter
`a signal appears at the input of the audio detector. The 15
`is turned OFF and the receiver is turned ON again as
`output from Schmitt trigger 80 is applied to a noise tim
`soon as the diver stops speaking.
`ing gate 82 and a silence timing gate 84. Both timing
`When the function selector switch 96 in in Position
`gates are outputted to a threshhold detector 86 which
`No. 3, the input section of the VOX system (reference
`produces an output signal if the output from either of
`numerals 58-62) is deactivated, so that the transmitter
`the timing gates exceed a predetermined level which 20
`does not operate when the diver speaks. This "listen
`represents an abnormal condition. The output of
`only' position is useful when the diver is under great
`threshhold detector 86 is used to actuate an alarm
`exertion and is generating so much noise (gasping, for
`audio oscillator 88 which is coupled to the transmitter
`example) that the throat microphone 54 receives a suf
`28 through lead 90. The output of the threshold detec
`ficiently strong signal to activate the VOX system or in
`tor is also applied to the input of the VOX Schmitt trig- 25
`situations in which the diver is forced to observe com
`ger 68 through input terminal 92 (See FIG. 2).
`munications silence. However, the alarm system 98 and
`The overall result of the circuitry just described is
`the switching portion of the VOX system are fully oper
`that whenever a signal is present at the input of the
`ationable so that an abnormal condition can still be de
`audio detector 78 for a time longer than T or a signal
`tected.
`is absent from the input to the audio detector for a pe- 30
`When the function selector switch is in Position No.
`riod longer than T, the audio alarm oscillator 88 is ac
`4, the input section of the VOX is deactivated but a
`tivated, the VOX system is triggered through inputter
`control voltage B4 is applied to the VOX Schmitt trig
`minal 92, the transmitter 28 is activated, and a signal
`ger 68 so that the receiver power switch 72 is in the
`containing the audio alarm note is transmitted through
`OFF mode and the transmitter power switch 74 is con
`the water.
`tinuously ON. This mode of operation is useful when
`The alarm circuit is deactivated if the diver resumes
`the diver's speech and/or breathing must be monitored
`normal breathing. If the abnormal condition persists,
`continuously. An alarm condition causes no change in
`the radiated alarm signal can be used in conjunction
`the system other than the application of the audio
`with underwater homing devices to locate the victim.
`alarm signal to the input of the transmitter 28.
`Although the alarm circuit just described is specifi
`Finally, when the function selector switch 96 is in Po
`cally designed for evaluating speech and breathing
`sition No. 5, the underwater communication system is
`noises, it will be appreciated that it is also suitable for
`in the full "Alarm' mode continuously. This mode of
`evaluating cardiac rhythm. In that situation, the noise
`operation is useful when the diver wishes to inform oth
`timing gate 82 is omitted and the time constant of the
`ers of an extremely dangerous situation, or wishes to be
`45
`silence timing gate 84 is greatly reduced to accommo
`located by means of his radiated signal.
`date a normal human heart rhythm.
`It will be appreciated from the preceding description
`Power for operating the underwater communication
`of the VOX system of my invention that numerous
`system 10 is obtained from a conventional power sup
`modifications can be made without departing from the
`ply 94 and distributed to the appropriate circuits
`50
`scope of the invention as defined in the appended
`through a function selector switch 96. Looking at both
`figures of the drawings, the function selector switch 96
`claims.
`comprises a two-pole, five position switch which can be
`claim:
`1. In a voice communication system having a trans
`operated by the diver by means of an external knob
`mitter, a receiver, and a speech microphone acousti
`(not shown). The speech processor 18, the alarm sys
`55
`cally coupled through a gaseous medium to a speech
`tem 98, the tuned power amplifier 32 and the Schmitt
`source signal for generating a modulation signal for
`trigger, receiver power switch, and transmitter power
`switch portions 68,72, and 74, respectively, of the
`said transmitter, a voice actuated transmit switch com
`prising:
`VOX system are connected to the power supply 94
`l, a separate tissue-conduction microphone; and,
`whenever the function selector switch is in any position
`60
`2. means responsive to an output signal from said mi
`other than Position No. 1. The tuned power amplifier
`crophone for enabling said transmitter.
`32 does not draw any current unless the VCO30 is acti
`wated, since it operates in the class-C mode.
`2. The transmit switch of claim 1 wherein said tissue
`conduction microphone is a throat microphone.
`When the function selector switch is in Position No.
`2, the receiver power switch 72 is ON and delivers cur- 65
`3. The transmit switch of claim wherein said tissue
`conduction microphone is a chest microphone.
`rent to the receiver circuits identified in the block dia
`gram by the reference numerals 40 through 50, the
`4. The transmit switch of claim 1 further character
`transmitter power switch 74 is OFF, and the transmitter
`ized by means for disabling said receiver.
`
`40
`
`35
`
`- 7 -
`
`

`

`3,746,789
`9
`5. In a voice communication system having a trans
`mitter, a receiver, and a speech microphone acousti
`cally coupled through a gaseous medium to a speech
`source for gener