US2121778A - Sound translating apparatus - Google Patents

Description

June 28, 1938. s. BALLANTINE 2,121,778

4 SOUND TRANSLATNG APPARATUS Filed Feb. 12, 1935 2 Sheets-Sheet 1 June 28, 1938. s. BALLANTINE l SOUND TRANSLATING APPARATUS Filed Feb- 12, 1955 2 Sheets-Sheet 2 Patented June 28, 1938Y I l Il Y l UNITED STATES PATENT OFFICE n A'2,121,228 t .SOUND TRANSLATING APPARATUS Stuart Ballantine, Mountain Lakes, N. I. u Application February 12, 1935, sci-iai Nc. 6,245

' 14 omme- (ci. 11s-1) This invention relates to sound translating apessary for intelligible transmission in a, tele-.

paratus and particularly 'to sound translating phone circuit, since these are the principal com. apparatusadapted for incorporation in an elec- .-stituents Yof the majority of consonants.

trical communication system. Iv have discovered that the higher speech fre- The ordinary ktypes of microphones, designed quencies are represented, in appreciable magni- 5 to be actuated by aerial sound waves, must be tudes, in the vibratory energy available at the supported in front of, and preferably. close to, larynx, and that the frequency-vibration ampli- ,the mouth. There are many applications and, tude characteristic is such that a high degree of uses for microphones, however, where this is quite naturalness in the reproduced sound may be ob A 1o 'inconvenient even in those cases where somel tained by restoring the balance in the frequency 101 means of support, other ythan the hands, are distribution oi energy. Broadly speaking, this is provided. Examples of such use areY in gas and accomplished by designing the pickup device and oxygen masks. According to this invention, I `its associated apparatus' to be much more ein-- abandon the method of picking up the sound cient at high frequencies than at low frequen- .lll5 from the air and instead pick up the sound as cles, or in other Words, to have a, rising fre- 15 it exists in the vibration o ,the upper parts of quency response characteristic. the body which are set up by the acoustical and Objects of the invention are to provide novel mechanical actions of 'the voice. These vibramethods and apparatus for converting into elections are particuia'riy strong ini-,he region of the trical currents, for transmission over an electri- 20 larynx and associatedV cartilaginous structures cal system, the acoustical vibrations of certain 20.

such as the thyroid cartilage. Devices embody- Darts 0f the body. Objects are t0 provide mething the invention include apparatus that is apods of and apparatus for producing electrical plied to the vibrating parts so as to be directly R currents from mechanical vibrations of theactuated by the mechanical vibrations thereof throat, the electrical currents being such that,

25 and 'acoustical transmission through the air plays when converted back to sound, the sound reg5 -a minor role. production will closely resemble the original voice Throat microphones have been proposed but, of the person as heard in the ordinary way or so far as I know, none .of the prior devices has as picked up by a microphone of conventional been successful in yielding an electrical output type designed for aerial sound `waves.' 3o which was a faithful replica of the'natural sound A further object is to provide apparatus for. 3th of the voice. The sound reproduced by the 'prior use in an electrical communication system, the. devices was composed largely of low frequency apparatus being designed for actuation by mecomponents and was muilled,rdeep and booming, chanical vibrations of the throat and being capabeing so highly unintelligible as to be totally unble of producing electrical currents which are t for electrical communication purposes. comparable with the currents produced by a good 36;

I have made an experimental study. by wave microphone when favorably positioned for actuanalysis, of the mechanical vibration amplitudes ation by the voice or sound waves in air. More of the larynx at different .frequencies and have particularly, an object is to provide a mechano-` found the reason for the failure of the prior electrical transducer adapted to be placed in con-z 40.V devices. I, have found that the vibration amtact with and actuated by the mechanical vibra.. i0-` Pltlldes 0f the larynx' diminish Very rapidly as tions of the body and a. transmission network cothe frequency increases and, COHSeCillently, the operating with the transducer to deliver electrical priorart devices did not yield intelligible sounds, Waves ,vmh when impressed u'pon a telephone for the devices were designed Simply t0 reproduce line or other electrical communication system,

the vibrations as they exist with total disregard result .in a reproduction of sound that has the.

oi their spectral distribution.` acoustical quality of the voice. This frequency distribution of the vibratory. T] energy is reconcuable with the fact that many invenzsiirlnwdilcl'tllxier ggarnngrlhgfhgw;

of the sounds of speech, involving high frequency constituents, are formed in the upper specification when taken with the accompanying liti drawings, in which:

cavities by the lips and tongue and probably reach the larynx, much attenuated, by a back- F15 11S a curve sheet ShW-lng 9' frequency-Ye'- ward acoustical' transmission which may be more spons@ 'hm'actel'istc Which Will Yield highly ar, aerial than mechanica. It is wcii established ticulated Speech from the mechanical fibratn that these higher frequency components are nec-A ot the larynx;

Fig. 2 is a circuit diagram of a sound reproducing system which embodies the invention;

Fig. 3 is a curve sheet showing the frequency response characteristicspf the Fig. 2 apparatus;

Fig. 4 is a fragmentary circuit diagram of another embodiment of the invention; Fig. 5 is a curve sheet showing the frequency response characteristics of the microphone and of the complete system of Fig. 4; and

Fig. 6 is a fragmentary circuit diagram illustrating a modification of the Fig. 4 circuit.

I have determined, from my studies of the relation between frequency and the vibration amplitudes of the larynx, that there is a general progressive decrease in amplitude towards the higher end of the useful voice frequencies of from- 200 to about 4,000 cycles and have determined empirically a. frequency response characteristic which may be employed with satisfactory results to correct for the non-uniform energy input at different frequencies. The frequency response characteristic of the apparatus can be described,

for explanation of this invention and for engineer-ing purposes, by means of a curve showing the output voltage of the system when the micro-` phone is vibrated mechanically by a piston which is driven sinusoidally at constant velocity at all frequencies. The term constant velocity" means that the maximum value of the velocity (which varies sinusoidally) is maintained constant. This method of defining the frequency characteristic has been adopted in view of the difculty of actuating the microphone over the entire frequency range for measurement purposes when it is worn on Vthe throat. The mechanically driven piston is used merely as a convenient way of imparting a definite and reproducible actuation to the microphone.

The frequency response or transmission characteristic shown by curve A of Fig. 1 has been found, experimentally, to yield highly articulated speech from the mechanical vibration of the larynx. The ordinates represent the Velectrical voltage output while the abscissae indicate the corresponding frequencies at which the microphone was vibrated, as explained above, by a constant velocity piston. To obtain this constant velocity, the amplitude of vibration varied inversely as the frequency. As shown by the dotted line A', the output of the pickup system varies as the 1.3 power of the frequency over the greater Dari. of the curve A. It willv be seen from this curve that a considerable readjustment of the relation between high and low frequencies is required to compensate for the frequency distribu- Ation in the larynx. It must be understood, of

course, that the shape and slope of the curve is not critical, but may be varied somewhat from that shown in Fig. 1. The diagram is intended to show the order of the effect and to indicate why a device of conventional design (for uniformresponse at all frequencies) is doomed to failure. It must also be pointed out that the anatomical and other variations between individuals and of the position of the microphone on the throat would make a closer specification of a desirable characteristic of doubtful practical value.

The microphone should respond over as great a frequency range as possible, but it has been found in practice that good telephone communication can be obtained by the inclusion ci' frequencies up to from 3000 to 5000 cycles. 'I'he curve in Fig. 1 represents the performance of an actual apparatus including a microphone of the piezoelectric type as described below and in more vfound economical or expedient. It may be, for

example, incorporated solely in the microphone, or partly in other apparatus. It will usually be advantageous to obtain the frequency characteristic as near to the microphone as possible, in order to avoid the necessity of altering existing circuits into which the apparatus is to work.

A representative embodiment of this invention is shown in Fig. 2. This represents a complete telephone system of a simple type. Here the microphone I comprises a carbon-granule cell of conventional type which is provided with a button 2 which is driven by the vibration of the larynx. The microphone output is fedinto the network Lilla, RiRz, C1C2 of the constant resistance type, then through transformer T1, the line, and transformer T2 to the telephone receiver 3. 'I'he condenser Ca serves to confine the direct current from battery 4 to the microphone circuit. The desired frequency characteristic, shown in Fig. 1, is obtained by means of the network in combination with the mechano-electrical characteristic of the microphone. If the microphone is so designed as to have a constant output for a given displacement at all frequencies, its output voltage, for constant velocity excitation, will vary inversely as the first power of the frequency as shown by curve B, Fig. 3. The network provides the rising transmission characteristic, curve C, and these transmission characteristics combine to provide the required characteristic, curve D.

'Ihe apparatus shown to the left of line ab of Fig. 2 maybe replaced by the equipment illustrated in Fig. 4. The device I which converts in parallelso that their effects are additive electrically, and are shielded electrically by the metallic housing 5 and shielded cord 6, one terminal of `each piezoelectric unit being grounded to the shield.

The output terminals of the microphone are connected to the input terminals of the iirst tube i of a two-stage amplifier. These input terminals are shunted by a resistance R1 which serves as a grid leak to fix the direct current bias on the grid G1 of tube 1 and which has the further function of establishing a definite relation between the alternating current voltage applied to the grid and the voltage generated by the microphone, the relation being such that the input voltage rises with frequency. A part of the desired frequency characteristic of the whole system may thus be obtained at this portion of the circuit. This eiect is obtained by selecting a value for resistance R1 which is lower than the capacitive reactance of the piezoelectric crystals in the microphone at the higher frequencies. A rela-- tively low value of resistance R1 has other advantages in that it tends to minimize any variation in output due to leakage in the microphone crystals or the connecting cord.

The relation between the voltage across resistance R1 and the voltage generated by the typical dotted line curve `quency components of the electrical currents were generated from the mechanicalvibrations of the body at conversion ratios that increased with frequency in the useful voice frequency range. The conditions under which this curve was obtained were as follows: the two crystals connected in parallel each had a capacity of I.001 microfarad atV 20 C., and resistance R1 was 50,000 ohms.

The rising frequency characteristic of the microphone is supplemented by designing the amplifier, principally the first stage, to have a. rising frequency characteristic, i. e. a gain that rises with frequency. `The plate circuit of the pentode tube 1 of this stage includes the primary winding of the coupling transformer T1 in paralcircuit.

lel with a resistance Rs, the reactance of the transformer being low compared with the resistance R3 and plate resistance in parallel, thus providing a gain characteristic which rises with frequency. The transformer T1 is in fact made resonant at a frequency near the upper limit of f the desired frequency range. The frequency characteristic of the second stage, tube 8; is substantially flat.' By shunting the output transformer T2 th a resistance R5, the output impedance, viewed from the output terminals, may be made a substantially pure resistance at all frequencies. In this way, the microphone and associated amplifier may be adapted to replace the conventional carbon microphone of about 100 to 200 ohms resistance without aifecting the frequency characteristics of other parts of the To limit regeneration, the terminals of transformer T1, as shown, are reversed from their normal polarity. Grid biases for vthe two amplifier tubes are obtained from the voltage drop through the cathode resistances R2, The condensers C1, C2, C3 are the usual by-pass condensers to limit intercircuit coupling, resisty ances ReRi are decoupling resistances in the plate circuits, and resistance Ra is a so-called bleeder. The overall frequency characteristic of the ampliflerisshown by the curve F of Fig. 5 in which the ordinates represent the output voltage plotted against frequency for a fixed voltage applied across the resistance R1. The relation between the voltage output terminals of the amplifier andfrequency, for a constant velocity excitation of the microphone, is shown by curve A of Fig. 1.

It is not always convenient mechanically to the-'microphone and in such of the micrephone is in a radio frequency radiation iield, some radio frequency energy may leak to the amplifier. Examples of such situations are found in open cockpit airplanes where the microphone may be used to modulate a radio transmitter. radio frequency voltages may be prevented from reaching the grid of the amplifier tube by a filter network. such as shown in Fig. 6. The general form of the amplier circuit may be substantially identical with that shown in Fig. 4, and certain elements of the Fig. 6 circuit are therefore identifled by the correspondingreference characters of Fig. 4. The filter circuit for blocking the transmission of radio frequency voltages comprises the choke coil L and the shunt condenser C which are connected between the microphone terminals and the input terminals of the amplifier.

completely shieldcases, when the wearer case is shown by the l and R4, respectively. v

Such stratl It will be understood that there is considerable elements of the sound translating apparatus, and in the frequency response characteristics of the several elements. The desired frequency response characteristic of the complete system may be obtained in a variety of ways other than those herein described. In the Fig. 4 circuit, for example, the resistor R1 could be made larger and the resulting reduction in the power of the response versus frequency relation can be compensated in the design of the microphone or the amplifier.

It will therefore be apparent that the invention is not restricted to the particular methods and embodiments herein described and that various changes which will occur to those familiar with this art fall within ,the scope of my invention' as set forth in the following claims.

I claim:

1. Sound translatingapparatus comprising a mechano-electric transducer substantially nonresponsive to sound waves in air and actuated by the vibrations of the body due to the voice, and an electrical network for transmitting the electrical output of said transducer, said transducer and network having an overall frequency response characteristic which rises progressively with frequency from 200 to about 4000 cycles when the transducer is vibrated at constant velocity, whereby the electrical current output of said network may be converted into sound waves hav ing an acoustical quality comparable to that of the voice..

2. Sound translating apparatus as claimed in claim 1, wherein said transducer has a frequency responseLcharacteristic which falls off with increasing frequency, and said network has a characteristic which rises with frequency at a rate which overcompensates the falling vcharacteristic of said transducer.

3. Sound translating apparatus as claimed in claim l, wherein both said transducer and said network have frequency response characteristics which rise with frequency.

4. Sound translating apparatus comprising a mechano-electric transducer susbtantially nonresponsive to sound waves in air and actuated'by the vibrations of the body due to the voice, and

an electrical network for transmitting the electrical output of said transducer, said transducer and network having an overall frequency response characteristic which rises as a predetermined power of the frequency over the voice frequency rangev of from 200 to about 4000 cycles when the transducer is vibrated at constant velocity, whereby the electrical work may be converted into sound waves having an acoustical quality comparable to that of the voice. m5,.,-Sound translating apparatus comprising a A a'rfi'echano-electric transducer substantially nonresponsive to sound waves in air and actuated by the vibrations of the body due to the voice, and an electrical network for transmitting the elec.- trical output of said transducer; said transducer and network having an overall frequency response characteristic which rises as'V a power of the frecurrent output of said netquency which is not lessithan unity over the use ful voice frequency ,range of from 200 to about 4000 cycles when-the transducer is vibrated at constant velocity, whereby the electrical current output of said network may be converted into sound waves having an acoustical quality ,comparable to that of the voice.

6. 'An electrical communication system comprising a mechano-electrical transducer substantially non-responsive to sound waves in air and actuated by the mechanical vibrations of the body due to voice, a transmission circuit, and a sound reproducer, said system having an overall transmission eiiiciency which produces an output that rises progressively with frequency in the useful voice frequency range of from 200 to about 4000 cycles when the transducer is vibrated power of the frequency over a substantial part of the useful voice frequency range of the system of from 200 to about 4000 cycles when the transducer is vibrated at constant velocity, whereby the reproduced sound has the acoustical quality of the voice.

8. In sound translating apparatus, combination with a throat microphone of the piezoelectric type, said microphone having a capacitive reactance over its frequency range of response, of a resistor shunted across the output terminals of said microphone, the resistorhaving a resistance less than the capacitive reactance offs'aid microphone over the greater part of the said frequency range of response, whereby the relation between the output voltage across the microphone terminals and voltage generated by the microphone rises with frequency.

9. In sound translating apparatus, the combination with a throat microphone of the piezoelectric type, said microphone having a capaci.

tive reactance over its frequency range of respouse, a resistance shunted across the output terminals of the microphone, a vacuum tube and connections between the input terminals thereof and the microphone terminals, and an impedance network cooperating with said tube to effect operation thereof as an amplifier, `the overall transmission emciency of said microphone and amplier varying with frequency to produce a relation between 4amplifier output voltage and the voltage generated by the microphone which rises with frequency.

.10. In sound translating apparatus, the combination with a throat microphone including a. transducer unit, and a. vacuum tube amplifier stage having output terminals; of means for developing across the output terminals of said amplifier stage an output voltage which rises with frequency when said microphone is vibrated at constant velocity, said means including an. additional amplifier stage between said microphone and said first-mentioned amplifier stage, the additional amplifier stage having a rising frequency characteristic.

11. 'Ihe invention, as claimed 'in claim 10, wherein the additional amplifier stage includes a transformer resonating at a frequency near the upper limit of the range of frequencies vto be transmitted thereby.

12. The invention as claimed in claim 10, in combination with a lter network between said microphone and the additional'amplifier stage to prevent the application of radio frequency voltages to the additional amplifier stage.

13. 'Ihe process of reproducing speech which comprises the steps of translating mechanical vibrations of the body due to the voice into elecc trical currents, selectively amplifying the several frequency components of the vibration-produc d currents at gains which increase with the frquency of such components in the useful voi `e frequency range of.irom 200 to about 4000 cycles, transmitting the amplified electrical currents t a. desired reproduction point, and converting the transmitted and amplified currents into sound Waves.

14. 'I'he process of reproducing speech which f includes the steps of selectively generating electrical currents from mechanical vibrations of the body due to the Voice at conversion ratios that increase progressively with frequency in the useful voice range of from about 200 to about 4000 cycles, amplifying the resulting electrical currents at gains which increase progressively with frequency within the said useful voice range, transmitting the amplified electrical currents, and converting the transmitted and amplified currents into sound waves. i

STUART BALLANTINE.

Images