| United States Patent |
4,821,326
|
|
MacLeod
|
April 11, 1989
|
Non-audible speech generation method and apparatus
Abstract
A non-audible speech generation apparatus and method for producing
non-audible speech signals which includes an ultrasonic transducer or
vibrator for projecting a series of glottal shaped ultrasonic pulses to
the vocal track of a speaker. The glottal pulses, in the approximate
frequency spectrum extending from fifteen kilohertz to one hundred five
kilohertz, contain harmonics of approximately 30 times the frequency of
the acoustical harmonics generated by the vocal cords, but which may
nevertheless be amplitude modulated to produce non-audible speech by the
speaker's silently mouthing of words. The ultrasonic speech is then
received by an ultrasonic detector disposed outside of the speaker's mouth
and electrically communicated to a translation device which down converts
the ultrasonic signals to corresponding signals in the audible frequency
range and synthesizes the signals into artificial speech.
| Inventors:
|
MacLeod; Norman (Sunnyvale, CA)
|
| Assignee:
|
Macrowave Technology Corporation (San Jose, CA)
|
| Appl. No.:
|
121659 |
| Filed:
|
November 16, 1987 |
| Current U.S. Class: |
704/261; 381/70; 623/9; 704/258 |
| Intern'l Class: |
G10L 007/02 |
| Field of Search: |
381/36-40,51,70,86
364/513.5
623/9
|
References Cited
U.S. Patent Documents
| 3914550 | Oct., 1975 | Cardwell, Jr. | 381/70.
|
| 4292472 | Sep., 1981 | Lennox | 381/70.
|
| 4338488 | Jul., 1982 | Lennox | 381/70.
|
| 4473905 | Sep., 1984 | Katz et al. | 381/70.
|
| 4502150 | Feb., 1985 | Katz et al. | 381/70.
|
| 4550427 | Oct., 1985 | Katz et al. | 381/70.
|
| 4571739 | Feb., 1986 | Resnick | 381/70.
|
| 4612664 | Sep., 1986 | Walsh et al. | 381/70.
|
| 4627095 | Dec., 1986 | Thompson | 381/70.
|
| 4633864 | Jan., 1987 | Walsh | 623/9.
|
| 4672673 | Jun., 1987 | Katz et al. | 381/70.
|
Primary Examiner: Wong; Peter S.
Assistant Examiner: Voeltz; Emanuel T.
Attorney, Agent or Firm: Rosenblum, Parish & Bacigalupi
Claims
I claim:
1. An artificial speech generation device, comprising:
transmitting means for producing an ultrasonic signal and introducing said
ultrasonic signal into the vocal track of a speaker for modulation of said
ultrasonic signal by said speaker into ultrasonic words;
detector means for receiving said ultrasonic words and generating an
ultrasonic word signal corresponding to said ultrasonic words; and
converter means for receiving said ultrasonic word signal and shifting said
ultrasonic word signal into a corresponding artificial speech signal which
can be utilized to create an audio signal corresponding to said speaker's
ultrasonic words.
2. An artificial speech generation device as recited in claim 1, wherein
said ultrasonic signal is a series of glottal shaped pulses.
3. An artificial speech generation device as recited in claim 1, wherein
said ultrasonic signal is an oscillatory shaped wave form.
4. An artificial speech generation device as recited in claim 1, wherein
said transmitting means is comprised of:
signal generator means for producing and outputting said ultrasonic signal;
and
dispatching means for receiving said ultrasonic signal from said signal
generator means and transmitting said ultrasonic signal into the vocal
track of said speaker.
5. An artificial speech generation device as recited in claim 4, wherein
said dispatching means is an ultrasonic transducer positioned so as to
direct said ultrasonic signal into the vocal track of said speaker.
6. An artificial speech generation device as recited in claim 5, wherein
said converter means is comprised of:
speech translation means responsive to said ultrasonic word signal and
operative to convert said ultrasonic word signal to said artificial speech
signal; and
playback means for receiving said artificial speech signal and producing an
audio signal representation thereof for playback to said speaker.
7. An artificial speech generation device as recited in claim 6, wherein
said speech translation means includes:
translator means for receiving said ultrasonic word signal and operative to
separate said ultrasonic word signal into a plurality of frequency bands,
to track the frequency and amplitude component of each said frequency
band, and to develop separate corresponding amplitude and frequency
component signals for output therefrom; and
synthesizer means for receiving said amplitude and frequency component
signals input from said translator means and operative to produce said
artificial speech signal.
8. An artificial speech generation device as recited in claim 7, wherein
said synthesizer means includes:
a plurality of variable resonance circuits for receiving said amplitude and
frequency component signals and operative to produce corresponding
amplitude and frequency signals in the audio spectrum;
noise generating means for producing an electrical noise signal
corresponding to the non-vocal noises required for the production of
comprehensible speech; and
summation means for receiving said amplitude and frequency signals in the
audio spectrum and said electrical noise signal and operative to combine
said amplitude and frequency signals in the audio spectrum with said
electrical noise signals so as to produce said artificial speech signal.
9. An artificial speech generation device as recited in claim 8, wherein
said plurality of variable resonance circuits are operated at about
one-thirtieth of the frequency of said ultrasonic signal introduced by
said transmitting means.
10. An artificial speech generation device as recited in claim 6, wherein
said speech translation means is a vocoder circuit.
11. An artificial speech generation device as recited in claim 4, wherein
said dispatching means is an ultrasonic vibrator positioned so as to
introduce said ultrasonic signal into the vocal track of said speaker.
12. An artificial speech generation device as recited in claim 11, wherein
said converter means is comprised of:
speech translation means responsive to said ultrasonic word signal and
operative to convert said ultrasonic word signal to said artificial speech
signal; and
playback means for receiving said artificial speech signal and producing an
audio signal representation thereof for playback to said speaker.
13. An artificial speech generation device as recited in claim 12, wherein
said speech translation means includes:
translator means for receiving said ultrasonic word signal and operative to
separate said ultrasonic word signal into a plurality of frequency bands,
to track the frequency and amplitude component of each said frequency
band, and to develop separate corresponding amplitude and frequency
component signals for output therefrom; and
synthesizer means for receiving said amplitude and frequency component
signals input from said translator means and operative to produce said
artificial speech signal.
14. An artificial speech generation device as recited in claim 13, wherein
said synthesizer means includes:
a plurality of variable resonance circuits for receiving said amplitude and
frequency component signals and operative to produce corresponding
amplitude and frequency signals in the audio spectrum;
noise generating means for producing an electrical noise signal
corresponding to the non-vocal noises required for the production of
comprehensible speech; and
summation means for receiving said amplitude and frequency signals in the
audio spectrum and said electrical noise signal and operative to combine
said amplitude and frequency signals in the audio spectrum with said
electrical noise signals so as to produce said artificial speech signal.
15. An artificial speech generation device as recited in claim 14, wherein
said plurality of variable resonance circuits are operated at about
one-thirtieth of the frequency of said ultrasonic signal introduced by
said transmitting means.
16. An artificial speech generation device as recited in claim 12, wherein
said speech translation means is a vocoder circuit.
17. An artificial speech generation device as recited in claim 1, wherein
said converter means is comprised of:
speech translation means responsive to said ultrasonic word signal and
operative to convert said ultrasonic word signal to said artificial speech
signal; and
playback means for receiving said artificial speech signal and producing an
audio signal representation thereof for playback to said speaker.
18. An artificial speech generation device as recited in claim 17, wherein
said speech translation means includes:
translator means for receiving said ultrasonic word signal and operative to
separate said ultrasonic word signal into a plurality of frequency bands,
to track the frequency and amplitude component of each said frequency
band, and to develop separate corresponding amplitude and frequency
component signals for output therefrom; and
synthesizer means for receiving said amplitude and frequency component
signals input from said translator means and operative to produce said
artificial speech signal.
19. An artificial speech generation device as recited in claim 18, wherein
said synthesizer means includes:
a plurality of variable resonance circuits for receiving said amplitude and
frequency component signals and operative to produce corresponding
amplitude and frequency signals in the audio spectrum;
noise generating means for producing an electrical noise signal
corresponding to the non-vocal noises required for the production of
comprehensible speech; and
summation means for receiving said amplitude and frequency signals in the
audio spectrum and said electrical noise signal and operative to combine
said amplitude and frequency signals in the audio spectrum with said
electrical noise signals so as to produce said artificial speech signal.
20. An artificial speech generation device as recited in claim 19, wherein
said plurality of variable resonance circuits are operated at about
one-thirtieth of the frequency of said ultrasonic signal introduced by
said transmitting means.
21. An artificial speech generation device as recited in claim 17, wherein
said speech translation means is a vocoder circuit.
22. An artificial speech generation device, comprising:
transmitting means for producing an ultrasonic signal and introducing said
ultrasonic signal into the vocal track of a speaker for modulation of said
ultrasonic signal by said speaker into ultrasonic words;
detector means for receiving said ultrasonic words and generating an
ultrasonic word signal corresponding to said ultrasonic words;
converter means for receiving said ultrasonic word signal and converting
said ultrasonic word signal into a series of analog spectral signatures
representing said ultrasonic words and developing a digitized spectral
signature signal therefrom; and
generating means for receiving said digitized spectral signature signal and
operative to generate an audible speech communication signal from said
digitized spectral signature signal.
23. An artificial speech generation device as recited in claim 22, wherein
said converter means includes:
spectrum analyzer means for receiving said ultrasonic word signal and
developing said analog spectral signature therefrom; and
digitizing means for receiving said analog spectral signature signal and
operative to develop said digitized spectral signature signal.
24. An artificial speech generation device as recited in claim 23, wherein
said generating means includes:
means for receiving said digitized spectral signature signal and operative
to recognize and to separate each of a plurality of speech components
contained within said digitized spectral signature signal, and to develop
a corresponding speech component signal containing each of said plurality
of speech components; and
synthesizing means for receiving said speech component signal and operative
to synthesize each speech component of said plurality of speech components
contained within said speech component signal into said audible speech
communication signal.
25. An artificial speech generation device as recited in claim 24, wherein
said generating means further includes a broadcasting means for receiving
said audible speech communication signal and producing audible speech
therefrom.
26. A method for generating non-audible speech communication signals,
comprising the steps of:
transmitting an ultrasonic signal into the vocal track of a speaker;
modulating said ultrasonic signal within the vocal track of said speaker
into ultrasonic words;
communicating said ultrasonic words out of the vocal track of said speaker;
detecting said ultrasonic words; and
converting said ultrasonic words into non-audible speech communication
signals which may be utilized to silently communicate operation commands
to a plurality of devices responsive to said non-audible speech
communication signals.
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to artificial speech aids, and more
particularly, to a method and apparatus for generating non-audible
artificial speech.
2. Discussion of the Prior Art
A number of devices have been developed which either replace the larynx
(including the vocal cords) or otherwise artificially create
comprehensible speech. The production of speech begins in the lungs which
supply the air that is required to vibrate the vocal cords and carry the
tune that is produced through the throat or pharynx and out of the nose
and mouth. When a person desires to produce speech, the muscles
controlling the vocal cords are tightened, thereby allowing the air
passing through the larynx to cause the vocal cords to vibrate, which in
turn produces a tone. The amplitude of the tone produced by the vocal
cords is proportional to the amount of air supplied by the lungs.
Modulation of the tone into comprehensible speech is caused by variations
in the movement and positioning of the tongue, nasal cavity, throat and
mouth.
Prior art artificial speech generation devices have been primarily designed
to replace the larynx or vocal cords of persons whom have suffered either
a permanent or temporary injury to that portion of their vocal track. Due
to the injury suffered, the person is incapable of producing a tone for
modulation. To solve this problem, U.S. Pat. No. 4,571,739, issued Feb.
18, 1986, to Resnick, discloses using an artificial tooth inserted into
the mouth, constructed out of tone emitting components, for producing a
tone which may then be modulated by the mouth, pharynx and nasal cavity of
a laryngectomy patient. Similarly, U.S. Pat. No. 4,550,427, issued Oct.
29, 1985, to Katz et al, disclose a completely self-contained intraoral
artificial larynx, having its own power source, tongue activated control,
power saving signal generation circuitry, acoustic and audio amplifiers,
and intraoral speaker, all of which are contained within an otherwise
conventional dental prosthesis. Like the Resnick device, the device in
Katz et al is designed for usage by laryngectomy patients, and in addition
to tone production, provides some tone amplification and enhancement. Both
prior art devices are designed to be operated only when the larynx is
inoperative or missing.
Additional prior art devices also supply a certain quantity of air, in
addition to the tone, into the oral cavity. In U.S. Pat. No. 4,612,664,
issued Sept. 16, 1986, to Walsh et al, a solenoid reed valve capable of
producing an amplitude modulated tone is disclosed. The reed valve
replaces the amplitude modulation function of the lungs by regulating the
quantity of air passing through a pressurized air passage created in the
valve, which in turn creates a tone of sufficient amplification. U.S. Pat.
No. 4,627,095, issued Dec. 2, 1986, to Thompson, discloses an artificial
voice apparatus having a pump for producing an air flow within a tube
connected to a sounding mechanism, thereby combining an audible tone with
air flow. The tube is then coupled to a mouth piece and tone is
transported into the person's mouth where the tongue, lips and teeth
articulate the sound into speech. In these prior art devices, it is
assumed that an audible tone is required to replace the vocal cords of the
person using the device. However, if a person simply desires to speak
without using their vocal cords, the transmission of an audible tone into
the person's mouth would make it impossible to create artificial speech
while at the same time restricting the audible emission of that speech
from the mouth.
A large number of applications exist for nonaudible speech communication,
such as in surveillance operations, military operations, or even where a
person simply wishes to not be overheard while speaking in a private
telephone conversation. Likewise, there are a number of situations in
which surrounding or background noise is so great, that common levels of
conversation or even high amplitude speech cannot be heard, such as at
airports, on the battlefield, or in industrial environments. Finally,
there are a number of applications, where audible speech itself, although
desired, may be too awkward or distracting, such as in the case of
dictation where the dictator is apparently only talking to himself. In
such an instance, silently communicating to the dictation unit, which can
record and playback an audible recording, may be an easier and more
convenient means of communication than audible speech, and just as
effective. However, inaudible artificial speech applications have
previously been made unachievable with prior art artificial speech
generation devices.
SUMMARY OF THE PRESENT INVENTION
It is therefore an object of the present invention to provide a non-audible
artificial speech generation device.
Another object of the present invention is to provide a non-audible
artificial speech generation device which allows a speaker to produce the
artificial speech by simply talking, but without use of the vocal cords or
use of unvoiced speech (whispering).
Another object of the present invention is to provide a non-audible
artificial speech generation device which projects ultrasonic pulses to
the vocal tract of a speaker from a ultrasonic transducer or by direct
contact with the throat of the speaker.
A further object of the present invention is to provide a non-audible
artificial speech generation device which projects ultrasonic pulses
shaped like the glottal pulses which are produced by the larynx during the
natural production of speech.
A still further object of the present invention is to provide a non-audible
artificial speech generation device which uses a vocoder operating at
ultrasonic frequency levels to translate and synthesize the non-audible
artificial speech produced by a speaker into the audible range.
A still further object of the present invention is to provide a method for
producing an artificial speech communication signal which can be used to
operate a variety of devices.
Briefly, a preferred embodiment of the present invention comprises an
ultrasonic transducer or vibrator for projecting a series of glottal
shaped ultrasonic pulses to the vocal track of a speaker. The glottal
pulses, in the approximate frequency spectrum extending from fifteen
kilohertz to one hundred five kilohertz, contain harmonics of
approximately 30 times the frequency of the acoustical harmonics generated
by the vocal cords, but which may nevertheless be amplitude modulated to
produce non-audible speech by the speaker's silently mouthing of words.
The ultrasonic speech is then received by an ultrasonic detector disposed
outside of the speaker's mouth and electrically communicated to a
translation device which down converts the ultrasonic signals to
corresponding signals in the audible frequency range and synthesizes the
signals into artificial speech.
These and other objects of the present invention will no doubt become
apparent to those skilled in the art after having read the following
detail disclosure of the preferred embodiment which is illustrated in the
several figures of the drawing.
IN THE DRAWING
FIG. 1 is an anatomical diagram of human speech generating organs (the oral
cavity).
FIG. 2 is a graph illustrating a glottal pulse shaped waveform in
accordance with a preferred embodiment of the present invention, and
depicting the relationship between tone amplitude versus time.
FIG. 3 is a graph illustrating the ultrasonic frequency spectrum of a
glottal pulse shaped waveform in accordance with a preferred embodiment of
the present invention.
FIG. 4A is an illustration of an ultrasonic glottal pulse vibrator and
ultrasonic receiver application of a preferred embodiment of the present
invention.
FIG. 4B is an illustration of an ultrasonic transducer and receiver as
utilized in a telephonic application of a preferred embodiment of the
present invention.
FIG. 5 is a diagram schematically illustrating the application of an
ultrasonic vocoder in accordance with a preferred embodiment of the
present invention.
FIG. 6 is a diagram schematically illustrating an alternative embodiment of
the present invention for converting ultrasonic words into audible speech.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
Referring now to FIG. 1, an anatomical diagram of the speech generating
organs of the human body 10 is shown, which may be utilized to properly
illustrate the human production of speech. These organs and the area which
they encompass, such as the lips and everything behind the lips, is
referred to herein as the oral cavity. Air flows from the lungs 12,
through the wind pipe (trachea) 14 and to the larynx 16. By voluntarily
tightening the vocal cords 18, air passing across the vocal cords can be
made to vibrate the vocal cords and cause them to emit a tone, much like
the production of sound by a guitar string, except that air vibrates the
cord or string rather than a finger or pick. The tone is then carried from
the larynx to the throat 20 and the nasal cavity 22 and mouth for
modulation.
The modulation function of the oral cavity may be demonstrated by fixing
the mouth in an open position while causing the vocal cords to emit a
constant tone. It will be noticed that the only change that can be created
in the vocal output (without moving any parts of the mouth) is either a
change in the pitch of the tone, or the amplitude of that tone. Thus, it
can be seen, that speech is dependent upon two mutually exclusive groups
of organs: (1) the lungs 12, trachea 14, and larynx 16, which create a
tone and control the amplitude of that tone; and (2) the mouth and
pharynx, which combine to modulate that tone into words. Practically any
modification of any part of either group results in either no sound being
produced, or the quality of the sound that is produced to suffer. In
addition to the modulation of audible tones, the teeth 24, lips 26, and
tongue 28 are also responsible to the production of non-vocal hissing
noises that are required for the production of such sounds as s, t, f, p,
sh, ch and k.
Prior art artificial speech devices attempt to replace the larynx and vocal
cords by introducing a standard oscillatory tone into the mouth itself, or
blowing a similar tone on a path of air into the mouth, either of which
can be referred to as illuminating the oral cavity, whereupon the tone is
modulated by the mouth and detected by a listener as something resembling
normal speech. However, when the tone introduced into the mouth for
modulation is in the ultrasonic frequency range, some means must be
provided to translate the ultrasonic signals into an audible frequency
range. To produce speech from the translated signals representing speech,
the signals must be synthesized into artificial speech. When synthesizing
speech, the more the translated signals resemble human produced signals,
the more likely the artificial speech will resemble natural speech. Thus,
if the tone introduced into the mouth is a waveform resembling the
waveforms produced by the vocal cords (glottal pulses rather than
oscillatory waves as taught by the prior art), then the speech produced
will more closely resemble human speech.
FIG. 2 graphically illustrates a glottal pulse shaped waveform which can be
used in accordance with the preferred embodiment of the present invention.
The glottal pulse of FIG. 2 depicts the relationship between the amplitude
of the glottal pulse over time. Although the glottal pulse waveform some
what resembles a sine (oscillatory) waveform, significant differences
exist between the two waveforms, namely the glottal pulse experiences a
longer rise time and fast fall time. If an ultrasonic glottal pulse signal
is analyzed by a frequency analysis system or spectrum, analyzer, a
frequency spectrum, resembling the signal graphed in FIG. 3, can be
produced. It should be noted that the amplitude of the signal of FIG. 3 is
highest in the below 30 kilohertz range, where the fundamental frequency
and the first couple of harmonics are located.
Two applications of a preferred embodiment of the present invention are
illustrated in FIGS. 4A and 4B. FIG. 4A depicts a woman wearing a head
set, having two speakers 30 and a receiver and/or transmitter 32 held
before the woman's mouth by an arm 33. In place of the transmitter 32, a
vibration device 34 may be placed on the woman's throat, in the
approximate area of the larynx 16. The vibration device 34 causes the
vocal tract to oscillate at an ultrasonic frequency, thereby causing
ultrasonic signals vibration within the vocal track, which may be
modulated by the mouth into words. Either way, ultrasonic signals are
radiated from the mouth of the woman, whereupon they illuminate the oral
cavity and are in return picked up by the ultrasonic microphone contained
within arm 33. An ultrasonic microphone is used to filter out sounds
within the audible range (below 13,500 Hz) so that only ultrasonic speech
is transmitted to the translator and synthesizer along the wire 36.
Speakers 30 may be used to provide a side-tone to the woman after
translation and synthesization so she can monitor the quality and
amplitude of her speech as spoken.
FIG. 4B depicts a similar application in which a telephone 38, having a
transmitter and receiver in the mouth piece 40, and a speaker in the ear
piece 42, transmits and receives signals over the cable 44 for either
transmission or translation and synthesization. Cable 44 may also be used
to recommunicate synthesized signals back up to the speakers 42, where the
synthesized speech may be monitored as a side-tone in the ear of the
speaker.
Since the amplitude modulated signals detected by the microphone are not in
the audible range when radiated from the mouth, the signals must be
translated into corresponding signals in the audible frequency range, if
they are to be used to produce speech. The same signals can also be used
to operate certain machinery, and in such applications, there is no need
to convert the signals into audible tones. Translation may be accomplished
through the use of a modified vocoder, or similar device, which is capable
of translating the ultrasonic speech into audible speech signals. In FIG.
5, a pulse generator 50 is used to produce a series of glottal pulses in
the ultrasonic frequency range for communication to vibrator or
transmitter 52. The ultrasonic glottal pulses are introduced into the
vocal track and then reradiated by the mouth 54 as ultrasonic speech. The
ultrasonic speech can then be picked up by receiver 56 and communicated to
the translator 57.
Translator 57 has four filters which split the ultrasonic speech signals
into four separate bands. The 12 Kilohertz low-pass filter 58 basically
covers the fundamental frequency of both the average male and female
speaker, while the 9 kilohertz to 30 kilohertz band-pass filter 60, the 30
kilohertz to 99 kilohertz band-pass filter 62, and the 90 kilohertz
high-pass filter 64 covering the remainder of the frequency spectrum. The
four bands separated by the filters 58-64 are then transferred to separate
frequency tracking circuits 66A-D and amplitude tracking circuits 68A-D,
which track the ultrasonic speech formats. Frequency and amplitude
tracking circuits 66 and 68 separate the frequency and amplitude
components of the ultrasonic speech signals, track the ultrasonic speech
formats, and output the signals to synthesizer 69.
The output of frequency tracking circuit 66D is input to a pulse divider
70, which divides the frequency component of the 12 kilohertz band by 30
and outputs a synchronization pulse for controlling the variable resonance
circuits 72 and the 400 Hz low-pass filter 78. Variable resonance circuits
72 produce converted speech signals with corresponding frequencies in the
audio spectrum of 300 hertz to 3500 hertz for transmission to the summing
circuits 74. The output of noise generator 80 is switched by switch 82
with the output of pulse divider 70 to create the non-vocal hissing noises
that are made by the teeth, lips and tongue, and which are required to
make certain sounds. The control of switch 82 is maintained by solenoid
84, which operates under control of input signals from amplitude tracking
circuit 68A, which indicates when speech signals are in the higher
frequency range and noise generation is not required. Provisions may also
be made to shift the audible voice tones into pitches that more closely
resemble that of the speaker, for instance, male or female. Summing
circuits 74 recombine all of the signals for transmission to speaker 76,
where the artificial speech is audibly reproduced.
FIG. 6 illustrates an alternative embodiment of the present invention in
which ultrasonic noise generator 98 and ultrasonic transducer 52 drive a
signal into mouth 54, where it is modulated into ultrasonic speech and is
picked up by receiver 56 and communicated to a converter 100. The
converter 100 is comprised of a spectrum analyzer 102 which segments the
analog wave form communicated from receiver 56 into a plurality of
discernable values and an analog-to-digital converter 104 which converts
the segmented analog waveform into a digital representation, or in other
words, a digitized spectral signature of the ultrasonic speech.
The digitized spectral signature of the ultrasonic speech is then
communicated to speech generator 106, which translates the digitized
spectral signature into audible speech. Speech generator 106 includes a
pattern recognizer 108 which utilizes speech recognition techniques for
generating a signal which can be synthesized by speech synthesizer 110
into electrical signals representing audible speech. The output of speech
synthesizer 110 is then communicated to speaker 76, which produces audible
artificial speech from the electrical signals.
It should also be noted that the primary function of either the preferred
or alternative embodiment need not be speech production. As is
demonstrated by FIG. 6, the input to speaker 76 may also be used to input
signals to a peripheral device 112, such as a conveyor system or robot.
Such a feature allows a worker in a noisy environment to operate a machine
from a distance through ultrasonic voice communication.
Although the present invention has been described above in terms of a
preferred embodiment, it is contemplated that numerous alterations and
modifications of the invention will be apparent to those skilled in the
art after having read the above disclosure. It is therefore intended that
the following claims be interpreted as covering all such modifications and
alterations as fall within the true spirit and scope of the invention.
* * * * *