US 3881059 A
A system for speech training is described which stores and displays speech parameters (pitch, intensity, etc.) on a television screen as a trace or line. Two parameters from the same speech signal (for example, pitch and intensity) are stored and displayed simultaneously, one on the lower half of the screen and one on the upper half of the screen. The same parameters from the voice of a speech therapist can be displayed along with those of a trainee. The therapist's and trainee's parameters are separated using black traces for the trainee, and white traces for the therapist on a grey background. Both the therapist's traces and the trainee's traces may be independently moved about the screen so that they can be overlaid for direct comparison, thus assisting trainees, especially the deaf, to improve their speech. The digital system produces a visible trace when the sum of the line count number plus the signal value equals a predetermined value. Trace-broadening indicias are formed by a plurality of sum values.
Description (OCR text may contain errors)
United States Patent 1 91 1111 81,059
Stewart 1 1 Apr. 29, I975 SYSTEM FOR VISUAL DISPLAY OF SIGNAL Primary Examiner-William C. Cooper PARAMETERS SUCH AS THE Assistant E.raminerE. S. Matt Kemeny AR M T S 0F SPEECH SIGNALS FOR Attorney, Agent. or F irm-Martin Lu Kacher SPEECH TRAINING PURPOSES  Inventor: Leslie C. Stewart, Rochester, NY.  ABSTRACT A system for speech training is described which stores and displays speech parameters (pitch, intensity, etc.) on a television screen as a trace or line. Two paramel l Filedi ug. 16. I973 ters from the same speech signal (for example, pitch [21 1 App No; 388,771 and intensity) are stored and displayed simulta neously, one on the lower half of the screen and one on the upper half of the screen. The same parameters  Assignee: Center for Communications Research, Inc., Rochester, NY.
 US. Cl 179/1 SP f th oi of a s eech therapist can be displayed [5|] Int. Cl. G101 1/14 along ith tho e of a trainee, Tho therapist's and Fleld Search 179/l 43/5; 235/ trainees parameters are separated using black traces 340/324 AD for the trainee, and white traces for the therapist on a grey background. Both the therapists traces and the trainees traces may be independently moved about  References Cited the screen so that they can be overlaid for direct com- UNITED STATES PATENTS parison, thus assisting trainees, especially the deaf. to 2.416.353 2/1947 Shipman ct al. 179/1 improve their speech. The digital system produces a 2.416.353 2/1947 Shipman et a1... 179/1 SP visible trace when the sum of the line count number 2920.31 /196 o on t 43/5 plus the signal value equals a predetermined value.
Trace-broadening indicias are formed by a plurality of .9.. 9 9 un er sum values. 3.791483 2/1974 Bushnell 340/324 AD 17 Claims, 8 Drawing Figures pun wq- /I8 [I6 /Z4 TVILINES I0 INPUT neeiaeturm DISPLAY M -53 umr uznomes CONTROL '1' w BLK ion .--EoR-w met-1 1 '1 l I 'r-s IIIC sw Y I F /22 12!- I wane sw Tmne l I 1-1. F'Osv SW comm i l 1N1 vPossw PARAMETER: 5 T
SELECT sw l DISPLAY mmmmzsaszs SHE] 10F 6 mum L SYSTEM FOR VISUAL DISPLAY OF SIGNAL PARAMETERS SUCH AS THE PARAMETERS OF SPEECH SIGNALS FOR SPEECH TRAINING PURPOSES The present invention relates to visual display systerns and particularly to a system for providing a graphic display of signal parameters which remain displayed unless or until such time as it is desired to change the display.
The invention is especially suitable for displaying one or more speech parameters for speech training purposes. Features of the invention are generally applicable for providing displays of various signal parameters in the form of a graph which represents the parameter as a function of time (viz., a graphic display in rectangular coordinates) as well as for other visual signal display purposes.
Speech training and therapy depends in large measure upon the skill of the therapist and the cooperation of the trainee. The therapist must interpret the deficiencies in the various parameters of the trainees speech, teach the trainee how to correct his speech to accommodate for the deficient parameters and determine if the trainee has learned the necessary corrections by listening to his attempts. The speech training process is therefore necessarily iterative and time consuming. A visual speech training aid as provided in accordance with the invention can be of significant assistance to the therapist in analyzing the performance of the trainee and determining the speech parameters in which deficiencies exist and in communicating performance targets and directions to the students. lt provides for storage and display of speech parameters and characteristics such as pitch, intensity, nasality, voice, and voiceless speech, etc.
The system includes memories having storage for a portion of a speech parameter such as pitch or intensity and provides a graphic display of the parameter as a trace or line on a screen, such as a television monitor screen. In a typical operation the trainee presses a display button while he speaks. When the button is re leased, the last few seconds of the speech are translated into parameters of speech, say pitch and intensity, and are permanently retained in memory. The signals in memory are displayed on the screen in a frozen" or permanently displayed manner. A plurality of parameters from the same speech signal (for example, pitch and intensity) can be displayed simultaneously, say, one on the lower half of the screen and one on the upper half of the screen. The same parameters from the voice of the therapist can be displayed along with those of the trainee and may be distinquished by the trainee display being made up of black traces while the therapist display is in the form of white traces, or vice versa. The system provides facilities for independently moving the displays about the screen so that they can be overlaid for direct comparison.
Accordingly. it is an object of the present invention to provide an improved system for speech training.
It is another object of the present invention to provide an improved system which aids in the training of the deaf and other hearing impaired per ms to speak.
It is still another object of the present invention to provide an improved system for displaying speech which is especially useful as a visual speech training aid.
It is a further object of the present invention to provide an improved system for the storage and display of speech parameters, for example pitch and intensity.
It is a still further object of the present invention to provide an improved system for speech training and therapy which is easy to use by both the trainee and the therapist.
It is a still further object of the present invention to provide an improved system for speech training and therapy, whereby a plurality of parameters of the same speech signal can be displayed simultaneously.
it is a still further object of the present invention to provide an improved speech training system whereby a plurality of different speech parameters can be displayed simultaneously and independently moved on the display for comparison purposes (viz., display of speech parameters produced by a trainee and therapist can be overlaid on the screen for direct comparison).
It is a still further object of the present invention to provide a quantitative display of speech parameters.
It is a still further object of the present invention to provide an improved system for storing parameters, such as pitch and intensity, of segments of speech such as phonemes, words, and phrases, and displaying them for as long as desired until replaced by the display of parameters of other speech segments.
It is a still further object of the present invention to provide an improved system for visually aiding speech training which enables a trainee to visually observe when he is talking too loudly or too softly, with too high or too low a pitch, too much through his nose or too much through his mouth, and how well he makes speech sounds.
It is a still further object of the present invention to provide an improved system for visually aiding speech training whereby traces of speech parameters derived from the therapist's voice can be continuously displayed on a screen while traces of like speech parameters derived from the trainees voice are successively displayed as the trainee practices emulation of the therapists voice, the display thus providing a quantitative measure of the degree of learning achieved by the trainee.
Aspects of the invention are also generally applicable for providing graphic displays of functions, such as analog signals which vary with time.
It is thus a further object of the invention to provide an improved system for providing graphic displays of analog signals on parameters of such signals as in rectangular coordinates.
lt is a still further object of the invention to provide an improved system for graphic display of signals in rectangular coordinates wherein the display can be controlled so as to modify the trace representing the signal in selected portions of the trace.
it is a still further object of the present invention to provide an improved system of generating visual displays wherein the display is controllable to change the trace which depicts the signal as by causing a portion of the trace to disappear, become broadened or to be marked.
It is a still further object of the present invention to provide an improved display system wherein a trace of a signal to be displayed is obtained by digitally controlling the intensity of a writiing or scanning beam.
Briefly described, a system for providing a graphic display in accordance with the invention includes a recirculating memory having storage for signals, for ex ample, digital characters, representing successive val ues of the signal to be displayed. Display means such as a television monitor are provided which display signals as visible traces on the plurality of lines which are disposed one alongside the other in succession. The display is generated by reading out the memories concur rently with each of the lines. Display control means are responsive to the value to which each signal read out of the memory corresponds and the position of the line with respect to the other lines on the display. When the signal value and the line position number combine (sum) to provide a value equal to the predetermined value or values (as for trace broadening to provide indicia in selected portions of the trace), the trace is controlled to become visible as by being darkened or lightened (to provide a black or white trace). The sequence of visible traces constitutes a line or trace which represents the signal. Since the signals in memory recirculate, the display remains on the screen in a fro zen" or continuous manner until it is replaced with a new signal portion. Means are provided for incrementing or decrementing the combined value of signal and line position so as to change the position of the line or trace constituting the display vertically or horizontally, or to mark the display in a manner to indicate certain signal characteristics or features. Trace-broadening indicias are formed by a plurality of sum values.
The foregoing and other objects and advantages of the present invention as well as additional features thereof will become apparent from a reading of the fol lowing description in connection with the accompanying drawings in which:
FIG. I is a block diagram illustrating a display system embodying the invention;
FIG. 2 is a block diagram showing the input unit of the display system illustrated in FIG. 1;
FIG. 3 is a block diagram showing the timing unit of the display system shown in FIG. 1;
FIG. 4 is a block diagram illustrating the control unit of the system shown in FIG. 1;
FIG. 5 is a block diagram illustrating the recirculating memories of the system shown in FIG. 1;
FIG. 6 is a block diagram illustrating the display control unit of the system shown in FIG. 1',
FIG. 7 is a waveform diagram illustrating the timing relationship of pulses produced in the timing unit shown in FIG. 3; and
FIG. 8 is a waveform diagram similar to the diagram shown in FIG. 7, but on an expanded scale.
FIG. 1 illustrates a visual speech training system wherein the invention is embodied. Two microphones are provided. The T-MIC I0 is provided for the therapist. Parameters of the therapist's speech result in white traces (W) on the display 12. The display 12 used in this embodiment of the invention is a conventional television monitor having intensity control or Z axis, vertical sync V and horizontal sync H inputs. The trainee or student is provided with a microphone l4 and the parameters of speech, pitch and intensity in the case of this embodiment of the invention, which the student or trainee speaks into his microphone ll (S-MIC) are displayed as black traces on the screen of the display 12. The pitch parameters maybe displayed in the upper half of the screen while the intensity parameters are displayed in the lower half of the screen. The student's traces are black (BLK) while the therapist's traces are white (W). Their location on the screen identifies the parameter and the color of the trace represents whether it is the students or the therapist's speech which is generating the parameter.
Considering the display 12, it will be observed that the display is in rectangular coordinates with the ordinate or Y axis being the television line at the horizontal sync lines in the raster scanned on the screen of the monitor and the abscissa or X axis calibrated in terms of memory position. TV lines 16 through 240 contain the display, while memory position 1 to 240 are pro vided for each TV line. Of course the display is not calibrated in terms of TV lines and memory positions as indicated. The calibration is provided herein solely for purposes of explaining the operation of the system. The memory positions in the display correspond to the stages in the re-circulating memories 16 of the system. A re-circulating memory of the same length (240 stages in shift registers constituting the memory) is provided for each parameter produced by the student and the therapist, there being four such re-circulating memories, one for the intensity parameter of the therapist; one for the intensity parameter of the student, one for the pitch parameter of the therapist; and one for the pitch parameter of the student. The memories also con tain two control memories which, like the re circulating memories for the parameter values, store control signals representing characteristics of the therapists and students speech and control signals. There are two control memories, one for the therapist and one for the student.
The input unit 18 translates the speech signals produced by the microphones l0 and 14 into digital signals representing successive values of parameters and of the speech characteristics and inputs them into the parameter and control memories in the re-circulating memories 16.
The control unit 20 is operated by various switches. The T-S MIC switch is operative to control the selection of either the therapist microphone 10 or the student microphone 14, such that either the therapist microphone output or the student microphone output will be connected in the input unit 18 for parameter and control characteristics processing. There are two write switches, one for the therapist and the other for the student. When the therapists write switch is depressed the parameters and control characteristics of the therapists speech which are at the time spoken into the microphone 10 are stored in the memory 16 and used to update the display 12. only the therapists white traces will be updated, and the students traces, the black traces, remain frozen on the screen. When the student's write switch is depressed, signals from the student microphone 14 will be translated into parameters and control signals, which will be stored in the memory 16 and used to update the black traces constituting the student display. The horizontal position of either the student or the therapist traces can be controlled (viz., the traces can be moved horizontally from right to left across the screen by operating the horizontal position switches). Similarly, the vertical position switches will control the location of the student or therapist traces causing them to move up and down on the screen. The parameter selection switch is provided when it is desired to input another parameter, say nasality, as obtained from an accelerometer mounted on the student's nose, into the input unit 18.
The timing unit 22 generates all of the timing signals such that data is inputted once each 128 horizontal lines into the memories for purposes of updating the memories. The timing unit also controls the shifting or re-circulating of the data through and around the memory registers. The timing unit 22 further provides line count signals to the display control unit 24. The line count corresponds to the number of the horizontal line which is being scanned on the display monitor 12. The timing is such that the memories are read out in their entirety during each horizontal scan (viz., the memory re-circulates in less time than it takes the beam to scan a horizontal line). it will be noted that horizontal and vertical sync is provided by the timing unit to the horizontal and vertical sync inputs of the display so as to insure proper synchronism and coherency of read-out from the memories with the scanning of the horizontal lines and retrace of the entire raster (which occurs upon occurrence of the vertical sync pulse). The value of the digital signals read out of the memory and the line count value taken together determine whether a spot or portion of a trace should be displayed or written on a particular horizontal line. The sum of the line count and digital value is determinative. For example, the first line which is used in the display is horizontal line No. 16. If the parameter represents a large value it should be written at the top of the trace or on horizontal line 16. A digital number corresponding to the sum of the value of the digital signal and the line count is indicative of whether or not the signal is of high enough intensity to be written on the top line 16. Accordingly, when the sum of the line count and digital values equals a predetermined number or predetermined numbers, the display control 24 will be operated to cause the intensity of the beam to be increased in the case of the therapist (light) traces, or decreased in the case of the student (black) traces. As each horizontal line is scanned the intensity of the writing beam will be increased or decreased depending upon whether or not the value of the digital signals are sufficiently high to equal the predetermined or writing threshold. Inasmuch as the memory is read out in synchronism with the horizontal tracing of each line, successive portions of the line will have a spot written thereon. The synchronism of the memory readout with the line scanning thus provides for the generation of the traces constituting the display. In the illustrated display where 224 TV lines are used, and there are 240 stages in the memories, there is a possibility of 53,760 points which can be written. In this particular embodiment of the invention, it will be observed as the description proceeds that each parameter has a memory of 240 words or stages of seven bits each. Accordingly, only 6,720 bits in total of memory are required to provide a display with high resolution, thus affording a considerable memory savings.
It will be observed that there are two displays, one on the upper half and one on the lower half of the screen. These two displays are obtained by controlling four memories through the timing pulses provided by the timing unit 22. Two memories, one for the therapist and one for the student, provide the pitch parameter display during TV lines to 126. Two other memories, one for the therapist and one for the student provide the intensity parameter display during the TV lines 128 to 239.
The translation of the speech signals into digital signals may be accomplished by the input unit which is illustrated in detail in FIG. 2.
Referring to FIG. 2, the inputs T-MlC and S-MlC from the therapist and student microphones are selected by a switch 26 which may be relay operated from the control unit 20. The inputs from the switch are applied to a pitch analyzer 28 and intensity analyzer 30, a silence characteristic detector 32, and an unvoiced speech characteristic detector 33. The pitch analyzer 28 is a circuit of conventional design which follows the fundamental frequency or pitch of the voice. In essence it is a frequency discriminator which provides a slowly varying analog DC voltage-reference may be had to Remaking Speech," by Homer Dudley, J. Acoust. Soc. Am., Vol-ll, No. 2 pl69-l77 Oct. i939, for a suitable pitch analyzer.
The intensity analyzer 30 is a broadband amplifier which follows the peak or average amplitude of the voice signal from the microphone and may include automatic volume control circuits, sometimes called compression circuits, if desired.
The silence detector 32 is an amplifier 36, the output of which is applied to a Schmitt trigger or other suitable threshold circuit 38. A period of silence, i.e., when the input signal to the amplifier 36 falls below a threshold which may be controlled by varying the reference voltage to the Schmitt trigger input, allows the Schmitt trig ger output voltage to drop to a low output voltage state. An inverter 40 changes the logic level such that a period of silence is represented by a high logic level.
The unvoice detector 33 includes a detector circuit 42 which may contain a high pass and a low pass filter. The ration of the high pass to low pass filter outputs is obtained, as in a potentiometer and produces an output voltage which operates the Schmitt trigger circuit 44 when the high pass filter output is greater than the low pass filter output. Since a voice signal containing larger energy in the high frequency components thereof represents an unvoice condition, the Schmitt trigger output voltage will be at a high level when the speech signal has an unvoice characteristic. As the discussion proceeds, it will be observed that a high voltage level is taken to represent a logical 1" and a low voltage level a logical Another parameter may be provided by an acceler ometer 46 which may be placed on the students nose region. The output of the accelerometer is amplified in an amplifier 48 and may be applied to a switch 50. The switch 50 is operable by the parameter selection button of the control unit 20 when it is desired to display a speech parameter known as nasality.
A pair of analog digital converters A/D-l, and AID-2, 52 and 54, translate the analog signal which is applied to the input thereof into a seven-bit digital word. The converter 52 provides a digital word indicated as DW-l consisting of the bits D-l-O through D-l-6. The other converter 54 provides a digital word DW-2 consisting of seven bits D-2-0 to D-2-6. Conversion of succes sively occurring signals representing the pitch and intensity parameters is provided by operating the sampling circuits of the converters upon occurrence of a convert pulse, CP, produced by the timing unit 22. Upon each C? pulse a sample of the analog voltage presented from the analyzers 28 and 32 to the input of the converters 52 and 54 is converted into the digital word DW-l and DW-2. For purposes of simplification of the description of the system. only the pitch and intensity parameters is mentioned herein. It will be appreciated, of course, that other parameters such as the nasality parameter may be displayed.
The timing unit is illustrated in FIG. 3. Reference may also be had to FIG. 7 and 8 which illustrate the waveforms of the timing pulses. A 15.36 Kl-lz freerunning clock oscillator 56 controls a us, one shot multivibrator 58 which produces the horizontal sync pulse waveform H-SYNC. The pulses from the clock oscillator 56 are divided in frequency by an eighbstage binary counter 60. The counter thus divides by 256 which corresponds to the number of TV lines (horizontal lines) in the raster on the display screen 12(FlG. l). Decoders 62 which may be provided by AND gates connected to different stages of the counter 60 derives a pulse of the duration during which counts 247 to 255 are produced (i.e., during the occurrence of TV lines Nos. 247 to 255). This pulse is used as the vertical sync pulse V-SYNC, and may be applied to the vertical sync pulse input of the display 12 (FIG. 1). An OR gate 64 combines the horizontal and vertical sync pulse to produce the composite sync waveform C-SYNC, which is also shown in FIG. 7. The decoders 62 provide pulses at counts and 128 which are combined in an OR gate 66 to produce the start vertical clock pulse signal also illustrated in FIG. 7. The I28 pulse counted by the counter produces a square wave having a frequency of 120 Hz. The lowest counter stage produces a 60 cycle square wave pulse train. The 120 Hz counter output drives a 5 [.LS one-shot 68 to produce the sample pulse. The sample pulse controls the timing of writing of successive digital words in the memories, as will be explained more fully hereinafter. The trailing edge of the sampling pulse drives another 5 p.s one-shot 70 which produces the convert pulse CP. The convert pulse when produced allows the analog to digital converter 52 and 54 to sample a successively occurring speech parameter signal from the pitch and intensity analyzers 28 and 30.
Another free-running clock oscillator CLK-OSC-2, 72, produces clock pulses at a 4.2 MHz rate. The rate is selected so that greater than 240 horizontal clock pulses, HC, will be produced during each horizontal sync pulse interval. It is these 240 pulses which will enable all of the stages of the memories to be read out during the scan of each horizontal line. The trailing edge of the horizontal sync pulse sets a flip-flop 74 which enables the clock oscillator 72 and starts the horizontal clock. The horizontal clock pulses are counted by an eight-stage binary counter 76. An AND gate decoder 78 connected to different stages of the counter 76 decodes the 240th horizontal clock pulse and resets the flip-flop 74 thus stopping the clock oscillator 72. As shown in FIG. 8 in the waveform for the memory clock, MC, the reset occurs before the onset of the next horizontal sync pulse.
The start vertical clock pulse sets a flip-flop 80 which enables an AND gate 82 through which horizontal sync pulses are applied to the input of a seven-stage or divideby- I 28 binary counter 84. The leading edge of the horizontal sync pulse causes the counter 84 to change state. When the last stage of the counter returns to 0," the negative going edge drives a l as one-shot 86 through an inverter 88. This presets the counter 84 to a count of 16. The one-shot 86 output pulse also resets the flip-flop 80. The counter 84 will therefore initially be preset to a count of 16. The output of the counter 84 is the vertical word. It is a seven-bit binary number, the decimal equivalent of which corresponds to the TV line which is being scanned on the screen of the display l2. The first visible line corresponds to TV line 15 (i.e., the vertical word is made up of seven binary 0 bits). Line 16 will be a binary word having a binary I bit in the 2 position. line 17 will correspond to a vertical word with a binary I in the 2 and in the 2 positions, and all the rest of the bits binary Os. The vertical word will have all ls at line 128.
As mentioned previously, the screen is divided into two halves, the dividing point being at the l28th line. Each parameter will be displayed on 112 visible TV lines, thus it is suffeicient that the vertical word corresponds to a decimal maximum of I28. Selection of the upper and lower halves of the display is obtained by means of the top/bottom, or TB, signal, which is the 60 Hz waveform from the counter 60. During one-half of that waveform, the upper I28 TV lines are used, and during the other half the bottom I28 TV lines are used. The vertical words are translated into line-count words LC-l and LC-2. Adders 90 and 92 are used for vertical position control purposes for incrementing or decrementing the LC-l and LC-2 words, respectively when the LC-I word is incremented or decremented, the display in the upper half of the screen is moved vertically, while when the LC-2 word is incremented or decremented, display in the lower half of the screen is moved.
The line count words are incremented by vertical positioning words VPW-I and VFW-2, which are respectively added to the vertical words in adder 90 and adder 92, to produce the line count 1 and line count 2 words. An up/down counter 94 in the control unit 20 (see also FIG. 4) produces the VPW-l word while another up/- down binary counter 96 produces the VPW-2 word. In the control unit there are provided a pair of switches for moving the display in the upper half of the screen up or down. These are shown as the VPOS-UP-I switch and the VPOS-DN-l switch. Similar switches VPOS- UP-2 and VPOS-DN-Z control the counter 96 for mov ing the display in the lower half of the screen up or down. These switches control the counting direction either up or down in the counters 94 and 96. The counters receive the 60 Hz TB wavform. The stages of the counter 94 provide the VPW-l word and the eight stages of the counter 96 provide the VFW-2 word. The highest order or 2 bit of these words is used as a sign bit.
When the up switches are closed for example, the counter outputs increase at the 60 Hz rate. When the up switch is opened the counters hold their current count. The adders 90 and 92 add these vertical positioning words to the vertical word from the counter 84. The lower order 7 sum bits become the line count words LC-l and LC-Z. Of course when the counter 94 and 96 outputs are all 0's, the vertical words become the LG! and LC-2 words. For positive displacements the adders 90 and 92 perform a normal binary addition. When negative displacement occurs, which will be indicated by the sign bit in the VPW-l and VFW-2 words, the sign bit is used to convert the operation in the adder to two's complement addition, so as to subtract the vertical positioning words from the vertical words. There is therefore, facility for positive and negative displacements corresponding to l l l and I28 TV lines respectively. This is enough to move the trace on the screen completely off the screen in either direction. If the displacement is greater than the aforementioned 111 or I28 lines, the 2 sum bit from the adder 90 and the 2 sum bit from the adder 92 becomes a binary l This bit is inverted and used to provide the OB-l and 08-2 bits for offset blanking purposes. The OB-l bit will prevent traces in the upper half of the display below the I28 TV line, or above the 16 TV line, while the OB-Z bit will prevent tracing below the 240th TV line or above the I28 TV line.
Referring again to FIG. 4, there are provided controls for horizontal positioning and writing. These controls are operated by the horizontal position switch HPOSSW-l and WSW-l which control horizontal positioning and writing or updating of the display in the upper half of the screen. Two similar swithes HPOSSW-Z and WSW-2 control horizontal positioning and writing in the lower half of the screen. When the horizontal positioning switch HPOSSW-l is depressed a flip-flop 98 is set. This enables an AND gate 100 to pass a shift pulse through an OR gate 102 and another OR gate 104 to the memory clock MC-1 output of the control unit. The memory clock output provides pulses for shifting or advancing the memories. Since a shift or advance of data aroung the memory (viz., from one stage to the next) causes a horizontal displacement of the trace in the display 12, the traces will be horizontally shifted when the horizontal position switch is operated. Horizontal positioning by virtue of the generation of the shift pulse occurs only during the sampling pulse (SP) interval, (see the waveform for the memory clock in FIG. 8). The shift pulse is generated when the sampling pulse sp passes through an AND gate [06. The sampling pulse also passes through an OR gate 108. The leading edge of sampling pulse triggers a 1 ps, one-shot 110. After the l as interval of the pulse from the one-shot 110, another 1 ,us one-shot 112 is triggered. The pulse from the one-shot 112 constitutes the shift pulse which passes through the AND gate 100, OR gate 103, OR gate 104 to shift data in the memory which controls the upper half of the display. The next sample pulse resets the flip-flop 98.
The data in the memory may, if desired, be moved only one memory cell at a time. To obtain such operation, the horizontal position switch HPOSSW-l, for instance, must be closed for less than one-half second. The flip-flop 98 is then set, as was the case for horizontal positioning described above. The output of the flipflop 98 is transmitted via an OR gate 114 to set a flipflop 116. When the flip-flop 116 is set, a one-half second one-shot H8 is triggered. During the one-half second duration of the one-shot 118 pulse, the AND gate 106 is inhibited via an inverter 120 from passing the sample pulse, SP. Instead, another AND gate 122 is enabled by the flip-flop 116 in its set condition. The AND gate 122 then passes the SP pulse which triggers the one-shot 110 and the one-shot 112. The shift pulse is then generated and resets the flip-flop I16 so that no more shift pulses occur if the horizontal posiion switch HPOSSW-l has been released. If this switch is closed more than a half second, then the single shift pulse is generated. After one half second the one-shot 118 pulse is no longer produced and the inverter 120 enables the AND gate 106 which allows the train of sample puls t pass through the AND gate 106 for generating a series of shift pulses each during a successive sample pulse period.
When writing is desired, the write switch. say WSW- l, is depressed. A flip-flop 124 is then set. The output of the flip-flop enables an AND gate 126 via an OR gate 128. The AND gate 126 passes the sample pulses, SP, to provide the enable write, ER-l, pulses for the memories which control the display in the upper half of the screen. in other words, when the ER-l waveforem is high, the output of the analog to digital converters 52 and 54 (FIG. 2) are connected to the memory inputs. The control memories are also enabled to receive the silence and unvoice bits from the input unit.
The flip-flop 124 when set by operation of WSW-l also enables an AND gate 130. The AND gate 130 passes the shift pulse to the MC-] output. Since the shift pulse occurs during the horizontal retrace inter vals and are coincident with the ER pulses (see FIG. 8), the data is shifted into the memories during the horizontal retrace intervals. Since the sample pulse SP occurs every 128 horizontal sync pulses, a new data word or control word is shifted into the memories every 128 horizontal sync pulses. Between entries of new data or control words the memories recirculate 127 times.
When the write switch is released the next sample pulse following release resets the flip-flop 124. When the flip-flop is reset the negative going edge of its output pulse triggers a 20 ms one-shot 132 and provides the end of writing pulse EWl to denote the end of writing on the upper half of the screen. The pulse from the one shot 132 also enables the gates 126 and 130 which allows additional ER-l and shift pulses to be applied to the memories. Thus, writing continues during the 20 ms EW-l pulse. The analog-to-digital converters 52 and 54 (FIG. 2) are reset on the leading edge of the CP pulse which follows the sample pulse. Thus a new conversion is started on the trailing edge of the CP pulse which is complete before the next sample pulse.
The horizontal positioning and writing on the lower half of the screen is controlled by the HPOSSW-Z and the WSW-2 switches. These switches operate a write and horizontal position control logic 134 identical to the logic associated with HPOSSW-l and WSW-l switches. The logic 134 produces the ER-2, MC-2 and EW-2 pulses which control the memories associated with the lower half of the screen of the display 12 (FIG. 1).
In the event that shift pulses are not applied to the memory clock MCI and MC-2 outputs, the horizontal clock l-lC are applied to these outputs via the OR gate 104 and an OR gate 136.
The recirculating memories are shown in FIG. 5. There are two memories and 142 for data word DW-l. Two memories 144 and 146 for DW-Z and two control memories 148 and 150. The enable recirculating pulse ER-l must be present for writing to occur in the data memories 140 and 144, and in the control memory 148. Since the ER-l pulse will be produced only when the write switch WS-l used by the therapist is operated. only data and control words due to the therapist's speech will be written into memories 140, 144 and 148. Similarly, only operation of the write switch WS-2 used by the student or trainee will produce an ER-2 pulse. Data words representing parameters due to the student's speech will be written in data memories 142 and 146 and in control memory 150.
The data memory 140 is typical of all of the data memories. It contains memory logic 152 for the first data bit D-l-O of DW-l; memory logic 154 which is similar to the logic 152 for the second data bit D-l-l of DW-l, and additional memory logics for each of the remaining bits D-l-2 to D-l -6 of DW-l. Only the logic 152 for the first bit D-l-l is shown to simplify the illustration. The control memories have memory logic for two bits indicated as C-l-l and Cl2 for the control memory 148, and C2l and C-22 for the control memory 150. Since the memory logic 152 is typical, it alone will be described in detail.
A shift register 158 which may be a dynamic MOS register, has a length or capacity for 240 bits of data. In the absence of a writing or horizontal positioning op eration. the ER-] pulse is low and an AND gate 160 is enabled, via an inverter 162. The HC pulses are applied on the MC-l line to the shift input of the register 158. Each of the 240 MC pulses due to the horizontal clock cause the bits in the register 158 to shift one position for each HC pulse and thus the entire register is read out during each horizontal line time. In other words, during the time a TV line is scanned on the screen of the display, each bit in the register 158 shifts succes sively to all positions and is recirculated via the AND gate 160 and an OR gate 162. The bits may be read out at the output of the OR gate as they recirculate. The bits read out of this first memory logic 152 are indicated as being the RD-l-O bits of the RDW-1A word. It is the RD-W-lA word which controls the display to write the therapists trace representing the pitch parameter in the upper half of the display. When a new data word is to be entered in memory the ER'I output goes high. thus enabling a gate 164. The gate 164 passes the Dl0 bit of the DW] word which is entered into the shift register by the shift pulse which is produced and appears on the MC-l input line to the register 158.
Memories 144 and control memory 148 share the MC-l clock and produce the RDW-2A data word and the control word consisting of the RC-l-l and RC- -1-2 bits. The data memories 142 and 146 and the control memory 150 share the M02 clock and produce the data word RDW-1B and RDW-ZB. The control memory 150 produces the control word consisting of the bits RC-Z-l and RC-2-2.
The control memories receive the control word EW. SlL and UNV bits. The EW-l bit is applied to the first control memory 148 and the EW-Z bit to the other control memory 150. A control word generator 166 and another control word generator 168 respectively associated with the control memory 148 and the control memory 150 translate the three control characteristic signals into the two-bit control words which are stored in the control memories. The logic of the control word generator consists of inverters and gates as shown in the case of the generator 166 which provide the following logical operations. When EW is high Cl is a l and C2 is a 1. When SlL is high Cl is a l and C2 is a 0. When UNV is high Cl is a O and C2 is a l. When EW, SlL and UNV are all low C1 is O and C2 is also 0.
The control words read out of the memories 148 and 150 are reconverted into control bits by recorded control word regenerating logic 170 and 172. The logic 170 is typical. It produces and end of writing EOR-A level. a SlL-A level. and a UNV-A level, depending upon the control characteristics of the therapist's speech. The work regenerator logic 170 consists of in verters and AND gates which provide the inverse or decoding operation of the operation provided by the control word generators 166 and 168. The other control word regenerator 172 which is associated with the memory 150 which stores the student's speech characteristics produces the EOR-B. SlL-B, and UNV-B levels which depend upon the student voice characteristics.
The display control unit 24, shown in detail in FIG. 6 converts the binary data stored in the recirculating memories into a TV raster format in which the parameter stored in memory becomes a trace on the screen. Consider the formation of the trace corresponding to the data stored in memory 140, the words RDW-lA each consisting of seven bits are read out 240 times during each horizontal line scan period. Each data word is added to the LC-l line count word in a binary adder 174. This binary adder 174 and OR gate 176 and an AND gate 178 provides control of the format of the trace representing the pitch parameter in the upper half of the screen. Accordingly, it is called the trace l-A-UP control. Only the 2 2 2, 2 and 2 sum bits from the binary adder 174 are used. The OR gate 176 passes the 2 sum bit when UNV-A is low. When UNV-A is high, the input to the AND gate 178 from the OR gate 176 is always enabled. The offset blanking OB-l level and the SlL-A level are also applied to the AND gate 178. When SlL-A and OB-l are high, the AND gate 178 operates as a decoder of binary values corresponding to decimal 124, 125, 126 or 127 (the five most significant sum bits from the adder 174).
Consider, for example. that LG] is a binary number corresponding to decimal 16. which is the line at the top of the screen (see FIG. 1). As memory recirculates, if any of the RDW-lA words has a decimal value 108, 109, 110, or 111, then these RDW-lA words when added to 16 enable the AND gate 178 for the duration of the data word. The AND gate 178 output is operative to produce a white dot on the screen at the point of time that the data word of sufficient value (i.e., 108 or greater) is read out of the memory 140. In this way a trace consisting of dots depending upon the value of the RDW-1A words stored in the memory 140 will be written and displayed on the screen. At mid screen LC-1 is equal to decimal 127. Then a RDW-lA word having a value of 0 will be decoded by the gate 178. Note that if gate 178 were connected to all of the seven sum bits available from the adder 174, then only a single sum value of 127 would be operative to produce the writing of a point on the screen. By using only the five most significant bits. a broader trace covering four TV lines is written on the screen. lndicia are the wider trace portions.
1n the event that the UNV-A signal is high. the 2 input to the AND gate 178 will always be high. Thus. the OR gate 176 is operative to broaden the trace to eight TV lines when the UNV-A waveform is high. The display control also utilizes similar logic including a binary adder. an OR gate and an AND gate connected as shown for the trace lA-UP control for the control of trace 2Adown. trace 1B up. and trace ZB-down associated with the data memories 142, 144, 146, respectively. The output of the trace controls is translated into a video control signal, by video control logic 180. The video control logic also utilizes the EOR-A and EOR-B and the TB wavefonn. and determines whether the top or bottom of the screen is to be used. The video control encodes the signals on a three-level grey scale, white for the therapist traces, black for student traces, and grey for no trace. It combines the video signal with the C-sync composite sync signal and applies that the signal to the intensity or Z input of the video monitor which produces the display on a screen.
Four OR gates 182, 184, 186 and 188 receive the trace control and EOR signals. The upper and lower half of the screen are selected by the AND gates 190, I92, 194 and 196. When the TB waveform is high, the top half of the screen is selected by the gates 190 and 194. High level pulses at the output of an OR gate 198 become white spots on the screen, while high level pulses at the output of another OR gate 200 become black spots on the screen. An AND gate 202 in conjunction with an inverter 204 acts to inhibit black pulses from appearing in the presence of white pulses. The white spots get precedence. The supply voltage V,- is slightly more positive than the high logic level voltage. The resistor 206 and the potentiometers 208 and 210 establish the voltages for the white, grey, and black levels of the composite video signal. When white is produced, the output of the OR gate 198 and an inverter 212 are high; there is then very little voltage drop across the resistor 206 to provide the white signal which is amplified by the video amplifier 214. For grey level, the OR gate 198 output is low, while the output of the inverter 212 is high; current is then drawn through the resistors 206 and 208 causing a voltage drop at the input to the video amplifier 214. The black level is established when both the gate 198 and the inverter 212 outputs are low; then the potentiometers 208 and 210 are both allowed to draw current through the resistor 206, thus dropping the video signal amplitude to black level. The adjustment provided by the potentiometers 208 and 210 allow for the adjustment of the grey and black levels.
The composite C-sync signal is added to the video signal after inversion in an inverter 216 and after passing through an isolation diode 218. Accordingly, a succession of white and black dots is written on the screen of the video monitor to provide the traces representing the signal parameter.
From the foregoing description it will be apparent that there has been provided an improved visual display system which is especially suitable for use as a speech training system. Variations and modifications in the herein described system, within the scope of the invention, will undoubtedly suggest themselves to those skilled in the art. Accordingly, the foregoing description should be taken merely as illustrative and not in any limiting sense.
What is claimed is:
1. A system for providing a graphic display of a parameter which comprises:
a. means having storage for signals corresponding to successive values of said parameter,
b. display means for displaying signals as visible traces on a plurality of lines, one alongside the other in succession, each of said lines having a value corresponding to its position with respect to the other of said lines,
c. means for reading out of said storage means all of said signals concurrently with each of said lines, and
d. display control means responsive to the sum of the value to which each said signal read out of said storage means corresponds and the value corresponding to the position of each line with respect to the other of said lines on said display means for operating said display means to provide a visible trace during readout of each said signal when said sum equals a certain value whereby to provide a sequence of visible traces on said display means which provides said graphic display which represents said parameter.
2. The invention as set forth in claim 1 wherein said display means comprises means providing a TV display wherein each of said lines is a succesive horizontal line of said display, and means for controlling the intensity of said lines to provide said traces.
3. The invention as set forth in claim 2 wherein said storage means comprises a recirculating, plural stage digital memory, and said readout means comprises means for reading out digital signals presented in a stage in said memory, each during a successive portion of each of said horizontal lines.
4. The invention as set forth in claim 3 wherein said display control means comprises means for generating digital controls signals concurrently with the generation of each of said horizontal lines, each of which digital control signals represents the relative position of the horizontal line for which it is concurrently generated, and means for providing a signal to said intensity control means for rendering said line visible when the sum of the values represented by said digital control signal and said digital signal exceeds a certain value.
5. The invention as set forth in claim 4 including means for changing the location of said digital signals in said stages of said recirculating digital memory for horizontally moving the position of said graphic display of said parameter.
6. The invention as set forth in claim 4 including means for changing the values of said digital control signals for vertically moving the position of said graphic display of said parameter.
7. The invention as set forth in claim 4 which comprises a control recirculating plural stage digital memory for storing control digital signals in successive stages thereof, means for successively reading out digital signals from said control memory in synchronism with the digital signals from said first named memory, and means for controlling said means for providing said signal to said intensity control means in accordance with the value of said read-out control digital signals.
8. The invention as set forth in claim 4 wherein a plurality of said recirculating memories are provided each having storage for a different parameter, and wherein means are included in said intensity control means responsive to digital signals from difierent ones of said memories for providing the graphic displays which represents different parameters vertically offset from each other.
9. The invention as set forth in claim 4 wherein a plurality of said recirculating memories are provided, and wherein means are included in said intensity control means separately responsive to digital signals from dif ferent ones of said memories for providing said visible traces with different intensity whereby to simultaneously provide a plurality of said graphic displays each consisting of traces having a certain distinct intensity.
10. The invention as set forth in claim 4 further comprising an analog to digital converter for providing said digital signals for storage in said memory, and means operative once during each repetitive display of said parameter for reading a digital signal from said converter into said memory.
11. A speech training system which comprises a. means for providing signals representing a parameter of voiced speech,
b. recirculating memory means for storing said signals which are voiced during a period of time,
c. display means for providing a trace representing the signals stored in said memory,
d. timing means for circulating said signals in said memory in timed relationship with the read out of signals into said display means, and
e. control means operated by said timing means for controlling said memory means and said signals which are read out of said memory means for varying the location of said trace.
12. The invention as set forth in claim It further comprising a plurality of said memories, a plurality of said means for providing said signals at least one from the voice of a trainee and another from the voice of a therapist, means for storing said trainee signals and said therapist signals in different ones of said memories, and means for operating said display means for providing a plurality of said traces of each of the signals from a different one of said memories.
13. The invention as set forth in claim 12, including display control means for moving said plurality of traces with respect to each other so that they can be overlaid for comparison purposes.
14. The invention as set forth in claim 12 wherein said signals providing means includes means for providing a plurality of signals from the voice of said trainee and a plurality of signals from the voice of said therapist each representing a plurality of speech parameters,
means for storing each of said signals in a separate one of said plurality of memories, and means included in said display operating means for providing a plurality of separate traces each corresponding to the signals stored in a different one of said memories.
15. The invention as set forth in claim 14 further comprising means responsive to each of said speech signals for providing signals representing the certain characteristics of said speech, a recirculation control memory for storing said signals, and means included in said display control means responsive to said signals stored in said control memory for changing the size of said traces representing said characteristics.
16. The invention as set forth in claim 11 wherein said display means comprises a screen means for scanning a raster of successive, vertically displaced horizontal lines across said screen and means for controlling the intensity of said lines, said recirculating memory comprises register means for storing digital signals corresponding to successive values of said parameter, means for providing digital signals corresponding to the vertical position of each of said horizontal lines, means for reading out said memory during the scanning of each of said horizontal lines and means for operating said intensity control means to provide a trace on said screen when the sum of said vertical position signal and said parameter value signal exceeds a predetermined value.
17. The invention as set forth in claim 16 including control means for changing said trace including means operative to generate digital control signals, and means for combining said digital control signals and said sum operative to change said sum.