Apparatus for Producing a Vocal Sound Signal in an Electronic Musical Instrument

This invention relates to an apparatus for producing a vocal sound signal in an electronic musical instrument.

It has been known from a study of speech sounds that sounds resembling the human voice can be produced by using a formant filter. However, there has not yet been produced a simple apparatus for producing a consonant and subsequently thereto a vowel.

An object of the invention is to provide a simple apparatus for producing a consonant and a vowel subsequent thereto.

According to a first feature of this invention, in an apparatus in which a formant filter comprising plural filters connected in parallel to one another is interposed in a passing circuit for a musical tone signal obtained from a musical tone signal generator by operation of a key, a voltage controlled type filter is connected before or after the formant filter, and a control electrode thereof is connected to an output terminal of a control signal generating circuit arranged to be operated by the operaton of the key, so that a peak of the voltage controlled type filter is shifted to one side by an attack signal of an output of the control signal generating circuit and is shifted to the other side by a decay signal subsequent to the attack signal.

According to a second feature of this invention, in an apparatus in which a formant filter comprising plural filters connected in parallel to one another is interposed in a passing circuit for a musical tone signal obtained from a musical tone signal generator by operation of a key, at least one of said plural filters of the formant filter is composed of a voltage controlled type filter and is so arranged that, by an output signal of a control signal generating circuit arranged to be operated by operation of the key, a peak thereof is shifted from a lower frequency to a higher frequency and is then set in a sustained condition.

According to a third feature of this invention, in an apparatus in which a formant filter comprising plural filters connected in parallel to one another is interposed in a passing circuit for a musical tone signal obtained from a musical tone signal generator by operation of a key, the formant filter comprises a first formant filter for a consonant and a second formant filter for a vowel and said first and second formant filters are interposed in the passing circuit through first and second gates, respectively, and further there are provided first and second control signal generating circuits arranged to be operated by the operation of the key, and respective output terminals of the first and second control signal generating circuits are connected to respective control electrodes of the first and second gates so that at the moment when the key is operated, the first gate is opened to produce a consonant signal, and sequentially to the consonant signal the second gate is opened to produce a vowel signal.

FIG. 1 is a block diagram showing one embodiment of this invention,

FIG. 2 (A) is an output characteristic diagram of a keying signal generator,

FIG. 2 (B) is an output characteristic diagram of a control signal generator,

FIG. 3 is characteristic diagram of a formant filter,

FIG. 4 is a characteristic diagram of a voltage controlled filter,

FIG. 5 is a block diagram showing another embodiment of this invention,

FIG. 6 is an output characteristic diagram of another control signal generator,

FIG. 7 is a characteristic diagram of filters 5d, 5e,

FIG. 8 is a characteristic diagram of filter 5f ,

FIG. 9 is a characteristic diagram of the filters 5d, 5e, 5f,

FIG. 10 is a block diagram showing another embodiment of this invention,

FIG. 11 (A) is an output characteristic diagram of a keying signal generator,

FIG. 11 (B) is an output characteristic diagram of a first control signal generating circuit,

FIG. 11 (C) is an output characteristic diagram of a second control signal generating circuit,

FIG. 12 is a characteristic diagram of a formant filter for a consonant and p FIG. 13 is a characteristic diagram of a formant filter for a vowel.

Referring to FIG. 1 showing one embodiment of the invention, numeral 1 denotes a musical tone signal generator comprising a main oscillator 1a and a frequency divider lb . A plurality of output terminals of devider 1b are connected through a gate circuit 2 to a common passing circuit 3 for a musical tone signal. A voltage controlled type filter 4 and a formant filter 5 comprising plural filters 5a, 5b, 5c connected in parallel one to another are interposed in series in the passing circuit 3. An output terminal of the passing circuit 3 is connected through an amplifier 6 to a speaker 7.

Numeral 8 denotes a keyboard circuit, and numeral 9 denotes a keying signal generator provided in the keyboard circuit 8. An output terminal of the keying signal generator 9 is connected to a control signal generator 10, and an output terminal of the generator 10 is connected to a control electrode of the voltage controlled type filter 4.

The musical tone signal generator 1 is arranged to generate a waveform includng a number of harmonic wave components such as a saw-toothed wave, a pulse wave or the like. The control signal generator 10 is so arranged that a control signal comprising an attack signal A, a decay signal D, a sustaining signal S and a release signal R as shown in FIG. 2 (B), which is a so-called envelope signal, may be generated by a keying signal as shown in FIG. 2 (A) generated from the keying signal generator 9.

If it is now intended to generate a vocal sound "PA", the respective filters 5a, 5b, 5c of the formant filters 5 are so set as to have peaks at 800 Hz, 1200 Hz and 2800 Hz, respectively, as shown in FIG. 3. Additionally, the voltage controlled type filter 4 is so set that, as shown in FIG. 4, the filter 4 has a peak near 800 Hz at the sustaining signal S of the control signal of the control signal generator 10, and the peak may be shifted to 1500 Hz at the attack signal A and be shifted back to 800 Hz at the decay signal D.

If, with the above arrangement, the keyboard circuit 8 is operated by operation of a keyboard, a corresponding gate of the gate circuit 2 is opened and a corresponding musical tone signal is applied to the passing circuit 3. Meanwhile, the keying signal (FIG. 2(A)) is generated by the keying signal generator 9, and the control signal (FIG. 2(B)) is generated by the control signal circuit 10, so that, as shown in FIG. 4, the peak of the voltage controlled type filter 4 is shifted from 800 Hz to 1500 Hz by the attack signal A of the control signal, and sequentially thereto is shifted from 1500 Hz by the decay signal D thereof and remains at 800 Hz by the sustaining signal D. Thus, in the formant filter 5 the peak at 800 Hz is shifted to 1500 Hz as shown by the dotted lines in FIG. 3 and is moved back to 800 Hz. During this operation, the sound of the consonant "P" is produced, and emitted from the speaker 7. When the peak settles at 800 Hz, the vowel sound "A" is produced and emitted. Consequently, the vocal sound "A" is produced.

For producing other vocal sounds, for instance, "PI", "PU", "PE", "PO", the respective filters 5a, 5b, 5c of the formant filter 5 are set as shown in the following Table 1.

Additionally, the voltage controlled type filter 4 is so set, for each case, that the peak thereof may be shifted from 100 Hz, which is the minimum cutoff, to each of those values as shown in the following Table 2 by the attack signal A and is shifted back to each of the sustaining frequencies by the decay signal D.

Referring to FIG. 1, numeral 11 denotes a low frequency oscillator for vibratomodulation which serves to make the produced sound a more realistic voice sound.

The voltage controlled type filter 4 is provided in front of the formant filter 5 in the illustrated embodiment, but the filter 4 may be connected after the filter 5.

FIG. 5 shows a second embodiment of this invention. Therein, control signal generator 10' connected to the keying signal generator 9 is so constructed that there may be generated therefrom a signal whose output voltage is gradually increased from a time point when a key is depressed and, after about 30 m sec, is set in its sustained condition as shown in FIG. 6. The circuit can be constructed, for instance, as an integration circuit.

Among filters 5d, 5e, 5f constituting the formant filter 5, filter 5f is constructed to be a voltage controlled type filter. The remaining parts corresponding to those in FIG. 1 are indicated by the same reference numerals.

It is now intended to produce a vocal sound "RA", the filters 5d, 5e are to have peaks at 800 Hz and 2800 Hz, respectively as shown in FIG. 7. Additionally, the voltage controlled type filter 5f is so set that it ordinarily has its peak at 800 Hz, but its peak is shifted from 800 Hz to 1200 Hz as shown in FIG. 8 by the output of the control signal generator 10' as shown in FIG. 6.

Thus, a sound "R" is generated when a key is depressed, and with lapse of time, the peak of the voltage controlled type filter 5fis shifted to 1200 Hz, whereby the sound is changed into the sound "A" having its peaks at 800 Hz, at 1200 Hz and at 2800 Hz as shown in FIG. 9, and as a result the vocal sound "RA" is produced, as a whole, and emitted from the speaker.

For producing the vocal sounds RI, RU, RE, RO, the filters 5d, 5e are so set as to have their peaks at respective frequencies as shown in the following Table 3, and additionally, the voltage controlled type filter 5f is set, for each case, so that its peak is given such a shift as shown in the middle column of Table 3, by the output of the control signal generator 10'.

FIG. 10 shows a third embodiment of this invention. Therein, a series circuit of a formant filter for a consonant 5-1 and a first gate 12 and a series circuit of a formant filter for a vowel 5-2 and a second gate 13 are connected in parallel to one another in the passing circuit 3, and an output terminal of the keying signal generator 9 is connected to first and second control signal generating circuits 14, 15, respectively. The output terinals of the circuits 14, 15 are connected to respective control electrodes of the first and second gates 12, 13. The first control signal generating circuit 14 comprises a differential circuit 14a and an integration circuit 14b, and the second control signal generating circuit 15 comprises an integration circuit.

If, by operation of a keyboard, a keying signal as shown in FIG. 11 (A) is generated from the keying signal generator 9, a waveform comprising an attack signal and a decay signal as shown in FIG. 11 (B) is obtained at the output terminal of the first control signal generating circuit 14, and a waveform as shown in FIG. 11 (C) is obtained at the output terminal of the second control signal generating circuit 15. The first and second gates 12, 13 each comprises a transistor 16, a luminous diode 17 connected in series thereto, and a photoconductive element 18 comprising Cds facing the diode 17. The formant filter 5-1 comprises two filters 5g, 5f and is so set, for instance, in the case of producing a "R" sound, as to have its peak at 1200 Hz and 2400 Hz as shown in FIG. 12. The formant filter 5-2 comprises three filters 5i, 5j, 5k connected in parallel one to another, and is so set as to have its peak at 800 Hz, 1200 Hz and 2800 Hz as shown in FIG. 13. The remaining parts of the circuit are the same as those in FIG. 1 and FIG. 5.

If, now, a keyboard is operated and thereby a corresponding gate in the gate circuit 2 is opened, a musical tone signal is applied to the respective formant filters 5-1 and 5-2. Additionally, a control signal (FIG. 11 (B)) generated from the first control signal generating circuit 14 by the keying signal (FIG. 11 (A)) generated from the keying signal generator 9 serves to open the first gate 12, and thereby a signal passed through the formant filter 5-1 having the characteristics shown in FIG. 12 is allowed to pass therethrough and as a result the sound "R" is produced and is emitted from the speaker 7. The control signal (FIG. 11 (C)) generated from the second control signal generating circuit 15 gradually increases and serves to gradually open the second gate, and thereby a signal passed through the formant filter 5-2 having the filter characteristics as shown in FIG. 3 is gradually allowed to pass therethrough. As a result a vowel sound "A" is produced sequentially to the sound R, and thus the vocal sound "RA" is obtained at the speaker 7.

For obtaining vocal sounds RI,RU,RE,RO, the formant filter 5-1 remains unchanged, while the respective filters 5i, 5j, 5k constituting the formant filter 5-2 are so set as to have their peaks at the frequencies shown in the following Table 4.

The musical tone oscillator 1 in any of the embodiments shown in FIGS. 1, 5 and 10 is not limited to one comprising the main oscillator 1a and the frequency divider lb, and substantially the same results can be obtained an arrangement in which the oscillator is a voltage controlled type oscillator, for instance, and the keyboard circuit 8 is formed as a voltage generating circuit for generating a voltage corresponding to a key, so that a so-called synthesizer is constructed. With this arrangement, a keying signal may be received at an output of the voltage generating circuit, or from a keying signal generating circuit provided separately.