JPS598838B2 - An electronic musical instrument that generates human voice sounds - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an electronic musical instrument that generates a human voice sound at a pitch corresponding to the pitch of a pressed key by operating a keyboard.

In recent years, attempts have been made to generate sounds resembling the human voice, that is, human voice sounds, in keyboard electronic musical instruments such as electronic organs. For example, there are electronic musical instruments that generate human voice sounds based on one specified phoneme to achieve a choral effect. However, the only phonemes that could be generated were the vowels ``a'', ``u'', and ``su'', and it was not possible to sing the lyrics. Therefore, for example, JP-A-53-303
As seen in Publication No. 13, the lyrics are encoded character by character and stored sequentially, and the stored lyrics information is sequentially read out from the beginning by operating the lyrics readout operator, and at the same time, the pitch specification operation panel An electronic singing device has been proposed that allows singing performance by specifying a pitch by uttering lyrics at the specified pitch.

However, in such a device, the lyrics reading operator and the pitch specifying operation panel must be operated synchronously, which not only makes the operation cumbersome, but also allows changes in the pitch (melody) and phonological changes of the lyrics. There is a problem in that unnecessary deviations in timing tend to occur, impairing playability.

This invention has been made in view of the above points, and it generates a human voice sound corresponding to the lyrics at a pitch corresponding to the pitch of the pressed key in synchronization with the key press operation on the keyboard section, and sings the lyrics. The purpose is to provide an electronic musical instrument that can be used as an electronic musical instrument.

Hereinafter, the content of the present invention will be explained with reference to embodiments shown in the accompanying drawings.

FIG. 1 is a block diagram showing the basic configuration of an embodiment of the present invention, in which numeral 1 is an operation section including a keyboard section, 2 is a phonological information memory section consisting of a read-only memory (ROM), etc. for reading only; Phonological information for each phoneme of vocal sounds is stored.

3 is a writable/readable random access memory (R
4 is a phoneme information memory control unit, 5 is a phoneme information recording control unit, and 6 is a multiplex signal generation unit for multiplexing a large number of phoneme information for each phoneme. .

Reference numeral 1 denotes a musical tone control signal generating section that generates a musical tone control signal based on a signal read out from the phonetic information memory section 2 or the phonetic information recording section 3, and 8 is a musical tone forming section, which generates tone name information KV and key press signals KON based on key presses. generates a sound source signal, the musical tone parameters are controlled by various control signals generated by the musical tone control signal generating section 7, and a human voice sound signal of a pitch corresponding to the pitch of the pressed key is formed; Sound is generated through a sound system 9 consisting of an amplifier, speakers, etc.

The appearance of the operation section 1 is shown in FIG. 2, and the keyboard section 10
, a phoneme designation switch group 11 consisting of a phoneme designation switch for each phoneme (50 sounds, each voiced sound, each voiced sound, etc.), a manual switch SMN, an auto switch SAT, a light switch SWT, an up switch Su, and a down switch S.

, reset switch SR, etc. are provided. In addition, 1
Reference numeral 2 denotes a display section indicating which phoneme is being processed. Next, an outline of the operation of this embodiment will be explained.

First, press the light switch SWT to set the light mode, and press each phoneme designation switch in the phoneme designation switch group 11 in sequence along the lyrics. The phoneme information of the designated phoneme stored in is sequentially read out. In synchronization with this, the phoneme information recording control section 5 operates, and the phoneme information read out from the phoneme information memory section 2 is transferred to the phoneme information recording section 3.
are recorded sequentially. After recording all the lyrics, press the auto switch SAT to set the auto mode and play the keyboard section 10. Each time you press a key, the phoneme information recording control section 5 operates and the phoneme information recorded first is recorded from the phoneme information recording section 3. The information is read out in the order in which it was recorded and transmitted to the musical tone control signal generator 7.

Thereby, the musical tone control signal generating section 7 generates a large number of control signals for generating the human voice of that phoneme, and based on these control signals, corresponds to the pitch of the pressed key generated by the musical tone forming section 8. Various parameters of the sound source signal are controlled to form musical tones, and human voice sounds are sequentially generated via the sound system 9. In this way, by sequentially operating the keyboard, lyrics corresponding to the pitch of the operated keys can be sung.

When the manual switch SMN is pressed to enter the manual mode, by playing the keyboard section 10 with one hand and specifying a phoneme in the phoneme designation switch group 11 with the other hand, the phoneme information for that phoneme can be retrieved from the phoneme information memory section 2. is read out and transmitted to the musical tone control signal generator 7. Thereby, it is possible to generate the human voice of the specified phoneme at the pitch of the key pressed at that time. The configuration and operation of this embodiment have been briefly described above, and will now be described more specifically and in detail with reference to FIGS. 3 and 4.

FIG. 3 shows a musical tone control signal generator (musical tone forming section 8 in FIG. 1).
1 (excluding the sound system 9), and parts corresponding to those in FIG. 1 are given the same reference numerals.

In the operation unit 1, the phoneme designation switch group 11 is set to A, B,
When each phoneme designation switch such as U, ......Yu, ......, etc. is turned on, "1" is added to that line.
”, and this signal is designated as SW.

Further, when any one of the phoneme designating switch group 11 is turned on, the signal outputted via the 0R circuit 21 is assumed to be AS. Also, light switch SWT, auto switch SAT
, manual switch SMN, up switch SUL, down switch SD, and reset switch SR are respectively on, the signals that become "1" are sent to WT, AT, MN, UP, respectively.
Let DW and RS be. Note that KON is a key press signal that is always 601 and becomes "11" only while any key is pressed. The phoneme information memory section 2 is composed of a ROM, and
2 is assigned 100 addresses for each phoneme, for example, addresses 0 to 99 are assigned "a", 100 to 19
"I" for address 9, "U" for addresses 200-299, etc.
・・・・・・・・・、2000-2099 has "Yu"
, . . . , the phoneme information of all the phonemes that can be specified by the phoneme designation switch group 11 is stored. The contents of this phoneme information will be described later. The phoneme information memory control unit 4 includes a phoneme designation switch group 1
1, and an encoder 22 that inputs the signal SW from 1 and outputs an address signal of the address corresponding to the specified phoneme (in the previous example, the upper address of the 3rd digit or higher), an adder 23, and an 0R circuit 24. Become.

The 0R circuit 24 outputs 8 when the signal WT or MN becomes "1".
1゛ is output and added to the lead terminal RO of ROM2.
Make M2 readable.

The adder 23 specifies an address in the ROM 2 using an address signal obtained by adding the address signal of the upper address from the encoder 22 described above and the lower two digits of the address signal from the multiplexed signal generator 6, which will be described later. The phoneme information recording unit 3 consists of a RAM, and this RAM 3 requires a capacity that can be used to write the phoneme information of ordinary song lyrics by allocating 100 addresses to each phoneme, but this capacity is not necessary. You can choose as appropriate.

The phoneme information recording control section 5 includes an up-down counter 25, an adder 26, AND circuits 27, 28,
It consists of 0R circuits 29 and 30 and differentiating circuits 31 and 32.

The AND circuit 27 ANDs the signals WT and AS that output 31'' in the write mode, that is, when the light switch SWT is pressed, and when the phoneme designation switch is pressed, and the AND circuit 28 outputs the signals WT and AS in the auto mode, that is, when the phoneme designation switch is pressed. When the switch SAT is pressed and when the key is pressed, the signals AT and KON, which output 11'' respectively, are ANDed. 0R circuit 29 is AND circuit 2
7 and 28, and the signals UP and DW which output "1" when the up switch Su is pressed and the down switch SD are pressed, respectively, are output through a differentiating circuit 31. Then, the rising differential pulse of the output signal is inputted to the counter input terminal CK of the up-down counter 25.

Normally, this input pulse causes the up/down counter 2 to
5 is counted up, but the down switch SD is pressed and the down terminal DN is set to 611 by the signal DW.
When it gets old, it will count down.

The 0R circuit 30 outputs signals WT, AT, and RSf that output 111 when the light switch SWTl auto switch SAT and reset switch SR are pressed.
)0R, the rising differential pulse of the output signal is inputted to the up-down counter 25 set terminal R via the differentiating circuit 32, and this up-down counter 2
Reset 5.

The output of the up-down counter 25 becomes an upper address signal corresponding to the order of phoneme information to be written into or read out from the RAM 3.

The adder 26 uses an address signal obtained by adding the address information from the up-down counter 25 and the lower two digit address signals from the multiplexed signal generator 6, which will be described later, to input R.
Specifies the AM3 write or read address.

The multiplexed signal generator 6 includes a counter 33 to which a high-speed clock pulse Cp is always input to the count input terminal, and a flip-flop circuit (hereinafter abbreviated as "FF") whose output terminal Q is connected to the reset terminal R of the counter 33. 34
AND circuits 35 to 38, 0R circuit 39, and differentiation circuit 40
Consisting of

The AND circuit 35 outputs "11" to the set terminal S of the FF 34 when each bit of the output of the counter 33 becomes "1" (maximum count value).
is set and its output terminal Q becomes "1", so the counter 33 is reset and the set state is maintained.The AND circuit 36 ANDs the aforementioned signals WT and AS, and the AND circuit 37 Signal AT, 5KON and AN
The AND circuit 38 ANDs the signal MN which outputs "1" in the manual mode, that is, when the manual switch SMN is pressed, and the signal AS described above.
The 0R circuit 39 indicates that any one of the AND circuits 36 to 38 is "1".
", it outputs "1", and the rising differential pulse is input to the reset terminal R of the FF 34 via the differentiating circuit 40.

As a result, FF34 is reset, and its output terminal Q
Since the value becomes "0", the counter 33 is de-set and restarts its counting upon the arrival of the clock pulse Cp. The output of this counter 33 is sent to the adder 23 as a signal for a lower address signal for time-division multiplexing and reading or writing a large number of phoneme information for one phoneme.
, 26. Musical tone control signal generation section 7 is ROM2
Or a D/A converter 41 that converts the phonetic information read from the ROM 3 into an analog signal, and a demultiplex and sample hold circuit that demultiplexes the converted analog signal and converts it into each control voltage signal etc. described later and holds it. Consists of 42. Note that if the musical tone forming section 8 is constructed of a digital type, the D/A converter 41 becomes unnecessary. Furthermore, musical tone forming section 8
If this is also performed in a time-division manner, the demultiplexing and sample hold circuit 42 will also become unnecessary. However, in that case, it is necessary to synchronize with the tone forming section 8. Next, the operation of the musical tone control signal generator configured as described above will be explained for each mode.

(1) Light mode This mode is used to perform the auto mode described later.
RA the phonological information stored in M2 along with the lyrics in advance.
This is a mode for writing to M3.

When the write switch SWT is turned on, the signal WT becomes ``11'', and the lead terminal RD of ROM2 is connected to 6 through the 0R circuit 24.
11 to enable reading, and set the write terminal W of RAM 3 to "1" to enable writing.

Further, when the signal WT reaches 61'', a rising differential pulse is output from the differentiating circuit 32, and the up-down counter 2
Reset 5. Next, when one of the switches in the phoneme designation switch group 11 is turned on to designate a phoneme, the encoder 22 outputs an address signal for designating that phoneme.

At the same time, the signal AS becomes 611, so the AND circuit 27 outputs 311, and the up-down counter 25 is incremented to "1" via the 0R circuit 29 and the differentiation circuit 31.
This is displayed on the display section 12. Also, the AND circuit 36
outputs "17" and resets the FF 34 via the 0R circuit 39 and the differentiating circuit 40. Therefore, the counter 33 is unset and starts counting the clock pulses Cp. The adder 23 outputs a count value as a lower address signal for specifying the address of a plurality of phoneme information. Yotsute ROM2
Specify the address within and read the phoneme information of the specified phoneme. In synchronization with this, the adder 26 specifies an address in the RAM 3 using the upper address signal from the up-down counter 25 and the lower address signal from the counter 33, and transfers the phoneme information read from the ROM 2 to the RAM 3. The address for the first phoneme (for example, addresses 1 to 100) is written. This transmission of phoneme information for one phoneme from the ROM2 to the RAM3 is performed until the counter 33 reaches a full count.

When the counter 33 reaches a full count, the output of the AND circuit 35 becomes "1", so the FF 34 is set, and its output terminal Q becomes 611, so the counter 33 enters the reset state and waits for the next phoneme specification.In this way, By sequentially specifying phonemes using the phoneme designation switch group 11, the phoneme information of lyrics (hereinafter referred to as "lyrics information") can be written into the RAM 3 in the order of the phonemes to be pronounced.

In addition, when correcting the written part, turn on the down switch SD, the signal DW becomes 81n, and the down terminal DN of the up/down counter 25 becomes "11n".
At the same time, a differential pulse is input to the count input terminal CK, and the count is counted down by one. Therefore, when a phoneme is specified next using the phoneme designation switch group 11, the up-down counter 25 increments by one and outputs the same address signal as before counting down, and the phoneme for one phoneme written at the end of the RAM 3 is output. The information is erased and new phonological information is written in its place. Similarly, down switch S
After pressing D the same number of times as the number of phonemes to be corrected, all corrections can be made by specifying a new phoneme using the phoneme designation switch group 11. Also, in parts where there is only accompaniment and no lyrics, you can skip writing the phonological information, but in such cases, you can press the up switch Su as many times as necessary.
The up-down counter 25 is counted up by the rising differential pulse of the signal UP, and only addressing is performed without writing phoneme information into the RAM 3.

). ) Auto mode In this mode, R is synchronized with the KON signal generated when a key is pressed.
This is a mode for reading lyrics information from AM3 and generating musical tones (human voice sounds) corresponding to the pitches of pressed keys.

When the auto switch SAT is turned on, the signal AT becomes "11", and the lead terminal RD of the RAM 3 becomes "11", making it ready for reading, and the up-down counter 25 is reset by the rising differential pulse.

A KON signal is generated in response to the key press operation, and the AND circuit 2
The output of 8 becomes "17," and the 0R circuit 29 and the differential circuit 3
The up-down counter 25 is incremented by 1 to designate a phoneme address.

Further, each time a KON signal is generated, the FF 34 of the multiplexed signal generator 6 is reset, the counter 33 starts counting, and outputs address signals for a plurality of control signals of one phoneme to the adder 26. These address signals cause R
AM3 is addressed once, and one phoneme's worth of phoneme information is read out in the written order for each key press, that is, lyrics information is read out. This read information is converted into an analog signal via the D/A converter 41 of the musical tone control signal generating section 7, demultiplexed in the demultiplexing and sample hold circuit 42, and then sent to the musical tone forming section 8 (
(details of which will be described later). The timing of this demultiplexing is determined by the multiplexing signal generator 6
This is done in synchronization with the address designation of the RAM 3 by a signal from the counter 33. When the reset switch SR is turned on, the signal SR becomes "11", and the up-down counter 25 is reset via the 0R circuit 30 and the differential circuit 1 32, and the lyrics information stored in the RAM 3 is used to reset the up-down counter 25 from the beginning. However, you can sing at any pitch (melody). (3) Manual mode
In this mode, by simultaneously operating the phoneme designation switch group 11 and the keyboard section 10 (see Figure 2), a musical tone (human voice sound) with a pitch corresponding to the pitch of the pressed key is specified.
This is the mode for generating.

When the manual switch SMN is turned on, the signal MN becomes "1", and the lead terminal RD of ROM2 is connected via the 0R circuit 24.
is set to "11" to enable reading.

By specifying a phoneme using the phoneme designation switch group 11, the phoneme address (upper address) of the ROM 2 is specified in the same manner as in the write mode, and the multiplex signal generator 6 outputs multiple phoneme information for one phoneme. address (lower address) is specified, and the phoneme information is read from the ROM2.

This read information is not written to the RAM 3, but is directly D/A converted, demultiplexed, and sample-held in the musical tone control signal generating section 7, and is transmitted to the musical tone forming section 8 as a plurality of voltage signals for musical tone control. .

Next, a specific example of the musical tone forming section 8 shown in FIG. 1 is shown in FIG. 4, and will be explained along with its function. The names of the respective blocks in FIG. .

VCO is a voltage controlled variable frequency oscillator, VCA is a voltage controlled variable gain amplifier, VCF is a voltage controlled filter, B
PF is a voltage controlled variable bandpass filter, EG is a control waveform generator, WS is a waveform converter, LFO is a low frequency oscillator,
OS is a one-shot multi, PS is a phase shifter, BBD is a charge transfer element, and DL is a delay circuit, and these abbreviations will be used in the following description.

In addition, VIL, VAL. VSL,VAT,VD
T and VRT are initial I signals for generating control waveforms as shown in Fig. 5A and C in response to a key press signal (hereinafter abbreviated as "KON"), respectively, as shown in Fig. 5A.
These are voltage signals that determine the attack level AL, the sustain level SL1, the attack time AT1, the decay time DT1, and the release time RT.

Note that the sustain time ST ends at the falling edge (key release) of the KON signal, so no separate signal is required. Also, VCA
Since the control waveform for controlling the voltage rises from the zero level and returns to the zero level as shown at the beginning of FIG. 5, the voltage signal VIL for determining the initial level IL is unnecessary. Since the control waveform for controlling VCF or BPF has a sustain level SL of zero level as shown in FIG. 5C, a voltage signal VSL for determining it is not necessary. In addition to these voltage signals, there are VL that controls the amplification degree of the VCA, VCF that controls the center frequency and selectivity of the BPF, Vf and VQ that control the selectivity, Vvsp that controls the vibrato speed, VVDP that controls the vibrato depth, and frequency deviation. ΔF to control the shift value, V to select the conversion waveform
A voltage signal such as VT is used to control the output pulse width of ct, Vβ, V, bi...,0S.

Although the above signals are shown with the same symbols for each block in FIG. 4, they actually have different values, and these voltage signals correspond to the musical tones in FIGS. 1 and 3. These are control signals generated by the control signal generating section 7, and digital information for generating these control signals for each phoneme is phoneme information.

Now, based on the above premise, the operation of the musical tone forming section 8 shown in FIG. 4 will be explained in order.

Pitch voltage K as KON and pitch name information by key press operation
occurs.

Using this pitch voltage K, the VCO5O for generating the sound source signal,
The oscillation frequency of 5l is controlled to output a signal with a frequency corresponding to the pitch of the pressed key. However, in order to obtain a chorus effect, the oscillation frequency of the VCO 5l is slightly shifted from the frequency corresponding to the pitch of the pressed key by Vljf. Also, LFO
A low frequency signal generated by 52 is applied to VCO 5O, 5l via VCA 53, imparting a vibrato effect through frequency modulation and simulating vocal fold vibration.
Vibrato speed is controlled by Vvsp, and pibrato depth is controlled by VVDP. The output signals of the VCOs 5O and 5l are converted to Va, Vβ, V, . . . by the WSs 54 and 55, respectively.
For example, the WS54 outputs a sawtooth wave, and the WS55 outputs a rectangular wave, both of which are output as sound source signals containing many harmonic components. Ru.

These two series of sound source signals and a noise signal from a VCA 78, which will be described later, are mixed via VCAs 56 to 58. The mix rate is adjusted by the magnitude of VL applied to each VCA 56-58, but the ratio of noise signals caused by the VCA 58 is made small. The reason for this is that, as will be explained later, this series is mainly a circuit that generates vowel signals, but the VCA5
This is because the noise signal input to 8 is mainly a signal for creating consonants, and is not so necessary for vowels. The mixed sound source signal is divided into three series and BBD59 to 61.
The delay time is modulated by .

62-64 are VCOs that generate transfer clock signals for each BBD 59-61.

VCO 62 is approximately 0.7H generated by oscillator 65
z generated by the very low frequency signal and the oscillator 66
It is modulated by a signal obtained by superimposing a low frequency signal of Hz via mixing resistors Rl and R. The VCO 63 also outputs signals generated by the oscillators 65 and 66 to the PS 67 and
68, and are modulated by signals superimposed via mixing resistors R, , R4. Furthermore, the VCO 64 outputs the phase-shifted signals by PSs 67, 68 and 2/2 by PSs 69, 7O, respectively.
The signal is phase-shifted by 3π and is modulated by a signal superimposed via mixing resistors R5 and R6. Therefore, B.B.D.
The modulation position by 59, 6O, 6l is 2/3π (1206
), the chorus effect becomes even more pronounced when these three series signals are mixed.

This mixed musical tone signal is envelope-formed by the VCA 7l.

EG72 is controlled by VAL, VAT, VS and VDT9RT and generates a control waveform to control VCA7l in response to KON, but KON is slightly delayed by DL73. This is because the signal whose envelope is formed by the VCA 7l is mainly a signal that becomes a vowel of a human voice, and therefore needs to be generated a little later than a signal that becomes a consonant, which will be described later. Next, a circuit for generating signals mainly for consonants will be explained.

74 is a white noise generator whose output signal is V
The frequency is selected by CF75.

The center frequency and selectivity of the VCF 75 are controlled by Vf and VQ, and transient characteristic changes are given by the control waveform from the EG 76. 0S77 pulses the KON signal while the pulse width is controlled by VT and adds it to EG76.
EG76 is L, VAL, VAT, VDT with its pulse width.
, VRT generates a control waveform.

Therefore, the change in the transient characteristics of the VCF 75 is extremely short. The noise signal whose frequency is selected by the VCF 75 has an envelope formed by the VCA 78 .

EG79 receives the pulsed signal of KON by 0S80 as well as VAL, VAT, VSL, VDT, VR as described above.
A control waveform is generated by T to control the VCA 78. Therefore, the noise signal output from the VCA 78 is a very short signal mainly for consonants of human voice sounds.

A signal mainly for vowels whose envelope was formed by the above-mentioned VCA 7l and a signal mainly for consonants whose envelope was formed by the above-mentioned VCA 78 are connected to CA8l and 8.
2 to become vowel signal A, and C
Signal B for consonants mixed by A83 and 84
becomes.

The mix rate in this case is also adjusted by the magnitude of VL added to each VCA 8l to 84, but VCA 82 and 8
The ratio by 3 is extremely small. Signal A is divided into three series,
Frequencies are selected by BPFs 85, 86, and 87, respectively,
The signals are mixed again via VCAs 88, 89, and 90. The center frequencies Fl, f2, f3 and selectivities Ql, Q2, Q3 of each BPF 85 to 87 are controlled by f and VQ applied to each BPF, and the output level ratio of each series is
It is adjusted by VL which controls A88-90. Therefore, as shown in FIG. 6, the center frequencies f1 to F3, selectivities Q1 to Q3, and levels L1 to L of each BPF 85 to 87 are
3 is determined, a fixed formant for each vowel is formed. Furthermore, in this embodiment, VIL, VAL, VAT, VD are set according to KON by EG9l to 93, respectively.
A control waveform determined by T and VRT is generated, and the VCA
Changes in the transient characteristics of BPFs 85 to 87 are applied via filters 94 to 96, making it possible to more faithfully simulate human voice sounds. D
L97 to 99 are KON according to the delay time due to DL73
It is inserted to delay the signal. Signal B is BP
The frequency is selected by FlOO and mixed with the vowel signal via VCAlOl.

The center frequency and selectivity of BPFlOO are determined by the Vf, V, applied to it, and also by the pulsed signal of KON by 0S102 and the VIL9VAL/VA signal.
T! A control waveform generated at EGlO3 by the VDT bow VRT controls the transient characteristics of the BPF via VCAlO4 to form a consonant signal of a specified phoneme. In this way, the signal C mixed by the VCAs 88 to 90 and 101 is a combination of vowels and consonants to become a human voice signal of the specified phoneme, and is transmitted to the sound system 9 (see FIG. 1) via the batua 105. is sent to and pronounced.

Therefore, according to the phoneme information (lyrics information) read out from the ROM 2 or RAM 3 in FIG.
You can make the lyrics sing.

In this embodiment, a VCO system is used as the sound source generator, but an open/close system or a digital variable frequency tone generator system may also be used. In that case, the pitch name information K will be multiple bits.

As explained above, the electronic musical instrument according to the present invention stores lyrics in advance, and generates human voice sounds with phonemes corresponding to the lyrics at a pitch corresponding to the pitch of the pressed key in synchronization with key press operations. This makes it easy to operate, and the lyrics and melody do not deviate, allowing the same lyrics to be sung at various pitches (melody). Therefore, it can be used for chorus practice and when composing and arranging vocal music, making it extremely effective.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the basic configuration of an embodiment of the present invention, Fig. 2 is an external view of its operating section, Fig. 3 is a block circuit diagram showing a specific example of a musical tone control signal generating device, and Fig. 4 is a block diagram showing the basic configuration of an embodiment of the present invention. A block circuit diagram showing a specific example of a musical tone forming section, FIG. 5A is an explanatory diagram of a key depression signal, and FIG. 6 is an explanatory diagram of control waveforms, and FIG. 1...Operation section, 2...Phonological information memory section (ROM) 3...Phonological information recording section (RAM)
, 4... Phonological information memory control unit, 5...
・Phonological information recording control unit, 6...Multiplex signal generation unit, 7...Musical tone control signal generation unit, Total...
Musical sound formation section, 9...Sound system, 10.
. . . Keyboard section, 11 . . . Phoneme designation switch, 12 . . . Display section.

Claims (1)

[Claims] 1. A keyboard section, a group of phoneme specification switches for specifying each phoneme of human voice sounds, a phoneme information memory section that stores phoneme information for each phoneme of human voice sounds, a phoneme information recording section, and the phoneme information recording section. A musical sound that forms a human voice with a pitch corresponding to the pitch name information and a phoneme corresponding to the phoneme information by the phoneme information read from the keyboard section and the note name information of the pressed key generated from the keyboard section. a forming section; means for sequentially reading out phoneme information of the phoneme specified by the phoneme designation switch group from the phoneme information memory section and recording it in the phoneme information recording section; means for sequentially reading out information for each key press operation in synchronization with the key press operation of the keyboard section, and transmitting the information to the musical tone forming section together with the note name information of the press key; An electronic musical instrument that produces human voice sounds at high frequencies.