JP3405170B2 - Music synthesizer - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a musical tone synthesizing apparatus for generating musical tone signals such as musical tones and voice sounds of natural musical instruments by synthesizing formants.

[0002]

2. Description of the Related Art Conventionally, as this kind of musical tone synthesizer, there is one disclosed in, for example, Japanese Patent Laid-Open No. 4-251297. This tone synthesizer reads formant parameters that change in time series from a memory, generates a plurality of formants based on the read parameters, and synthesizes the generated formants to generate a tone signal based on the formants. To do. In this way, since the musical tones are synthesized based on the formants that change in time series by sequentially reading the parameters and generating the formants, it is possible to reproduce the timbre peculiar to the musical tones and voice sounds of natural musical instruments.

[0003]

However, the shape of the formant generated based on the parameters is, for example, a symmetrical shape centered on the center frequency.
The formant shape cannot be finely controlled.
Further, it is impossible to process only a specific (for example, one) formant shape to change the timbre. further,
Formants include voiced sound formants and unvoiced sound formants, but these formants cannot be independently processed in various ways. For this reason, when synthesizing a new musical tone, it is difficult to enrich the variation of the timbre to be synthesized by the formants and to improve the processing accuracy of the timbre.

SUMMARY OF THE INVENTION An object of the present invention is to provide a tone synthesizer for synthesizing tone signals by synthesizing formants so as to enrich the variations of tone colors to be synthesized and to improve the accuracy of tone color processing.

[0005]

According to a first aspect of the present invention, there is provided a musical tone synthesizing apparatus which comprises a formant generating means for generating musical tone signals having a plurality of formant components on a plurality of channels, and a musical tone signal generated by the formant generating means. and external output means for outputting selectively the external tone signal of any channel among the output from the external output unit, an external
A tone signal input means for inputting a processed tone signal; a tone signal synthesizing means for synthesizing the tone signal of the formant component generated by the formant generating means and the tone signal input by the tone signal input means; It is characterized by having.

According to the musical sound synthesizer of the first aspect of the present invention, the external output means selects the musical sound signal of the formant of any channel generated by the formant generating means and outputs it to the outside. By doing so, it is possible to externally process even a single formant musical tone signal, the processed musical tone signal is inputted by the musical tone signal input means, and the inputted musical tone signal is processed by the musical tone signal synthesizing means. By synthesizing with the musical tone signal generated by the formant generating means, it is possible to enrich the variation of the timbre. Further, since the tone signals of one formant can be individually processed, the processing accuracy of the tone color is improved.

A musical sound synthesizer according to claim 2 of the present invention is
A voiced sound formant generating means for generating a musical sound signal of a voiced sound formant, an unvoiced sound formant generating means for generating a sound signal of an unvoiced sound formant, a musical sound signal input means for inputting a musical sound signal from the outside, and the voiced sound formant generating means. The generated tone signal, the tone signal generated by the unvoiced sound formant generating unit, and the tone signal input unit that selectively inputs the tone signal input unit, and a tone signal synthesizing unit. A tone signal, an external output means for selectively outputting the tone signal generated by the voiced sound formant generation means and the tone signal generated by the unvoiced sound formant generation means to the outside , the tone signal input means, External output
The tone signal output from the unit and processed externally is input.
It is characterized by being a force .

According to the musical tone synthesizing apparatus of the present invention, the musical tone signal synthesizing means generates the musical tone signal generated by the voiced sound formant generating means and the musical tone signal generated by the unvoiced sound formant generating means. And the musical tone signal synthesized by the musical tone signal synthesizing means is output to the outside by the external output means, whereby a musical tone signal is obtained by formant synthesis of the voiced sound formant and the unvoiced sound formant. Further, the external output means selects the musical tone signal generated by the voiced sound formant generation means or the musical tone signal generated by the unvoiced sound formant generation means and outputs the selected musical sound signal to the outside, so that the musical tone signal of the voiced sound formant and the unvoiced sound formant are output. The musical tone signal can be processed externally, the musical tone signal thus processed is inputted by the musical tone signal input means, and the inputted musical tone signal is generated by the musical tone signal synthesizing means and the voiced sound formant generating means. By synthesizing the musical tone signal and the musical tone signal generated by the unvoiced sound formant generating means, it is possible to enrich the timbre variation. Further, since the voice signal of the voiced sound formant and the tone signal of the unvoiced sound formant can be individually processed, the processing accuracy of the timbre is improved.

A musical sound synthesizer according to claim 3 of the present invention is
Voiced form that produces a musical tone signal
Generates voice signal of unvoiced formant and cloak generation means
Unvoiced sound formant generation means to be generated and musical tone transmission from the outside
And a voiced sound form
Tone signal generated by the sound generating means, the unvoiced sound
Musical tone signal generated by the cloak generating means and the musical tone signal
No.1 input means for selectively synthesizing tone signals
Synthesized by the sound signal synthesizing means and the musical tone signal synthesizing means
Tone signal, generated by the voiced sound formant generating means
Generated by the unvoiced sound formant generating means.
External output hand that selectively outputs the generated tone signal to the outside
And the tone signal synthesized by the tone signal synthesizer,
For selectively synthesizing with the musical sound signal synthesizing means together with the musical sound signal generated by the voiced sound formant generating means, the musical sound signal generated by the unvoiced sound formant generating means and the musical sound signal input by the musical sound signal input means. in, characterized by comprising a feedback means for feeding back a musical tone signal synthesized by the musical tone signal synthesizing means to said tone signal synthesis means.

According to the musical sound synthesizing apparatus of the third aspect having the above structure, in addition to the operational effect of the second aspect, the musical tone signals synthesized by the musical tone signal synthesizing means are processed by the voiced sound formant generating means. Generated tone signals, tone signals generated by unvoiced formant generation means, externally processed voiced tone formant tone signals and unvoiced tone formant tone signals can be combined, respectively, providing a wider variety of tone colors Can be

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below. FIG. 2 is a block diagram of an electronic musical instrument system to which the musical sound synthesizing apparatus according to the embodiment of the present invention is applied, and shows a case of a keyboard type electronic musical instrument including a CPU 10 or a personal computer. For example, when configured as an electronic musical instrument, the program memory 20 is a ROM and the data / working memory 30 is a RAM. The input device 40 is various switches on the operation panel, and the performance operator 50 is a keyboard or the like. Further, the display 60 is a liquid crystal panel or the like on the operation panel surface, and the communication interface 70 is an interface provided with a MIDI terminal or the like. The formant sound generation circuit 1 is a musical sound synthesizer according to an embodiment of the present invention, and is composed of an LSI chip and mounted on an electronic musical instrument.

In the case of a personal computer or the like, the program memory 20 is a hard disk device, a CD-ROM device or other external storage device, and the data / working memory 30 is a RAM or the like. The input device 40 and the performance operator 50 are keyboards and the like, the display 60 is a CRT or a liquid crystal panel, and the communication interface 70 is a MIDI interface and a modem connected to various networks. The formant sound generation circuit 1 is a sound source board or sound source device having an LSI chip corresponding to the musical sound synthesizer according to the embodiment of the present invention.

In either the electronic musical instrument or the personal computer as described above, the process for the formant sound generating circuit 1 according to the embodiment of the present invention is performed by various parameters and various parameters for determining the formant waveform and the finally output musical tone signal. Since this is a process for outputting data or various control signals, description will be made using the names of the blocks in FIG. 2 without distinguishing between electronic musical instruments and personal computers.

The CPU 10 controls the entire system using the working area of the data / working memory 30 based on the control program stored in the program memory 20. The program memory 20 stores various parameters for generating various formants corresponding to the various tones such as various natural musical instruments and voice sounds. When the tone color is selected by the input device 40 or the like, the parameters are stored. The parameter corresponding to the timbre is read from the program memory 20 and stored in a predetermined area of the data / working memory 30. Then, the CPU 10 outputs data / data according to the key code and touch data detected by the performance operator 50.
The parameters of the working memory 30 are modulated (or changed) and output as formant data to the formant sound generation circuit 1.

Further, the CPU 10 has mixer data for designating the volume and panning set by the input device 40,
As described later, mute data for muting a 16-channel formant signal for each channel is output to the formant sound generation circuit 1. Furthermore, CP
U10 outputs a control signal to the formant sound generation circuit 1 for designating a musical tone generation timing, a pitch, a designation of input / output selection switching described later.

The formant sound generation circuit 1 includes a CPU 10
It is composed of a CPU interface 11 having various registers, which is an interface with respect to, a formant waveform calculation unit 12 for generating formants and performing various kinds of synthesis, and an input / output interface 13 which is an interface with an external system (or expansion system) 2. . Based on the various data and control signals from the CPU interface 11, the formant waveform calculator 12 generates a formant tone signal and outputs it to the external system 2 via the input / output interface 13 to generate a tone. .

The external system 2 includes various filters,
A DSP (digital signal processor) or a computer and an input / output interface are used to process a musical tone signal output from the formant tone generating circuit 1 and add various effects. The external system 2 processes the formant tone signal output from the formant tone generating circuit 1 and inputs the processed tone signal or the individually generated tone signal to the formant tone generating circuit 1. Formant sound generation circuit 1
To synthesize a new musical tone signal.

The formant waveform calculator 12 generates a set of 16 channels of voiced and unvoiced sound formants by time division processing with one set of one voiced sound formant and one unvoiced sound formant as one channel.
FIG. 3 is a diagram illustrating a configuration for generating a one-channel formant. Voiced sound formant generation circuit (VF
G) 12a is a voiced sound formant basic waveform generator 12a
-1, a voiced sound formant envelope generator 12a-2 and a multiplier 12a-3,
Formant data of voiced sound formants and control signals are input to the VFG 12a.

Further, an unvoiced sound formant generation circuit (UF
G) 12b is an unvoiced formant basic waveform generating section 12b
-1, unvoiced formant envelope generator 1
2b-2 and a multiplier 12b-3, and U
Formant data of unvoiced formants and control signals are input to the FG 12b. In the following description and notation of parameters and the like, "V" is added to the voiced sound formants, and "U" is added to the unvoiced sound formants.

The formant data of the voiced sound formant is a pitch shift amount PSHIFTV indicating a shift amount of a musical tone pitch modulated based on touch data, a center frequency CFREQV of the formant, a shape parameter FSHAPV for designating the formant shape, an envelope of the formant. Is an EG parameter FEGV that specifies the sound volume, and a volume parameter VOLV that specifies the volume of the formant. Similarly, the formant data for unvoiced formants has the same center frequency CFRE.
QU, shape parameter FSHAPU, EG parameter FEGU, and volume parameter VOLU.

Further, the control signal is a tone generation instruction signal KON for instructing the start of tone generation, that is, generation of a tone signal of a formant, a pitch frequency PITCH for instructing a pitch corresponding to the generated formant, and an input for determining an input selection described later. Selection signal SYSW1, output selection signal SY that determines output selection
SW2. The control signal input selection signal SY
SW1 and the output selection signal SYSW2 are used by the voiced sound formant generation circuit (VFG) 12a and the unvoiced sound formant generation circuit (U
It is not output to the FG) 12b.

With the above configuration, the voiced sound formant basic waveform generating section 12a-1 generates the basic waveform of the voiced sound formant based on the formant data of the voiced sound formant, the pronunciation instruction signal KON and the pitch frequency PITCH. The voice sound formant EG12a-2 outputs the envelope value of the voiced sound formant, the basic waveform and the envelope value are multiplied by the multiplier 12a-3, and output as the tone signal V of the voiced sound formant. In addition, unvoiced formant formant data and pronunciation instruction signal KON
Based on the unvoiced formant basic waveform generator 12b
-1 generates the unvoiced formant basic waveform, the unvoiced formant EG12b-2 outputs the unvoiced formant envelope value, the basic waveform and the envelope value are multiplied by the multiplier 12b-3, and output as the unvoiced formant tone signal U. To be done.

The formant data is stored in the register of the CPU interface 11, and the formant data for each channel is output in each slot of the 16-channel time division processing, and is combined with the musical tone signal V of one voiced sound formant. Two unvoiced formant musical tone signals U are output as a set, one set for each channel. In the following description, each of the 16 channels that generate this formant is referred to as a "formant CH" to distinguish it from the left and right stereo channels.

FIG. 1 is a block diagram of essential parts of the formant sound generation circuit 1. Group data register 11a, mixer data register 11b, mute data register 11
c, the formant data register 11d, and the control signal register 11e are provided in the CPU interface 11, respectively. The group data register 11a stores group data that specifies the dry-output formant CH in association with each bit of 16 bits. In the mixer data register 11b,
Parameter that specifies the overall volume of the voiced sound formant
V-Vol, parameter U-Vol that specifies the overall volume of unvoiced formants, parameter PAN-Left that specifies the volume of the left channel that determines stereo panning, and parameters SENDLVL1 to SEND that specify the send output level.
LVL4 is stored under the control of the CPU 10 (FIG. 2).

Further, in the mute data register 11c, mute data MUTE for designating whether or not to mute the musical tone signal for each formant CH in correspondence with each bit of 16 bits is stored under the control of the CPU 10,
The formant data register 11d and the control signal register 11e store the formant data and the control signal under the control of the CPU 10, respectively.

As shown in FIG. 1, the output of the mute data register 11c (mute data MUTE) is input to the AND circuit a, and a timing signal T, which will be described later, is input to the AND circuit a, and the AND circuit a outputs. The output is the output of the mixer data register 11b (mixer data V-Vol, ...
Etc.) to the OR circuit b. The output of the OR circuit b is input to the EXOR circuit c and the fourth selector (SEL4) d, and the timing signal T is input to the EXOR circuit c. The output of the EXOR circuit c is input to the panning curve generator (PCG) e, and the output of the PCGe is input to the fourth selector d. The timing signal T is input to the fourth selector d, and the output of the OR circuit b and the output of PCGe are selected by the fourth selector d according to the timing signal T and output to the interpolation circuit f.

The output of the formant data register 11d (formant data PSHIFTV, ..., And the like) is input to the voiced sound formant generation circuit (VFG) 12a and the unvoiced sound formant generation circuit (UFG) 12b, and the output of the control signal register 11e is generated. Instruction signal KON
Is inputted to the interpolation circuit f, the VFG 12a and the UFG 12b, and the pitch frequency PITCH is inputted to the VFG 12a.
In addition, the input selection signal of the control signal register 11e
SYSW1 is output selection signal SYSW to the first selector (SEL1) g.
2 is input to the second selector (SEL2) h.

The tone signal V (CH) of the voiced sound formant which is the output of the VFG 12a is supplied to the third selector (SEL3) i and the second selector.
The unvoiced formant tone signal U (CH) output from the UFG 12b is input to the third selector i, and the output SEL1OUT (CH) of the first selector g is input to the third selector i. Three types of timing signals T are input to the third selector i, and the output of the VFG 12a, the output of the UFG 12b, and the output of the first selector g are selected by the third selector i according to the timing signal T.

The output of the interpolation circuit f and the output of the third selector i are input to the multiplication circuit j, and the multiplication outputs of the multiplication circuit j are input to the dry-out accumulator k, the send accumulator m and the VU synthesis accumulator n, respectively. . The outputs of the dry-out accumulator k and the send-out accumulator m are input to the input / output interface 13, and the output of the VU synthesis accumulator n (V +
U ACC OUTPUT (CH)) is a first selector g and a second selector h.
And the output of the second selector h is input to the input / output interface 13.

The input / output interface 13 outputs the output of the dry-out accumulator k to the dry-out 1 output DRY.
OUT1 and dry out 2 output DRY OUT2 output, send out accumulator m output send out 1,2
Output SEND OUT1,2 Send out 3,4 Output SEND OUT3,4
Output as. In addition, DRY OUT1, DRY OUT2, SEND OUT
Each of 1,2 and SEND OUT3,4 has two left and right stereo channels. The input / output interface 13 also outputs the output of the second selector h to the channel output CHANNE.
Output as L OUTPUT, and input the tone signal supplied from the outside as channel input CHANNEL INPUT.This channel input is the external channel input EXT CHANNEL INPUT (C
H) is input to the first selector g.

Formant sound generation circuit 1 of this embodiment
Is a sampling cycle of the amplitude of the digital musical tone signal, that is, one conversion period for converting the digital musical tone signal into an analog signal is divided into 128 slots to perform time-division processing. Is in sync with. Each slot is "0 to 127"
The timing signals T, which correspond to the numbers No., and are input to the AND circuit a, the EXOR circuit c, the fourth selector d, and the third selector i, switch in a predetermined slot period.
That is, the set of numbers "xy" shown in each timing signal T in FIG. 1 indicates that the timing signal T is "1" during the slots x to y and "0" during the other periods. ing.

The first selector g selects the output of the VU synthesis accumulator n when the input selection signal SYSW1 is "0" and outputs it to the third selector i. When the input selection signal SYSW1 is "1", the external channel input EXT CHANNEL INPUT (C
H) is selected and output to the third selector i. Also, the second
The selector h selects the output of the VU synthesis accumulator n when the output selection signal SYSW2 is “0” and outputs it to the input / output interface 13. When the output selection signal SYSW2 is “1”, the output of the VFG 12a is selected and input / output. Output to the interface 13.

Further, the third selector i selects the output of the VFG 12a in the period of slots 0 to 15 by the timing signal T, and the UFG 12b in the period of slots 16 to 31.
Output of the first selector g and the output SEL1OUT (CH) of the first selector g are selected and output to the multiplication circuit j. The fourth selector d selects the output of PCGe in the period of slots 32 to 63 according to the timing signal T and outputs it to the interpolation circuit f, and O in the other periods.
The output of the R circuit b is selected and output to the interpolation circuit f.

4 to 8 are timing charts showing an example of the operation of the embodiment, and the circuit operation will be described with reference to the figures. 4 to 8, the uppermost stage shows 128 arithmetic slots and the second stage shows the formant CH slots. First, since the third selector i selects the output of the VFG 12a in the slots 0 to 15 by the timing signal T, as shown in FIG. 4, the musical tone signal V (CH) of the voiced sound formant is transmitted to each formant CH. Output to the multiplication circuit j in synchronization with slots 0 to 15. Since the output of the UFG 12b is selected in the slots 16 to 31, the tone signal U (CH) of the unvoiced sound formant is output to the multiplication circuit j in synchronization with the slots 16 to 31 of each formant CH.

On the other hand, in the mixer data register 11b, the volume parameter V-Vol (CH) of the voiced sound formant is set in slots 0 to 15 in slots 0 to 1 for each formant CH.
5, the unvoiced formant volume parameter U-Vol (CH) is output in slots 16 to 31 in synchronization with slots 16 to 31 for each formant CH. As shown in FIG. 4, the mute data register 11c
Outputs mudata MUTE (CH) in slots 0-31.
Now, assuming that each bit of the mute data MUTE (CH) is "0" and each formant CH is not muted, A
Since the output of the ND circuit a becomes "0", the OR circuit b outputs the value (0/1) of each bit of the data output from the mixer data register 11b as it is. Further, the fourth selector d selects the output of the OR circuit b in slots 0 to 31 (other than 32 to 63) and outputs it to the interpolation circuit f.

That is, the parameter V-Vol (CH) of the voiced sound formant volume in slots 0 to 15 is
Volume parameter of unvoiced formant in 6 to 31 U-Vo
l (CH) is interpolated by the interpolator f via the fourth selector d.
Is output to. The interpolation circuit f is for removing click noise caused by an extreme change in the volume parameter, and the parameter V-Vol is changed by the sounding instruction signal KON.
The interpolation of (CH) and U-Vol (CH) is started, and the interpolated parameters V-Vol (CH) and U-Vol (CH) are output to the multiplication circuit j.

Then, the multiplication circuit j has slots 0 to 15
Parameter to the voice signal V (CH) of the voiced sound formant.
V-Vol (CH) is multiplied for each formant CH and sequentially output, and the multiplication output is stored in each adder register corresponding to the formant CH of the VU synthesis accumulator n. In addition, the multiplication circuit j uses the parameters V- to the unvoiced formant musical tone signal U-Vol (CH) in slots 16 to 31.
Vol (CH) is multiplied for each formant CH and sequentially output, and the multiplication result is added to the multiplication result for the voiced sound formant in the adder register corresponding to the formant CH of the VU synthesis accumulator n.

As described above, the VU synthesis accumulator n determines the synthesis result of the tone signal of each voiced and unvoiced formant, and as shown in FIG. 4, the VU synthesis accumulator n outputs the synthesis result to V +. Output as U ACC OUTPUT (CH) for each formant CH.

As in the above case, the mute data is
If mute is not specified by MUTE (CH), the output of the OR circuit b is the same as the output of the mixer data register 11b, but the content of the mute data MUTE (CH) is the bit corresponding to the formant CH to be muted. When it is "1", the mute is applied as follows. That is, the AND circuit a, during the period of slots 0 to 31, the OR circuit b corresponding to the bit for which mute is designated.
Since the output to "1" is set to "1", the output of the OR circuit b becomes all 1s in the channel corresponding to the bit. That is, in this embodiment, the volume parameter V-Vol (CH),
U-Vol (CH) uses a negative value corresponding to the decibel, and when each bit of this parameter V-Vol (CH) and U-Vol (CH) is all 1, the absolute value is the maximum. It takes a negative value and the volume is muted.

Next, as shown in FIG. 4, the third selector i selects the output SEL1OUT (CH) of the first selector g in slots 32 to 127 and outputs it to the multiplication circuit j. Further, the mixer data register 11a has slots 32 to 6
Output the panning parameter PAN-Left in the period of 3,
The parameter PAN-Left is output to the fourth selector d and the EXOR circuit c via the OR circuit b, but the fourth selector d receives the timing signal T and outputs P in slots 32 to 63.
The output of CGe is selected. On the other hand, the EXOR circuit c outputs the bit of the parameter PAN-Left as it is because the timing signal T becomes "0" in the slots 32 to 47, but the timing signal T becomes "1" in the slots 48 to 63. Inverts and outputs the bit of the parameter PAN-Left.

The value of the panning parameter becomes the maximum value by combining the values of the left channel and the right channel, and the inverted value of the bit of the parameter PAN-Left of the left channel becomes the parameter PAN-Right of the right channel. Become. That is, in slots 32 to 47, the parameter PAN-Left of the left channel is input to PCGe and slot 4
In 8-63, the parameter PAN-Right of the right channel is P
Input to CGe. PCGe converts the panning parameters with a curve or the like so that the volume of the left and right channels produces a panning effect on the actual hearing, and the converted parameters PAN-Left and PAN-Right are input to the fourth selector d. It And this parameter PAN-Left, PA
N-Right is input from the fourth selector d to the multiplication circuit j via the interpolation circuit f.

In this way, as shown in (4) of FIG. 4, in the slots 32 to 47, the left channel panning parameter PAN-Left and the output SEL1OUT of the first selector g output via the third selector i. (CH) is the multiplication circuit j
Input to the slots 48 to 63, and the right channel panning parameter PAN-Right and the output of the first selector
SEL1OUT (CH) is output to the multiplication circuit j. Then, in the slots 32 to 47, the multiplication circuit j multiplies the output SEL1OUT (CH) of the first selector g by the parameter PAN-Left for each formant CH and sequentially outputs the same.
At 63, the parameter is set to the output SEL1OUT (CH) of the first selector g.
The PAN-Right is multiplied by each formant CH and sequentially output, and each multiplication output is stored in the left channel register and the right channel register corresponding to the formant CH of the dryout accumulator k, respectively.

As described above, the result of multiplying the output of the first selector g by the left and right panning parameters PAN-Left and PAN-Right by the dry-out accumulator k is determined, and shown in FIG. As described above, the group data GROUPSEL1 (CH), GROUPSEL2 (C) output from the group data register 11a in the slots 48 to 111 are stored.
Based on H), the tone signal of the specified formant CH is output. That is, the input / output interface 1
LEFT1 (CH) of the left channel and RIGHT1 (CH) of the right channel are output for dry out 1 output DRY OUT1 in 3. In addition, LEFT2 (CH) of the left channel and RIGHT2 (CH) of the right channel are output for dry out 2 output DRY OUT2. Group data GROUPSEL1 (CH), GROUPSEL2
(CH) is composed of 16 bits corresponding to the formant CH, the formant CH corresponding to the bit "1" is output, and the formant CH corresponding to the bit "0" is not output.

Next, the mixer data register 11a sends the send level parameter SEND during the slots 64 to 79.
Send LVL1 (CH) to SENDLVL2 (CH) in the period of slots 80 to 95
, SENDLVL3 (CH) in the period of slots 96 to 111, and SENDLVL4 (CH) in the period of slots 112 to 127 are output in synchronization with slots 0 to 15 for each formant CH. This send level parameter SENDLVL1
(CH) to SENDLVL4 (CH) are output to the fourth selector d via the OR circuit b, and are output to the multiplication circuit j via the fourth selector d and the interpolation circuit f as shown in FIG. It The third selector i selects the output SEL1OUT (CH) of the first selector g in slots 64-127 and outputs it to the multiplication circuit j. Therefore, from the multiplication circuit j, as described above,
Each parameter SEND to the output SEL1OUT (CH) of the first selector g
LVL1 (CH) ~ SENDLVL4 (CH) are each formant CH
Each of them is multiplied and sequentially output, and the multiplied output is stored in each register corresponding to the formant CH of the send-out accumulator m.

As shown in FIG. 4, from the sendout accumulator m, the sendouts 1 and 2 outputs SEND OUT1,2 of the input / output interface 13 and the sendouts 3 and 4 outputs SEND OUT3,4 are used. , Send-out output SEND1 OUTPUT (CH), SEND2 OUTPUT (CH),
Output as SEND3 OUTPUT (CH) and SEND4 OUTPUT (CH).

FIGS. 7 and 8 are timing charts of input / output in the input / output interface 13, and output data DRY OUT1, DRY of the dry-out accumulator k.
Output data SEND of OUT2, send-out accumulator m
1 OUTPUT (CH) to SEND4 OUTPUT (CH) and output data CHANNEL OUTPUT of the second selector h are sequentially output for each formant CH in a period of 8 slots. In the figure, SEND3
The OUTPUT (CH) and SEND4 OUTPUT (CH) are not shown. In addition, the input data CHANNEL INPUT of the channel input of the tone signal supplied from the outside is the formant C
When each H is sequentially input in a period of 8 slots and is output to the first selector g as an external channel input EXT CHANNEL INPUT (CH), as shown in FIG. After the input of the formant CH is confirmed, it is input to the first selector g in synchronization with the slot of the formant CH after the slot 32.

Here, the second selector h outputs the output selection signal SY.
When SW2 is “0”, the output of VU synthesis accumulator n is output. When the output selection signal SYSW2 is “1”, VFG1
The outputs 2a are selected and output to the input / output interface 13. Therefore, by setting the output selection signal SYSW2 as desired, the input / output interface 1
V + U, which is the synthesized result of the musical sound signal of each voiced and unvoiced formant, as the 3 channel output CHANNEL OUTPUT
The ACC OUTPUT (CH) or the musical sound signal V (CH) of the voiced formant can be output to the external system 2.

The first selector g has an input selection signal SYSW.
Output of VU synthesis accumulator n when 1 is “0” V + U
ACC OUTPUT (CH) is output to the third selector i, and when the input selection signal SYSW1 is “1”, the external channel input EXT CHANNE
Since L INPUT (CH) is output to the third selector i, by setting this input selection signal SYSW1 as desired,
Parameters PAN-Left, PAN- performed after slot 32
Right multiplication processing, parameters SENDLVL1 (CH), SENDLV
Multiply L2 (CH), SENDLVL3 (CH), SENDLVL4 (CH) by V +
U ACC OUTPUT (CH) and external channel input EXT CHANNEL IN
It can be applied to any tone signal of PUT (CH). In addition, the results are dry out 1 output DRY OUT1, dry out 2 output DRY OUT2, send out 1 and 2 output SEND O
UT1,2, send out 3,4 output SEND OUT 3,4 can be output to the external system 2.

In the above example, as shown in FIG. 5A, for example, when the set values (0/1) of the input selection signal SYSW1 and the output selection signal SYSW2 are made constant in each slot. However, for example, as shown in FIG. 5B, the input selection signal SYSW1 and the output selection signal SYSW2 are changed to formant CH.
Alternatively, the switching may be performed in synchronization with. By doing so, it is possible to specify the synthesis processing of the musical tone signals as described above and the output of the musical tone signals of the formants for each formant CH. That is, it is possible to process one formant as a unit, such as outputting and processing one formant. .

In the above example, the musical tone signal V (CH) of the voiced sound formant of the VFG 12a is input to the third selector i and the second selector h, but as shown by the broken line in FIG. The unvoiced sound signal U (CH) of the formant is input to the third selector i and the second selector h,
The tone signal U (CH) of this unvoiced sound formant can be output to the external system 2.

As described above, the voice signal of the voiced sound formant, the music signal of the unvoiced sound formant, the music signal of the combination of the voiced and unvoiced sound formants, the dry-out music signal, and the send-out music signal are output to the outside. Further, it is possible to input a musical tone signal obtained by subjecting the outputted musical tone signal to various kinds of processing and other musical tone signals and to synthesize them with the various musical tone signals.

In the above embodiment, various data, parameters or control signals are set in the formant sound generating circuit 1 in accordance with the operation of the input device 40 of the electronic musical instrument or personal computer and the performance operator 50. Various data such as formant data, parameters, or control signals may be set by data input from the communication interface 70, the hard disk device of the program memory 20, the CD-ROM device, or other external storage device.

In FIG. 2, the formant sound generating circuit 1 is described as being composed of an LSI chip, but as another structural example, the digital signal processor DS is used.
An equivalent formant sound generation process can be performed by P and the microprogram. Further, the formant sound generation software may be implemented by causing the CPU 10 to execute the formant sound generation program. In that case, the hardware corresponding to the formant tone generation circuit 1 may be omitted.

[0054]

As described above, according to the first aspect of the present invention.
According to the musical tone synthesizer described, the musical tone signal of the formant of any channel generated by the formant generating means is selected and output to the outside, and the musical tone signal of one formant can be processed externally. To be able to
Further, since the processed musical tone signal can be combined with the musical tone signal generated by the formant generating means, it is possible to enrich the variation of the tone color,
The processing accuracy of the timbre is improved.

According to the musical sound synthesizer of the second aspect of the present invention, the musical tone signal of the voiced formant and the musical tone signal of the unvoiced formant are selected and output to the outside, and The unvoiced formant tone signal can be externally processed, and the processed tone signal can be combined with the voiced formant tone signal and the unvoiced formant tone signal. In addition to being rich, the processing accuracy of the timbre is enhanced.

According to the musical sound synthesizing apparatus of the third aspect of the present invention, the musical sound signals synthesized by the musical sound synthesizing means are generated by the musical tone signals generated by the voiced sound formant generating means and the unvoiced sound formant generating means. Since it is possible to synthesize a musical tone signal, a musical tone signal of an externally processed voiced sound formant, and a musical tone signal of an unvoiced sound formant, it is possible to further enrich the timbre variation.

[Brief description of drawings]

FIG. 1 is a block diagram of a main part of a formant sound generation circuit according to an embodiment of the present invention.

FIG. 2 is a block diagram of an electronic musical instrument system to which the musical sound synthesizing apparatus according to the embodiment of the present invention is applied.

FIG. 3 is a diagram illustrating a configuration for generating a one-channel formant according to the embodiment.

FIG. 4 is a timing chart of multiplication processing and multiplication result output in the embodiment.

FIG. 5 is a timing chart of an input selection signal and an output selection signal in the embodiment.

FIG. 6 is a timing chart of group data and dryout output in the embodiment.

FIG. 7 is a timing chart of slots in the first half of input / output data of the input / output interface according to the embodiment.

FIG. 8 is a timing chart of the latter half slot of input / output data of the input / output interface in the embodiment.

[Explanation of symbols]

1 ... Formant sound generation circuit, 2 ... External system, 11
b ... Mixer data register, 11d ... Formant data register, 11e ... Control signal register, 12
a ... voiced sound formant generation circuit, 12b ... unvoiced sound formant generation circuit, g ... first selector, h ... second selector, i ... third selector, k ... dry-out accumulator, m ... send-out accumulator, n ... VU synthesis accumulator , 13 ... Input / output interface.

─────────────────────────────────────────────────── ─── Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) G10H 1/00-7/12 G10K 15/00-15/06 G10L 13/00

Claims (3)

(57) [Claims] 1. Formant generating means for generating tone signal of a plurality of formant components on a plurality of channels, and externally outputting externally a tone signal of any channel among the tone signals generated by the formant generating means. Output means, tone signal input means for inputting the tone signal output from the external output means and processed externally , and the tone signal of the formant component generated by the formant generation means and the tone. A musical sound synthesizing device comprising: a musical sound signal synthesizing means for synthesizing a musical tone signal inputted by a signal input means. 2. A voiced sound formant generating means for generating a musical sound signal of a voiced sound formant, an unvoiced sound formant generating means for generating a musical sound signal of an unvoiced sound formant, and a musical sound signal input means for inputting a musical sound signal from the outside. Musical sound signal synthesizing means for selectively synthesizing the musical sound signal generated by the voice sound formant generating means, the musical sound signal generated by the unvoiced sound formant generating means, and the musical sound signal input by the musical sound signal input means, and the musical sound signal synthesizing synthesized musical tone signal by means and an external output means for outputting a musical tone signal generated by the generated musical tone signal and the unvoiced formant generating means selectively outside the voiced formant generator, said tone The signal input means outputs from the external output means.
Input the tone signal that has been processed externally.
Musical tone synthesizing apparatus characterized by that. 3. A tone signal of a voiced sound formant is generated.
Means for generating voiced sound formants and unvoiced sound formants for generating tone signals of unvoiced sound formants.
A cloak generating means, a musical sound signal inputting means for inputting a musical sound signal from the outside, and a musical tone signal generated by the voiced sound formant generating means.
No., tones generated by the unvoiced sound formant generating means
Signal and effortless sound signal inputted by said musical tone signal input means
Tone signal synthesizing means for selectively synthesizing the tone signal, the tone signal synthesized by the tone signal synthesizing means, and the voiced voice
The tone signal generated by the sound formant generation means and
The voice signal generated by the unvoiced sound formant generation means
An external output means for selectively outputting to the outside, and a musical tone signal synthesized by the musical tone signal synthesizing means, a musical tone signal generated by the voiced sound formant generating means by the musical tone signal synthesizing means, and an unvoiced sound formant generating means. to selectively synthesized with the input musical tone signal in the generated musical tone signal and the tone signal input means, a feedback means for feeding back a musical tone signal synthesized by the musical tone signal synthesizing means to said tone signal synthesis section , comfortable sound synthesizer comprising the.