JP3404953B2 - Music synthesizer - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention simulates a resonance part (for example, a flute tube) of an acoustic musical instrument with an electronic circuit, and the electronic circuit is oscillated by a vibration of an oscillating part of the acoustic musical instrument (for example, blown into the mouth of the flute). The present invention relates to a musical sound synthesizing device for synthesizing desired synthetic data by inputting drive data corresponding to breath) and causing resonance.

[0002]

2. Description of the Related Art In recent years, a musical tone synthesizer simulating the above-described musical tone generation structure of a musical instrument has been proposed (for example, Japanese Patent Laid-Open No. 62-109093).

A conventional tone synthesizer as described above will be described below with reference to the drawings.

FIG. 16 is a block diagram of a conventional tone synthesizer. In FIG. 16, reference numeral 1601 is a drive data generation circuit generated by reading out prestored drive data, 1602 is a delay device for delaying the data for a predetermined time,
Reference numeral 1603 denotes an adder for adding drive data and data output from the delay device 1602, and 1604 denotes a delay device 1602.
And an adder 1603 in a closed loop form.

The operation of the conventional tone synthesizing apparatus configured as described above will be described. Drive data generation circuit 160
The comb filter 16 outputs drive data 1 corresponding to vibrations (eg, force applied from a finger to a string when repelling a string) when driving a resonance part (eg, string or tube) of a musical instrument.
Enter 04. The comb filter 1604 is a circuit corresponding to the resonance part (string or tube) of the musical instrument, and acts as a comb filter that forms a resonance peak in an integer harmonic by using the frequency corresponding to the delay stage number j of the delay device 1602 as the first-order component. , Get the desired composite data. The delay unit 1602 uses a read / write memory, and if the write address precedes the read address of the read / write memory by m stages, it operates as a delay unit of m stages. That is, it is used as a ring memory.
Since the method of configuring the ring memory is generally known, its explanation is omitted. The drive data is the comb filter 1
It can be obtained by convolving the original sound of the musical instrument with the inverse characteristic of 604. The drive data is input to the comb filter 1604, and the drive data circulates in the comb filter 1604 to combine the combined data. By such an operation, a musical instrument sound can be reproduced. It is generally known that a resonance level is controlled by inserting a multiplier in the closed loop of the comb filter 1604 and changing the multiplier of the multiplier between the value 0 and the value 1. Here, the description of the control of the resonance level is omitted.

[0006]

However, as shown in FIG.
In the configuration as shown in FIG. 5, the comb filter 1604 can uniformly control the level of the integer harmonics of the fundamental frequency of the synthesized data, but the level of the odd harmonics is kept constant and the level of only the even harmonics is controlled. Therefore, it is impossible to realize the timbre change (level control of only even harmonics) according to the blowing angle to the singing mouth, which is a characteristic of flute-type instrument sounds.

Further, since the comb filter 1604 cannot act as a comb filter having a fundamental frequency (first harmonic) lower than the fundamental frequency of the synthesized data, the octave key, which is a characteristic of the flute type instrument sound, can be suppressed. For example, it is impossible to synthesize a sound in which a weak resonance sound having a fundamental frequency lower than the fundamental frequency by 1 octave or 1 octave and 5 degrees lower than the fundamental frequency is generated. In other words, it is not possible to realize a tone color change that sounds like a flute instrument sound according to the musical range.

Further, since the comb filter 1604 acts as a comb filter that amplifies each integer-order overtone by the same level, the order of the overtone is 3, 4 like a sax tube.
It is not possible to faithfully simulate the characteristic that the resonance level of the overtones below the peak is reduced and the resonance level of the overtones below the peak is reduced.

Further, since the drive data generation circuit 1601 is a circuit for reading digital data stored in advance,
It is not possible to sample analog acoustic signals such as voice in real time and resonate them in a circuit corresponding to the resonance part (string or tube) of a musical instrument.

Further, since the register word length of each circuit element such as the delay device 1602 is limited and the circuit is a recursive linear filter, when the loop gain approaches 1 and the resonance level is increased, the comb filter 1604 The data that circulates overflows, causing unpleasant noise.

In view of the above problems, the present invention can realize a timbre change (level control of only even overtones) according to the blowing angle into the singing mouth, which is a characteristic of a flute-type musical instrument sound. An object is to provide a musical sound synthesizer.

It is another object of the present invention to provide a musical tone synthesizing device capable of realizing a timbre change like a flute-type musical instrument sound according to a musical range.

It is another object of the present invention to provide a musical tone synthesizer capable of faithfully simulating the characteristics of a saxophone pipe in which the harmonics having harmonic orders near the 3rd and 4th peaks are peaked and the resonance level of the harmonics below that peak is reduced. To aim.

It is another object of the present invention to provide a musical tone synthesizer capable of sampling an analog acoustic signal such as voice in real time and resonating it in a circuit corresponding to a resonance portion (string or tube) of a musical instrument.

It is another object of the present invention to provide a musical tone synthesizer in which no unpleasant noise is generated even if the resonance level of a circuit corresponding to the resonance part (string or tube) of a musical instrument is increased.

[0016]

In order to achieve this object, the musical tone synthesizer of the present invention comprises a drive data generating means for reading drive data stored in advance and a circuit in which a delay device is formed in a closed loop. A circuit in which the drive data is circulated and the circulated data is extracted as the first waveform data and a circuit in which a delay device is configured in a closed loop are used to circulate the drive data to convert the circulated data into the second waveform data. A second comb filter that is taken out as waveform data, a mixing means that mixes the first waveform data and the second waveform data to generate desired combined data, and a mixing that controls the mixing ratio of the mixing means. And a ratio control means.

Further, the musical tone synthesizer of the present invention is such that the drive data generating means for reading the drive data stored in advance and the drive data are circulated by a circuit having a delay device in a closed loop form, and the data being circulated is the first one. And a second comb filter in which the driving data is circulated by a circuit having a closed loop-shaped delay device and the circulated data is extracted as the second waveform data. And mixing means for mixing the first waveform data and the second waveform data to generate desired combined data, and changing the delay time of the delay device in the second comb filter according to the fundamental frequency of the combined data. And a delay time changing means for changing the delay time.

[0018]

[0019]

[0020]

[0021]

With this configuration, the drive data generating means generates data corresponding to the vibration generated when the breath is blown into the tube, and the first comb filter generates the component corresponding to the odd harmonic of the fundamental frequency of the composite data. The second comb filter amplifies the components corresponding to the even harmonics
And second waveform data are generated, and the mixing means mixes the first and second waveform data to combine the combined data. At this time, the mixing ratio control means controls the mixing ratio of the second waveform data based on the tone color control parameter corresponding to the blowing angle, thereby changing the tone color according to the blowing angle to the song mouth (only the even harmonics Level control).

With this configuration, the drive data generating means generates data corresponding to the vibration generated when the breath is blown into the tube, and the first comb filter corresponds to the integer harmonic of the fundamental frequency of the composite data. And a second comb filter amplifies a component corresponding to an integer harmonic of a frequency lower than the fundamental frequency of the synthetic data to generate the first and second waveform data, and the mixing unit generates the first waveform data.
And the second waveform data are mixed to synthesize the synthesized data. At this time, the delay time changing means changes the delay time of the delay device in the second comb filter to about 2 times or about 3 times the delay time of the delay device in the first comb filter, thereby changing the fundamental frequency. Therefore, a sound in which a weak resonance sound having a low frequency of one octave below or one octave and 5 degrees below is superposed is synthesized. That is, it is possible to realize a timbre change that sounds like a flute-type musical instrument sound according to the musical range.

With this configuration, the drive data generating means generates data corresponding to the vibration of the lead, which is the nozzle of the tube, and the first comb filter corresponds to an integer harmonic of the fundamental frequency of the composite data. And the second comb filter outputs one of the integer harmonics of the fundamental frequency of the synthesized data.
By amplifying low-order components up to the second and second orders, and subtracting means subtracting the second waveform data from the first waveform data, the overtones having harmonic orders near the third and fourth order are peaked. This will faithfully simulate the characteristics of the sax tube that will reduce the resonance level of the following overtones.

With this configuration, the microphone converts an acoustic signal such as voice into an analog electric signal, the analog-digital converter converts the analog electric signal into digital data, and the comb filter resonates the digital data. Sound such as voice is sampled in real time and resonated in a circuit corresponding to the resonance part (string or tube) of the musical instrument.

With this structure, the drive data generating means generates drive data, and the comb filter circulates the drive data to synthesize desired synthesized data. At this time, the non-linear conversion means in the comb filter performs data level conversion processing so that the circulating data does not exceed a predetermined level, so even if the comb filter resonance level is increased, it is uncomfortable due to data overflow. The synthesized data is synthesized without generating any noise.

[0026]

【Example】

(First Embodiment) A first embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a musical sound synthesizing apparatus according to the first embodiment of the present invention. In FIG. 1, 101 is a drive data generation circuit that reads out pre-stored data and outputs it as drive data, 102 and 104 are delay devices that delay the data for a predetermined time, 103 is drive data and the data output from the delay device 102. Subtractor for subtracting and 10
5 is an adder for adding the drive data and the data output from the delay unit 104, 106 is a mixing circuit for mixing the waveform data S1 and S2, 107 is a mixing ratio control circuit for controlling the mixing level of S2, and 108 is Reference numeral 109 is a comb filter in which the delay device 102 and the subtractor 103 are configured in a closed loop, and 109 is a comb filter in which the delay device 104 and the adder 105 are configured in a closed loop.

FIG. 2 is a circuit diagram of the mixing circuit 106. In FIG. 2, 201 is a multiplier that multiplies the waveform data S2 and ratio data W that determines the mixing level of S2, and 202 is an adder that adds the waveform data S1 and the multiplication result of the multiplier 201.

FIG. 3 is a circuit diagram of the mixing ratio control circuit 107. In FIG. 3, reference numeral 301 is a table for reading out pre-stored data using the tone color parameter as an address.

FIG. 4 is a transfer characteristic diagram of the comb filter.
In FIG. 4, (A) is a transfer characteristic diagram of the comb filter 108, and (B) is a transfer characteristic diagram of the comb filter 109.
The horizontal axis represents the order of each overtone when the fundamental frequency of the synthetic data is the first order, and the vertical axis represents the amplitude value.

FIG. 5 is an amplitude spectrum diagram of various data. 5, (A) is an amplitude spectrum diagram of drive data, (B) is an amplitude spectrum diagram of waveform data S1,
(C) is an amplitude spectrum diagram of the waveform data S2, and (D) is an amplitude spectrum diagram of the combined data. The drive data corresponds to, for example, vibration of air generated in a blow mouth (song mouth) when a breath is blown into a pipe in a flute, and is a transfer characteristic of the comb filters 108 and 109 and the mixing circuit 106 (that is, drive data). It is obtained by convoluting the original sound data of the flute with respect to the inverse characteristic of the transfer characteristic of the circuit from the input point of the data to the output point of the synthesized data). The broken line in (A) represents a component having no periodicity such as noise (a level smaller than the periodic component).

The operation of the musical tone synthesizer configured as above will be described. The characteristic of the flute-type musical instrument sound is that the amplitude level of the even harmonics changes according to the blowing angle to the mouth. The drive data generation circuit 101 generates drive data corresponding to the vibration of air that occurs when breathing into the tube. Since the comb filter 108 is a negative feedback type comb filter, it functions as a filter having resonance peaks of odd order as shown in FIG. Here, if the frequency corresponding to the first-order component is f1, the relationship of (Equation 1) is established.

[0033]

[Equation 1]

On the other hand, the comb filter 109 is a positive feedback type comb filter, and therefore, it functions as a filter having resonance peaks of even-numbered orders as shown in FIG. 4B.

By superposing the transfer characteristics of FIGS. 4A and 4B on the amplitude spectrum of the drive data shown in FIG. 5A, the amplitude spectra of the waveform data S1 and S2 are shown in FIG. ), (C).
The mixing circuit 106 mixes these data, and synthetic data having characteristics as shown in FIG. 5D is obtained. The mixing ratio control circuit 107 generates the ratio data W via the table 301 according to the timbre parameter corresponding to the blowing angle of the flute to the mouth, and controls the weighting of S2 in the multiplier 201 in the mixing circuit 106. .

As described above, according to the first embodiment of the present invention, the comb filter 109 generates the waveform data S2 corresponding to the even harmonic component of the composite data, and determines the blowing angle of the flute to the mouthpiece. Since the mixing ratio control circuit 107 controls the mixing ratio of the waveform data S2 according to the corresponding timbre parameter, it is possible to change the amplitude level of the even-order overtone of the synthetic data. It is possible to realize a tone color change according to the blowing angle to the song mouth.

(Second Embodiment) A second embodiment of the present invention will be described below with reference to the drawings.

FIG. 6 is a block diagram of a tone synthesis apparatus according to the second embodiment of the present invention. In FIG. 6, 601
Is a delay device that delays data for a predetermined time, 602 is an adder that adds drive data and data output from the delay device 601, and 603 is a delay device that delays data for a predetermined time.
Reference numeral 604 is an adder for adding the drive data and the data output from the delay unit 603, 606 is a delay time change table for changing the delay time of the delay unit 603 according to the pitch data, and 607 is the delay unit 601 and the adder. Reference numeral 602 denotes a comb filter having a closed loop shape, and 608 denotes a comb filter having a delay unit 603 and an adder 604 having a closed loop shape.
The other circuits are the same as those of the musical sound synthesizer of the first embodiment. The ratio data W of the mixing circuit 106 has a value sufficiently smaller than 1.

FIG. 7 is a circuit diagram of the delay device 603. Figure 7
701 is a read / write memory that temporarily stores data, 702 is a counter that performs an increment operation based on the generation timing of the system clock SCK, 703 is a subtractor that subtracts the count value of the counter 702 and the delay stage number n, and 704 is a counter 702. It is a selector that selects the count value of 1 and the subtraction result of the subtractor 703. The system clock SCK is generated every sampling time Ts, and the selector 704 alternately selects the A input and the B input within the Ts time. When the A input is selected, the write address is output as the address, and when the B input is selected, the read address is output as the address. Further, the counter 702 repeatedly increments between values 0 and n which are sufficiently larger than n.

FIG. 8 is a longitudinal sectional view of a flute tube. FIG. 9 is a transfer characteristic diagram of the comb filter.

In FIG. 9, (A) is a comb filter 60.
7 is a transfer characteristic diagram of FIG. 7, and FIG. 7B is a transfer characteristic diagram of the comb filter 608. The horizontal axis represents the order of each overtone when the fundamental frequency of the synthetic data is the first order, and the vertical axis represents the amplitude value.

FIG. 10 is an amplitude spectrum diagram of various data. In FIG. 10, (A) is an amplitude spectrum diagram of drive data, (B) is an amplitude spectrum diagram of waveform data S1, and (C) is an amplitude spectrum diagram of waveform data S2.
(D) is an amplitude spectrum diagram of the combined data.

The operation of the musical sound synthesizing apparatus configured as described above will be described. The characteristic of the flute instrument sound is that when the pitch is raised by pressing the octave key, etc., two vibration modes occur in the sound pressure in the pipe, and one octave from the fundamental frequency corresponding to the pitch of the flute sound. It means that a low pitch or a low pitch of 1 octave and 5 degrees below is superposed although it is a weak level. This phenomenon will be described with reference to FIG. When the sound is played in the low range without pressing the octave key, resonance occurs in the tube in the vibration mode indicated by the broken line. When the octave key is pressed to produce a sound in the range one octave higher, resonance occurs in the tube in the vibration mode indicated by the solid line. However, at this time, although weak,
The resonance of the vibration mode indicated by the broken line also occurs. With the sound pressure as a reference, the broken line vibration mode has antinode positions near both ends of the pipe, and the solid line vibration mode has antinode positions near both ends of the pipe and the center of the pipe. The broken line vibration mode forms a vibration having a fundamental frequency about one octave lower than the solid line vibration mode. And if it is raised one octave further, it becomes a vibration mode in which there are four belly positions including the positions of both ends of the tube. Will be formed. The comb filter 607 performs a resonance operation corresponding to the solid line vibration mode. That is, it has the transfer characteristic shown in FIG. Here, the frequency corresponding to the first-order component is f
When set to 1, the relationship of (Equation 2) is established.

[0044]

[Equation 2]

Now, as pitch data, 261.63H
A case will be described in which pitch data corresponding to a frequency of less than 523.25 Hz from z is set. Since this range corresponds to the lowest range in which the octave key is not pressed in the flute, only the vibration mode shown by the broken line in FIG. 8 occurs. At this time, the comb filter 608 also acts so as to form vibration having the same fundamental frequency as the comb filter 607.
That is, the number of delay stages n of the delay device 603 in the comb filter 608
Is equal to k. Next, as pitch data, 523.2
When the pitch data corresponding to the frequency from 5 Hz to less than 1046.50 Hz is set, it corresponds to the range where the octave key is pressed, so both the vibration mode shown by the solid line and the vibration mode shown by the broken line in FIG. 8 occur. Will be done. At this time, the comb filter 608 acts so as to form a vibration about one octave lower than the fundamental frequency of the comb filter 607. That is, the delay stage number n of the delay device 603 in the comb filter 608 takes a value twice as large as k. Similarly, in the case of a tone range that is one octave higher, the number of delay stages n of the delay device 603 in the comb filter 608 takes a value three times k. To simplify the explanation, considering the case where the scale within the octave is fixed, the delay time change table 606 is a table having the relationship of (Table 1).

[0046]

[Table 1]

In FIG. 7, since the write address is always preceded by n stages from the read address by the counter 702, the subtractor 703, and the selector 704, the delay device 603 acts as a delay device with the delay stage number n. Further, since the mixing circuit 106 sets the mixing level of S2 to a value sufficiently smaller than the value 1, the vibration level corresponding to the vibration mode indicated by the broken line in FIG.
As shown in (D), when the frequency f1 is a first-order overtone, amplitude levels other than the integer order can be made small.

As described above, according to the second embodiment of the present invention, the delay time change table 606 sets the delay stage number n of the delay unit 603 to k, 2k, according to the tone range (pitch data).
3 k is changed, and the mixing circuit 106 mixes the mixed data in a state where the mixing level of S2 is lower than the mixing level of S1 to synthesize the synthesized data. Therefore, as shown in FIG. It is possible to weakly superimpose a low resonance component one octave lower than f1 or one octave and five degrees lower, and to synthesize a timbre like a flute.

(Third Embodiment) A third embodiment of the present invention will be described below with reference to the drawings.

FIG. 11 is a block diagram of a musical sound synthesizing apparatus according to the third embodiment of the present invention. In FIG. 11, 1
101 is a delay device for delaying data for a predetermined time, 1102
Is a high-frequency cutoff filter (described as LPF in the figure) that cuts off high-frequency components of the data output from the delay device 1101.
Reference numeral 03 denotes an adder for adding drive data and data output from the high-frequency cutoff filter 1102, 1104 denotes a subtractor for subtracting the waveform data S2 from the waveform data S1, 1105.
Is a comb filter in which the delay device 1101, the high-frequency cutoff filter 1102, and the adder 1103 are configured in a closed loop.
The other circuits are the same as those of the musical tone synthesizers of the first and second embodiments.

FIG. 12 is a transfer characteristic diagram of the comb filter. In FIG. 12, (A) is a transfer characteristic diagram of the comb filter 607, and (B) is a transfer characteristic diagram of the comb filter 1105.

FIG. 13 is an amplitude spectrum diagram of various data. 13, (A) is an amplitude spectrum diagram of drive data, (B) is an amplitude spectrum diagram of waveform data S1, (C) is an amplitude spectrum diagram of waveform data S2,
(D) is an amplitude spectrum diagram of the combined data.

The operation of the musical tone synthesizer configured as above will be described. The characteristic of the sound of the sax is that the harmonics of the harmonics of the fundamental frequency in the vicinity of the 3rd and 4th harmonics become peaks, and the resonance level of the harmonics below that becomes small. The transfer characteristic of the comb filter 607 is the characteristic shown in FIG. 12A, as in the second embodiment. The transfer characteristic of the comb filter 1105 is shown in FIG. 12B by setting the high cutoff frequencies of the high cutoff filter 1102 to around the second and third orders in FIG. 12A as shown in FIG. It can have such characteristics. Comb filter 60
By inputting drive data having characteristics as shown in FIG.
Waveform data S1, S2 having characteristics as shown in (C)
Is obtained. By subtracting S1 and S2 in the subtractor 1104, synthetic data as shown in FIG. 13D is obtained. Note that the connection order of the delay device 1101 and the high frequency cutoff filter 1102 may be changed.

As described above, according to the third embodiment of the present invention, the comb filter 1105 has the first and second orders (the second order level is lower than the first order level) of the fundamental frequency of the synthesized data.
S2 is generated by resonating only the low frequency component, and the subtractor 1104 subtracts S2 from S1 so that the harmonic overtones with respect to the fundamental frequency are peaks at the harmonic overtones near the third and fourth harmonics. It is possible to obtain synthetic data having a characteristic that the resonance level becomes small.

(Fourth Embodiment) A fourth embodiment of the present invention will be described below with reference to the drawings.

FIG. 14 is a block diagram of a tone synthesis apparatus according to the fourth embodiment of the present invention. In FIG. 14, 1
Reference numeral 401 is a microphone that converts an acoustic signal such as voice into an analog electric signal, and 1402 is an analog-digital converter that converts the analog electric signal into digital data. The other circuits are similar to those of the musical sound synthesizer of the second embodiment.

The operation of the musical tone synthesizing apparatus configured as described above will be described. A so-called unvoiced sound such as a plosive sound or a fricative sound is input to the microphone 1401. The analog-digital converter 1402 converts the unvoiced analog electric signal into digital data in real time, and the comb filter 6
Enter in 07. The comb filter 607 generates synthetic data by performing a resonant operation at a fundamental frequency corresponding to the number of delay stages m. Note that the circuit of the comb filter 607 may be complicated to configure a resonance portion of a musical instrument having a more complicated vibration mode such as a drum membrane. Further, it is not necessary to limit to unvoiced sound, and a general acoustic signal may be input to the microphone 1401.

As described above, according to the fourth embodiment of the present invention, the microphone 1401 and the analog-digital converter 1402 convert unvoiced sound into digital data in real time and input it to the comb filter 607. It is possible to obtain a sound that resonates with a resonance part of a musical instrument such as a string, a tube, or a film. The voice has a formant peculiar to the speaker, and the characteristic of this formant remains in the synthetic data generated from the comb filter 607. That is, it is possible to obtain synthetic data that makes the best use of the characteristics of each individual's voice.

(Fifth Embodiment) A fifth embodiment of the present invention will be described below with reference to the drawings.

FIG. 15 is a block diagram of a musical sound synthesizer in the fifth embodiment of the present invention. In FIG. 15, 15
Reference numeral 01 is a non-linear table for performing non-linear conversion of input data, and 1502 is a delay device 601 and a non-linear conversion table 1.
This is a comb filter in which 501 and an adder 602 are configured in a closed loop. The other circuits are the same as those of the musical tone synthesizers of the first and second embodiments.

The operation of the musical sound synthesizing apparatus configured as described above will be described. The non-linear conversion table 1501 is
When the absolute value level of the input data is small, it acts as a linear converter that outputs the input data in a through state or by multiplying it by a certain scale factor, and when the absolute value level of the input data is large, the input data is continuously output. It acts as a non-linear converter (limiter) that clips to and outputs. When the non-linear conversion table 1501 does not exist, the theoretical maximum absolute value of the data circulating in the loop when the comb filter 1502 resonates to the maximum is set to 10, and the practical maximum absolute value (in the loop). (Due to the limitation of the bit word length of the circuit element of the above). When the non-linear conversion table 1501 does not exist, an overflow occurs at ± 8, so unpleasant noise occurs due to the discontinuity of data, but the non-linear conversion table 1501 exists.
Assuming that the conversion characteristics are as shown in (Table 2),
Since the absolute value is less than 8, discontinuous noise does not occur.

It is assumed that a value obtained by interpolating two data values is present between each data in (Table 2) (for example, between 0 and 1 of the input data).

[0063]

[Table 2]

Instead of the non-linear conversion table 1501, the non-linear conversion may be performed by using another calculation means or the like. Further, the connection order of the delay device 601 and the non-linear conversion table 1501 may be changed.

As described above, according to the fifth embodiment of the present invention, the nonlinear conversion table 1501 has the comb filter 1
When the absolute value level of the data circulating in 502 is small (value 6 to (-6)), the input data is directly output,
When the absolute value level of the input data is large (value 6
As described above, since the value (−6) or less) and the input data are continuously clipped and output, discontinuous noise can be removed.

[0066]

According to the present invention, the drive data generating means generates a waveform corresponding to the vibration generated when the breath is blown into the tube,
The first comb filter amplifies the component corresponding to the odd harmonics of the fundamental frequency of the synthesized data, and the second comb filter amplifies the component corresponding to the even harmonics to obtain the first and second waveform data. Then, the mixing means mixes the first and second waveform data to combine the combined data. At this time, the mixing ratio control means controls the mixing ratio of the second waveform data based on the tone color control parameter corresponding to the blowing angle, thereby changing the tone color according to the blowing angle to the song mouth (only the even harmonics Level control) can be realized.

Further, the drive data generating means generates a waveform corresponding to the vibration generated when the breath is blown into the tube, and the first comb filter amplifies the component corresponding to the integer harmonic of the fundamental frequency of the synthetic data. , The second comb filter amplifies a component corresponding to an integer harmonic of a frequency lower than the fundamental frequency of the synthetic data to generate the first and second waveform data, and the mixing means generates the first and second waveform data. The waveform data is mixed to synthesize the synthesized data. At this time, the delay time changing means changes the delay time of the delay device in the second comb filter to about 2 times or about 3 times the delay time of the delay device in the first comb filter, thereby changing the fundamental frequency. Therefore, a sound in which a weak resonance sound having a low frequency of one octave below or one octave and 5 degrees below is superposed is synthesized. That is, it is possible to realize a timbre change that sounds like a flute instrument sound in accordance with the musical range.

Further, the drive data generating means generates a waveform corresponding to the vibration of the lead, which is the nozzle of the tube, and the first comb filter amplifies the component corresponding to the integer harmonic of the fundamental frequency of the synthetic data. , The second comb filter amplifies low-order components up to the first and second orders of the integer harmonics of the fundamental frequency of the composite data, and the subtracting means extracts the second waveform data from the first waveform data. By subtracting, it is possible to faithfully simulate the characteristic of the sax tube in which the harmonics having harmonic orders near the 3rd and 4th harmonics peak and the resonance level of the harmonics below that peak becomes small.

Further, the microphone converts the voice into an analog electric signal, the analog-digital converter converts the analog electric signal into digital data, and the comb filter resonates the digital data. Can be sampled in real time and resonated by a circuit corresponding to the resonance part (string or tube) of the musical instrument. In other words, it is possible to obtain synthetic data that makes use of the characteristics (formants) of each individual's voice.

Further, the drive data generating means generates drive data, and the comb filter circulates the drive data to synthesize desired synthesized data. At this time, the non-linear conversion means in the comb filter performs data level conversion processing so that the circulating data does not exceed a predetermined level, so even if the comb filter resonance level is increased, it is uncomfortable due to data overflow. It is possible to combine the combined data without generating any noise.

[Brief description of drawings]

FIG. 1 is a block diagram of a musical sound synthesizer according to a first embodiment of the present invention.

FIG. 2 is a mixing circuit 106 of the musical sound synthesizer of the present invention.
Schematic of

FIG. 3 is a circuit diagram of a mixing ratio control circuit 107 of the musical sound synthesizer of the present invention.

FIG. 4 is a transfer characteristic diagram of a comb filter of the musical sound synthesizer in the first embodiment of the present invention.

FIG. 5 is an amplitude spectrum diagram of various data of the musical sound synthesizer in the first embodiment of the present invention.

FIG. 6 is a block diagram of a musical sound synthesizer in a second embodiment of the present invention.

FIG. 7 is a circuit diagram of a delay unit 603 of the musical sound synthesizer according to the second embodiment of the present invention.

FIG. 8 is a vertical sectional view of a flute tube.

FIG. 9 is a transfer characteristic diagram of a comb filter of the musical sound synthesizer in the second embodiment of the present invention.

FIG. 10 is an amplitude spectrum diagram of various data in the musical sound synthesizer according to the second embodiment of the present invention.

FIG. 11 is a block diagram of a musical sound synthesizer in a third embodiment of the present invention.

FIG. 12 is a transfer characteristic diagram of the comb filter of the musical sound synthesizer in the third embodiment of the present invention.

FIG. 13 is an amplitude spectrum diagram of various data in the musical sound synthesizer according to the third embodiment of the present invention.

FIG. 14 is a block diagram of a musical sound synthesizer in a fourth embodiment of the present invention.

FIG. 15 is a block diagram of a musical sound synthesizer in a fifth embodiment of the present invention.

FIG. 16 is a block diagram of a musical sound synthesizer in a conventional example.

[Explanation of symbols]

101 Drive Data Generation Circuit 103, 703, 1104 Subtractor 105, 202, 602, 604, 1103, 1603
Adder 106 Mixing circuit 107 Mixing ratio control circuit 108, 109, 607, 608, 1105, 150
2, 1604 Comb filter 201 Multiplier 606 Delay time change table 1102 High-frequency cutoff filter 1401 Microphone 1402 Analog-digital converter

Front page continued (72) Inventor Atsuko Tanaka 1006 Kadoma, Kadoma City, Osaka Prefecture Matsushita Electric Industrial Co., Ltd. (56) Reference JP-A-6-95678 (JP, A) (58) Fields investigated (Int .Cl. ⁷ , DB name) G10H 1/00-7/00

Claims (5)

(57) [Claims] 1. A drive data generating means for reading prestored drive data, and a circuit in which a delay device is formed in a closed loop so that the drive data is circulated and the data being circulated is taken out as first waveform data. A first comb filter, a second comb filter in which the drive data is circulated by a circuit in which a delay device is configured in a closed loop shape, and the data being circulated is taken out as second waveform data; comprising a mixing means for generating a desired combined data is performed a mixing of the waveform data the second waveform data, and a mixing ratio control means for controlling the mixing ratio of said mixing means, said first comb filter is the A comb with the property of forming resonance peaks in the odd harmonics of the fundamental frequency of the desired composite data.
A second comb filter for forming resonance peaks at even harmonics of the fundamental frequency of the desired synthesized data .
A musical sound synthesizer characterized by being a comb filter having the following characteristics. 2. The first comb filter includes a delay device and a subtractor, and the data output from the delay device is subtracted from drive data and the subtraction result is input to the delay device in a closed loop. The second comb filter includes a delay device and an adder, and adds the drive data and the data output from the delay device and inputs the addition result to the delay device. 2. The musical tone synthesizer according to claim 1, wherein the closed loop is formed in various connection forms. 3. A drive data generating means for reading out prestored drive data, and a circuit in which a delay device is configured in a closed loop so that the drive data is circulated and the data being circulated is taken out as first waveform data. A first comb filter, a second comb filter in which the drive data is circulated by a circuit in which a delay device is configured in a closed loop shape, and the data being circulated is taken out as second waveform data; Mixing means for mixing the waveform data and the second waveform data to generate desired combined data, and a delay for changing the delay time of the delay device in the second comb filter according to the fundamental frequency of the combined data. A comb having a characteristic that the first comb filter forms a resonance peak at an integer harmonic of the fundamental frequency of the desired combined data.
A second comb filter , wherein the second comb filter is an integer harmonic of a frequency lower than the fundamental frequency of the desired synthesized data.
A musical tone synthesizer characterized in that it is a comb filter having the characteristic of forming a resonance peak at the bottom. 4. A connection form in which the first comb filter includes a delay device and an adder, and adds drive data and data output from the delay device and inputs the addition result to the delay device. The second comb filter includes a delay device and an adder, adds the drive data and the data output from the delay device, and inputs the addition result to the delay device. 4. The musical tone synthesizer according to claim 3, wherein a closed loop is formed in such a connection form. 5. The delay time changing means is an integer other than 1 such that the delay time of the delay device in the second comb filter is about twice or three times the delay time of the delay device in the first comb filter. 4. The musical tone synthesizing apparatus according to claim 3, wherein the musical tone synthesizing apparatus controls the musical tone synthesizing ratio so that