JP3926445B2 - Music synthesizer - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a musical sound synthesizer, and more particularly to a musical sound synthesizer suitable for generating nonharmonic musical sounds such as piano sounds and guitar sounds.
[0002]
[Prior art]
In general, a musical tone waveform is formed by superposing a plurality of sine waves. These sine waves are called harmonics because their frequencies are in an integral multiple of each other. In particular, the overtone with the lowest frequency is called a fundamental wave or a fundamental tone, and the frequency is called a fundamental frequency. The frequency of the other harmonics is an integral multiple of the fundamental frequency. The relative level relationship between overtones is called the overtone structure, and the tone of the musical tone is determined by the difference in overtone structure.
[0003]
In an electronic musical instrument, it is known to synthesize a musical sound by generating a plurality of sine waves at a harmonic frequency and superimposing them. This is a musical sound synthesis method called a sine wave addition method. In this method, the fundamental frequency for the pitch is stored in a memory as a frequency table, and the fundamental frequency is read from this table. Then, the frequency of each harmonic other than the fundamental frequency is obtained by multiplying the fundamental frequency by an integer, and these are synthesized to obtain a desired musical sound. Since the fundamental frequency for the pitch is common to all musical sounds, the memory capacity for storing the frequency table can be reduced. According to this sine wave addition method, a sound having the same harmonic structure as that of a natural musical sound can be generated, so that a musical sound close to a natural musical instrument can be generated.
[0004]
Usually, the fundamental frequency of each pitch is determined by equal temperament. In the equal temperament, the fundamental frequency of each pitch of one octave is determined so that the frequency ratio of adjacent sounds of twelve pitches in one octave is constant. That is, one octave is divided into 12 equal pitches, and one pitch is a semitone.
[0005]
In the equal temperament, when the pitch increases by one octave, its fundamental frequency doubles. In the case of the sine wave addition method, when the frequency of a harmonic is obtained as an integral multiple of the fundamental frequency, the frequency of the second harmonic of a certain pitch matches the fundamental frequency of the pitch one octave higher than that. That is, the frequency of an even overtone of a certain pitch matches the frequency of a harmonic overtone of one octave above.
[0006]
By the way, some natural musical instruments, for example, musical sounds such as pianos and guitars, have a frequency that is slightly shifted rather than an integral multiple. In these musical tones, the frequency of overtones is slightly higher than the integral multiple of the fundamental frequency. This shift increases as the harmonic order increases. This type of musical tone is called anharmonicity. The degree of anharmonicity varies depending on the musical instrument, and the same musical instrument differs between bass and treble.
[0007]
As an example, the inharmonicity of the piano sound will be explained. The anharmonicity can be expressed by a deviation δn (cent value) of the frequency fn of the n-th overtone from an integral multiple of the fundamental frequency f0 (the fundamental frequency of an ideal non-elastic string). Sought by.
[0008]
.delta.n = 1200 log2 (fn / nf0) (1)
FIG. 10 shows the anharmonicity of the piano sound obtained according to the equation (1). In the figure, the inharmonicity of bass, medium and treble is shown by curves L, M, and H, corresponding to harmonic orders. Anharmonicity is indicated by a cent. As shown in this figure, it can be seen that the higher the pitch, the greater the deviation from the integer multiple frequency. An example of piano sound inharmonicity is described in pages 9 to 15 of the Journal of the Institute of Electronics, Information and Communication Engineers, “Science Technical Report EA93-28 (1993-07)”.
[0009]
In view of such anharmonicity, the present inventor calculates a harmonic frequency from a preset anharmonic coefficient and a fundamental frequency, and generates a musical tone having anharmonicity by a sine wave addition method. An apparatus has already been proposed (Japanese Patent Application No. 9-70908: reference number 8522KWA).
[0010]
[Problems to be solved by the invention]
The tone synthesizer using the sine wave addition method has the following problems. First, in an inharmonic musical tone, if the frequency of each pitch is equal temperament, the frequency of an even overtone of a certain pitch is the frequency of the fundamental tone or harmonic overtone that is one octave higher than that. There is a problem that beats occur because of the slight deviation.
[0011]
Secondly, it is known that human ears tend to have a tendency that sounds below 100 Hz can be heard higher than actual sounds, and sounds above 1000 Hz can be heard lower than actual sounds. 85, “Kenji Kitamura, published by Uzikanova”). Therefore, in the acoustic piano, the low frequency range is lower and the high frequency range is higher and the fundamental frequency is adjusted.
[0012]
The fundamental frequency adjusted in this way and representing the deviation from the original frequency of the average temperament is called a tune curve, and an acoustic piano has a natural pitch. Therefore, it has been desired that a musical tone synthesizer that generates an anharmonic musical tone by a sine wave addition method can generate a musical tone having a frequency suitable for the characteristics of the human ear.
[0013]
A musical tone synthesizer has been proposed in which a tuning method implemented based on the frequency relationship of frequency components between octaves is realized by a delay control device equipped with an all-pass filter (Japanese Patent Laid-Open No. 8-22285). In the tone synthesizer described in this publication, as shown in FIG. 11, the delay time control unit 100 supplies the delay time Dn (p) corresponding to the key code KC to the delay circuit 110, and the excitation waveform generation unit. The excitation waveform output from 120 is circulated in a delay feedback loop including a low-pass filter 140 to synthesize a desired attenuated sound. In the delay time control unit 100, a data table for outputting the delay time Dn (p) using the key code KC as a read address is generated for each tone color.
[0014]
In this musical tone synthesizer, the pitch of the fundamental tone corresponding to the key code KC is uniquely determined from the delay length of the all-pass filter 130 that generates an unharmonic musical tone. That is, the tuning curve is uniquely determined corresponding to the inharmonicity of the generated musical sound. Therefore, it can be tuned so as not to cause the beat based on the inharmonicity, but according to the nature of the human ear, the pitch of the fundamental tone of each key is tuned to the extent that the beat is not noticeable. I can't. In other words, there is a problem that the degree of freedom of tuning is limited and a more natural pitch cannot be obtained.
[0015]
An object of the present invention is to provide a musical tone synthesizer that solves the above-mentioned problems, does not generate a beat due to anharmonicity, and enables tuning with a higher degree of freedom that matches the characteristics of the human ear. And
[0016]
[Means for Solving the Problems]
In order to solve the above-mentioned problems and achieve the object, the present invention provides an anharmonic coefficient generation that uses musical pitch information as an input and outputs an anharmonic coefficient that is set in advance corresponding to the pitch information. Means, a tone value generating means for receiving the pitch information of the musical tone as input, and outputting a deviation from a preset fundamental frequency corresponding to the pitch information as a tuning value, and output from the tuning value generating means A harmonic frequency generating means for calculating each harmonic frequency using the fundamental frequency tuned with a different tuning value and the anharmonic coefficient output from the anharmonic coefficient generating means, and generating a musical sound based on the harmonic frequency There is a first feature in that it includes overtone generating means.
[0017]
According to the present invention, in addition to the first feature, an anharmonic touch coefficient generating unit that receives touch information as an input and outputs an anharmonic touch coefficient set in advance corresponding to the touch information, and the anharmonic coefficient And a multiplier for multiplying the non-harmonic touch coefficient by the non-harmonic touch coefficient. The second characteristic is that the non-harmonic coefficient multiplied by the non-harmonic touch coefficient by the multiplier is supplied to the harmonic frequency generating means. There is.
[0018]
Also, the present invention provides an anharmonic coefficient table selection unit, wherein the anharmonic coefficient generating unit includes a plurality of anharmonic coefficient tables set for each timbre, and selects one of the anharmonic coefficient tables based on timbre information. And the tuning value generating means includes tuning curve table selection means for selecting one of the tuning curve tables based on tone color information, including a tuning curve table set for each tone color. There is a third feature.
[0019]
Further, according to the present invention, the anharmonic touch coefficient generation means includes a plurality of anharmonic touch coefficient tables set for each timbre or tone range, and one of the anharmonic touch coefficient tables is obtained based on the tone color or tone range information. A fourth feature is that it further includes a non-harmonic touch coefficient table selection means for selection.
[0020]
According to the first to fourth features, a musical tone is generated based on each harmonic frequency calculated using a fundamental frequency tuned by a tuning value and a preharmonic coefficient set in advance corresponding to the pitch of the musical tone. Is done. In particular, according to the third feature, the fundamental frequency and the anharmonic coefficient tuned by the tune value corresponding to the tone color of the generated musical tone are used for the calculation of each harmonic frequency. Further, according to the second and fourth features, the anharmonic coefficient is corrected by the anharmonic touch coefficient selected based on the touch information. In particular, according to the fourth feature, the nonharmonic touch coefficient to be used can be selected based on the tone color and tone range of the generated musical sound.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, the present invention will be described in detail with reference to the drawings. FIG. 2 is a block diagram showing a configuration of an electronic musical instrument including a musical tone synthesizer according to an embodiment of the present invention. In the figure, a keyboard unit 1 has a plurality of keys, and the key press and release states of the plurality of keys are detected by a plurality of sensors 2 provided corresponding to the keys. The sensor 2 is connected to the bus 3, and key codes and touch information detected by the sensor 2 are input to the CPU 4 via the bus 3. The CPU 4 is connected to an operation panel 5 having various switches such as a power switch, a tone color selection switch, and a volume switch, and a pedal 6 for giving a damper effect and a soft effect. Furthermore, an external device can be connected to the CPU 4 via the MIDI interface 7. The ROM 8 is used as a program memory 8a for storing programs executed by the CPU 4 and a timbre data memory 8b for storing timbre data. The RAM 9 is provided for storing change data such as registers and flags.
[0022]
The CPU 4 performs processing using programs and data stored in the ROM 8 and RAM 9 in accordance with signals input from the keyboard 1, the operation panel 5, and the pedal 6, and supplies an instruction based on the processing result to the tone generator 10. . The tone generator 10 synthesizes a tone signal based on an instruction from the CPU 4, and the tone signal is converted into analog data by the D / A converter 11 and then supplied to the sound system 12. Music is generated.
[0023]
Next, main functions of the musical sound generating apparatus according to the present embodiment will be described with reference to a block diagram. 1, an anharmonic table group 13 as an anharmonic coefficient generating means set in the ROM 8 stores anharmonic tables 13-1, 13-2,... 13-N for each tone color. In the anharmonic tables 13-1 to 13-N, as shown in an example in FIG. 3, a correspondence relationship between the basic anharmonic coefficient MK and a keyboard number, that is, pitch information is set. The basic anharmonic coefficient MK is corrected by an anharmonic touch coefficient TK, which will be described later, to become an anharmonic coefficient K, which is used for calculating the harmonic frequency.
[0024]
The nonharmonic table selection information generating unit 14 generates selection information for selecting one from the plurality of nonharmonic tables based on the timbre information input from the operation panel 5. The anharmonic table selection information generating unit 14 has a function of converting timbre information into an address of the anharmonic table. The pitch information by the sensor 2 is input as an address to the anharmonic table selected by the selection information, and the basic anharmonic coefficient MK corresponding to the address is read out.
[0025]
On the other hand, in the non-harmonic touch coefficient table 15, as shown in an example in FIG. 4, a correspondence relationship between the non-harmonic touch coefficient TK and touch velocity as touch information is set. The touch velocity is calculated by the CPU 4 based on the detection signal of the sensor 2. Since the configuration of the sensor 2 for detecting touch velocity and the method for calculating the touch velocity are well-known techniques, a detailed description thereof will be omitted. The anharmonic touch coefficient table 15 is stored in the ROM 8 and outputs an anharmonic touch coefficient TK according to the touch velocity.
[0026]
The basic anharmonic coefficient MK and the anharmonic touch coefficient TK are multiplied by the multiplication unit 16 to obtain an anharmonic coefficient K modified according to the touch velocity. The anharmonic coefficient K is input to the harmonic frequency generator 17. The overtone frequency generation unit 17 calculates the frequency of each overtone based on the fundamental frequency f 0 of the musical tone to be generated and the anharmonic coefficient K, and sends it to the overtone generation unit 18. The overtone generator 18 generates overtones of the input frequency.
[0027]
The frequency of each overtone is calculated according to the following equation (2).
fn = nf0 (1 + Kn ² ) ^1/2 (2)
In order to simplify the calculation of Expression (2) and reduce the scale of the arithmetic circuit, the following approximate expression (3) can be used.
[0028]
fn = nf0 + Kan ³ (3)
The anharmonic coefficient Ka is for the approximate expression (3), and is different from the anharmonic coefficient K in the expression (2).
[0029]
Tuning curve tables 19-1, 19-2,... 19-N for each tone color are stored in the tuning curve table group 19 as a tuning value generating means for outputting a cent value indicating a tuning value for tuning the fundamental frequency. Has been. In the tuning curve tables 19-1 to 19-N, as shown in an example in FIG. 5, a correspondence relationship between a cent value indicating a tuning value and pitch information (keyboard number) is set. FIG. 5 illustrates tuning curves corresponding to three kinds of timbres. Here, 100 cents = 1.059463 (true number).
[0030]
Returning to FIG. 1 again, the tuning curve table selection information generating unit 20 generates selection information for selecting one from the plurality of tuning curve tables based on the timbre information input from the operation panel 5. The tuning curve table selection information generating unit 20 has a function of converting timbre information into an address of the tuning curve table. In the tuning curve table selected by the selection information, pitch information from the sensor 2 is input as an address, and a cent value corresponding to the address is read out.
[0031]
The fundamental frequency generator 21 stores a fundamental frequency for each pitch. When pitch information is input, a fundamental frequency f0 'corresponding to the pitch is output. The multiplier 22 multiplies the fundamental frequency f0 ′ by the cent value read from the tuning curve table, and outputs the fundamental frequency f0 after tuning. The fundamental frequency f0 is input to the harmonic frequency generator 17 to calculate the frequency of each harmonic.
[0032]
Next, the operation of the musical tone synthesizer based on the above function will be described with reference to a flowchart. FIG. 6 is a flowchart of the main routine. After the power is turned on, in step S1, the CPU 4, the RAM 9, and the LSI constituting the musical tone generator 10 are initialized. In step S2, step S3, and step S4, panel event processing, pedal event processing, and keyboard event processing are performed, respectively. In step S5, other processing is performed. Panel event processing and keyboard event processing will be described later, but description of pedal event processing and other processing not directly related to the present invention will be omitted.
[0033]
First, panel event processing will be described with reference to the flowchart of FIG. In the figure, in step S20, it is determined whether or not the timbre selection switch is turned on. If the timbre selection switch is turned on, the process proceeds to step S21 to perform timbre selection processing. In the tone color selection process, a flag indicating the selected tone color is set, or a display lamp (LED or the like) corresponding to the selected tone color is turned on. A display lamp can be provided on the operation panel 5.
[0034]
In step S22, one of the unharmonic tables 13-1 to 13-N is selected based on the timbre information input from the timbre selection switch. In step S23, one of the tuning curve tables 19-1 to 19-N is selected based on the timbre information input from the timbre selection switch.
[0035]
If it is determined in step S20 that the tone color selection switch has not been turned on, the process proceeds to step S24 to determine whether or not the volume switch has been turned on. If the volume switch is turned on, the process proceeds to step S25 to perform a volume setting process. In the volume setting process, the volume is set according to the ON operation of the volume switch. For example, a volume change amount by a single ON operation of the volume switch is set in advance, and the set volume is changed by this volume change amount.
[0036]
Further, if it is determined in step S24 that the volume switch is not turned on, the process proceeds to step S26, where it is determined whether other switches are turned on. If the result is affirmative, in step S27. The schedule processing corresponding to other switches is performed. If all the switches have not been turned on, the process of this flowchart is finished as it is, and the process returns to the main routine.
[0037]
Next, keyboard event processing will be described with reference to the flowchart of FIG. In the figure, in step S40, it is determined whether or not the key of the keyboard 1 is turned on, that is, whether it is an on event. If it is an on event, the process proceeds to step S41, and overtone frequency calculation processing is performed. Details of this processing will be described later with reference to FIG. In step S42, musical tone parameter calculation processing is performed. Here, various parameters of musical tone such as envelope and decay time are calculated according to the pitch (key code) and keying strength (velocity) of the pressed key, and loaded into the tone generator LSI of the overtone generator 18. In step S43, the overtone generator 18 performs a sound generation process. Here, a sine wave is synthesized according to each harmonic frequency and an envelope is added to generate a musical sound.
[0038]
On the other hand, if it is not an on event, the process proceeds from step S40 to step S44, and it is determined whether the key is released, that is, an off event. If it is determined that the event is an off event, the process proceeds to step S45, and it is determined whether or not the damper pedal is turned on. If the damper pedal is on, the sound is not muted and the process returns to the main routine. If the damper pedal is not turned on, the process proceeds from step S45 to step S46, and in order to end the sound generation, a parameter for determining the scheduled release speed, that is, the time until the mute after key release, is set as the tone generator LSI of the overtone generator 18. To finish the pronunciation. If it is not an off event, the process skips steps S45 and S46 and returns to the main routine.
[0039]
FIG. 9 is a detailed flowchart of overtone frequency calculation processing. In the figure, in step S50, the fundamental frequency f0 'output from the fundamental frequency generator 21 is multiplied by the cent value read from the tuning curve table to obtain the fundamental frequency f0. In step S51, the key code that has been turned on, that is, the basic anharmonic coefficient MK corresponding to the pitch information is read. Here, the inharmonicity table selected in the panel event (step S22) is used. In step S52, the anharmonic touch coefficient TK is read from the anharmonic touch coefficient table 15 based on the pitch information. In step S53, the modified anharmonic coefficient K or Ka is calculated by multiplying the anharmonic coefficient MK by the anharmonic touch coefficient TK. In step S54, the harmonic frequency fn is calculated by the equation (2) or the equation (3) using the anharmonic coefficient K or Ka.
[0040]
Next, an example of creating the tuning curve table will be described. In the central one-octave region “C3” to “B3”, first, the tone above “C3”, that is, the fundamental frequency of “C4” is set as a tuning value so that it matches the second harmonic frequency of “C3”. Determine the cent value. The secondary harmonic frequency here is a value including anharmonicity represented by the anharmonic coefficient K. Next, the tuning value for one octave is equally divided into 12 scales, and the tuning values for “C3” to “B3” are determined.
[0041]
Similarly, for the “C2” sound that is one octave lower than the “C3” sound, the tuning value is determined so that the secondary harmonic frequency thereof matches the fundamental frequency of the “C3” sound. This is equally divided into 12 scales, and the tuning values for “C2” to “B2” are determined. Similarly, for other octaves, the tuning value of each scale is determined in relation to the adjacent octaves.
[0042]
By determining the rhythm value in this way, it is possible to eliminate beats due to inharmonicity, but furthermore, in accordance with the nature of the human ear that it sounds lower than it actually is in the high range and higher than it actually is in the low range, The tuning value can be determined so as to correct this.
[0043]
In the above-described embodiment, one type of non-harmonic touch coefficient table is used. However, this table may have a plurality of types of non-harmonic touch coefficient tables corresponding to timbres as in the case of the non-harmonic table group 13. Good. In this case, an anharmonic touch coefficient table selection information generating unit for selecting one of the anharmonic touch coefficient tables based on the timbre information is provided in the same manner as in the above example.
[0044]
Furthermore, you may provide a nonharmonic touch coefficient table for every sound range. For example, three types of anharmonic touch coefficient tables of low, middle, and high frequencies are stored in the ROM, the range is identified from the key code, the table corresponding to the range is selected, and the anharmonic touch coefficient TK is read. If it can be output, it is possible to generate more precise musical sounds for each range. Further, although the harmonic overtone generator 18 generates a musical tone by the sine wave addition method, the present invention is not limited to this method, and any other method can be used as long as the frequency can be controlled for each overtone.
[0045]
【The invention's effect】
As is apparent from the above description, according to the present invention, an inharmonicity can be imparted using a fundamental frequency common to all timbres and an anharmonicity coefficient corresponding to the timbre. Furthermore, the fundamental frequency of the fundamental tone can be tuned so as not to cause a beat due to anharmonicity, and can be tuned to match the properties of the human ear.
[0046]
In particular, in the present invention, since the anharmonic coefficient generating means and the rhythm value generating means are provided independently, the degree of tuning obtained by the rhythm value generating means is determined regardless of the anharmonicity obtained by the anharmonic coefficient generating means. it can. Therefore, not only is it possible to prevent beats, but there is also an effect that the degree of freedom for generating musical sounds that can be naturally felt by the human ear to the extent that beats are inconspicuous is high.
[Brief description of the drawings]
FIG. 1 is a block diagram showing main functions of a musical tone synthesis apparatus according to an embodiment of the present invention.
FIG. 2 is a block diagram showing a hardware configuration of the tone synthesizer according to the embodiment of the present invention.
FIG. 3 is a diagram illustrating an example of an anharmonic coefficient table.
FIG. 4 is a diagram illustrating an example of an anharmonic touch coefficient table.
FIG. 5 is a diagram illustrating an example of a tuning curve table.
FIG. 6 is a main flowchart of the musical tone synthesizer.
FIG. 7 is a flowchart of panel event processing.
FIG. 8 is a flowchart of keyboard event processing.
FIG. 9 is a flowchart of overtone frequency calculation processing.
FIG. 10 is a diagram illustrating an example of anharmonicity of a piano sound corresponding to a harmonic order.
FIG. 11 is a block diagram showing a configuration of a musical tone synthesizer according to the prior art.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Keyboard part, 2 ... Sensor, 4 ... CPU, 5 ... Panel, 13 ... Anharmonic table group, 14 ... Anharmonic table selection information generation part, 15 ... Anharmonic touch coefficient table, 16 ... Multiplication part, 17 ... Overtone Frequency generation unit, 18 ... harmonic overtone generation unit, 19 ... tuning curve table group, 20 ... tuning curve table selection information generation unit

Claims (4)

An anharmonic coefficient generating means for inputting the pitch information of the musical sound and outputting an anharmonic coefficient set in advance corresponding to the pitch information;
Rhythm value generating means for inputting pitch information of a musical tone and outputting a deviation from a fundamental frequency set in advance corresponding to the pitch information as a rhythm value;
Harmonic frequency generating means for calculating each harmonic frequency using the frequency of the fundamental tone tuned with the rhythm value output from the rhythm value generating means and the anharmonic coefficient output from the harmonic coefficient generating means;
Overtone generating means for generating a musical sound based on each harmonic frequency;
An anharmonic touch coefficient generating means for receiving touch information as an input and outputting a preset anharmonic touch coefficient corresponding to the touch information;
Multiplying means for multiplying the anharmonic coefficient by the anharmonic touch coefficient,
The tuning value is determined so that the fundamental frequency of a predetermined pitch matches the second harmonic frequency calculated using the anharmonic coefficient that is one octave lower than the predetermined pitch. ,
An apparatus for synthesizing a musical sound, characterized in that an anharmonic coefficient obtained by multiplying the anharmonic touch coefficient by the multiplication means is supplied to the harmonic frequency generating means . The anharmonic coefficient generating means includes a plurality of anharmonic coefficient tables set for each timbre,
Comprising an anharmonic coefficient table selecting means for selecting one of the anharmonic coefficient tables based on timbre information;
The tuning value generating means includes a tuning curve table set for each timbre,
2. A musical tone synthesizing apparatus according to claim 1, further comprising tuning curve table selection means for selecting one of the tuning curve tables based on tone color information. The anharmonic touch coefficient generation means includes a plurality of anharmonic touch coefficient tables set for each timbre,
3. The musical tone synthesizing apparatus according to claim 1, further comprising non-harmonic touch coefficient table selection means for selecting one of the non-harmonic touch coefficient tables based on timbre information. The anharmonic touch coefficient generating means includes a plurality of anharmonic touch coefficient tables set for each sound range,
3. The musical tone synthesizing apparatus according to claim 1, further comprising an anharmonic touch coefficient table selection unit that selects one of the anharmonic touch coefficient tables based on the range information.

Images