JP3551992B2 - Electronic musical instrument - Google Patents

Description

[0001]
[Industrial applications]
The present invention relates to an electronic musical instrument, and more particularly to an electronic musical instrument provided with a waveform readout circuit capable of reducing a memory capacity.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, in an electronic musical instrument such as an electronic piano using a tone generation circuit of a waveform reading method, a waveform corresponding to a specific pitch is recorded by PCM for each tone, and processed waveform data is stored in a waveform memory. The waveform is read out at an address interval proportional to the pitch to be sounded, and the address interval data is called a frequency number. Conventionally, in order to convert a key number corresponding to each key of a semitone unit of a keyboard into a frequency number, a conversion table that stores a frequency number corresponding to each key number has been used. However, this method has a problem that it is necessary to store a large number of frequency numbers corresponding to the number of keys. To solve this problem, for example, as described in JP-A-2-139599. Technologies have been proposed. This method focuses on the fact that the frequency number doubles each time the octave increases, converts the key number into octave information and note (note name) information, converts the note information into a frequency number, and responds to the octave information. Thus, a desired frequency number is obtained by bit shifting.
[0003]
[Problems to be solved by the invention]
In the conventional frequency number conversion method as described above, when obtaining a frequency number in semitone units, it is sufficient to store 12 frequency numbers in one octave. In order to cope with the frequency change, it is necessary to perform control in units of cents in which a semitone is divided into, for example, 100. If the frequency numbers corresponding to all cents are stored, the storage capacity becomes extremely large. Was.
SUMMARY OF THE INVENTION It is an object of the present invention to provide an electronic musical instrument having a tone generating means capable of coping with a fine frequency change with a small storage capacity by improving the above-mentioned problems of the prior art.
[0004]
[Means for Solving the Problems]
According to the present invention, first conversion means for inputting upper digit information of pitch information and performing exponential conversion,
The lower digit information excluding the upper digit information of the pitch information is input and multiplied by the second converting means for performing exponential conversion, the output information of the first converting means and the output information of the second converting means. Multiplication means for calculating a frequency number for reading a waveform from the waveform memory, wherein at least one of the first conversion means and the second conversion means performs exponential conversion of information other than the pitch information. It is also used as conversion means for performing the conversion.
[0005]
[Action]
According to the present invention, with the above-described configuration, the frequency number conversion means only needs to convert, for example, only 12 frequency numbers corresponding to the integer part of the key number, and the exponent conversion means has, for example, two digits after the decimal point (centennial). It is only necessary to convert the data into 100 or 128 exponent data corresponding to up to 7 bits after the decimal point in binary or binary. Therefore, when the conversion means is constituted by a table, the storage capacity of the table is smaller than that of the conventional example. Very small. Further, by using the conversion means in a time-sharing manner, it can be used for other purposes such as calculation for generating an envelope signal.
[0006]
【Example】
Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. FIG. 2 is a block diagram showing an example of a circuit configuration of an electronic musical instrument on which the present invention is based . The CPU 1 is a central processing unit that controls the entire electronic musical instrument based on a control program stored in the ROM 2. The CPU 1 also has a built-in timer circuit that issues a timer interrupt at a set predetermined cycle. The ROM 2 stores programs, timbre parameters, and the like. The RAM 3 is used as a work area or a data buffer, and may be backed up by a battery. The keyboard 4 is composed of, for example, a plurality of keys each having two switches and a scan circuit, and the panel 5 is composed of various switches, a liquid crystal display, a display device for displaying characters and figures by means of LEDs or the like, and an interface circuit therefor.
[0007]
As will be described later, the musical tone generating circuit 6 sequentially reads out the musical tone waveforms from the waveform memory in which the digital musical tone waveform information is stored at address intervals proportional to the pitch to be generated, and multiplies the envelope waveform signal to generate the musical tone signal. It has a plurality of tone generating channels to be generated, and in fact, it is configured so that a plurality of tone signals can be independently generated at the same time by time-division multiplexing one tone generating circuit. Further, it may include a sound image effect circuit, a reverberation effect circuit, and the like. The D / A converter 7 performs D / A conversion of the digital tone signal, and the analog signal processing circuit 8 performs a simple filtering process for removing noise. The amplifier 9 amplifies the tone signal to drive the speaker 10. The bus 11 connects each circuit in the electronic musical instrument. In addition, a MIDI interface circuit, a floppy disk drive circuit, a memory card interface circuit, and the like may be provided.
[0008]
FIG. 3 is a block diagram showing the configuration of the tone generator 6 shown in FIG. The waveform reading circuit 20 receives from the CPU 1 a key number signal including information after the decimal point, a signal indicating key-on or key-off, and timbre information, and generates a read address for the waveform memory 21 by a configuration described later. Further, the data of the decimal part of the read address is output to the interpolation circuit 22. The waveform memory 21 stores waveform data obtained by performing PCM recording of a waveform corresponding to a specific pitch (for example, C7) for each timbre, and processing the envelope or the waveform of the repeatedly read portion.
[0009]
The interpolation circuit 22 inputs the waveform data corresponding to the integer part of the read address and the waveform data corresponding to the address obtained by adding 1 to the read address from the waveform memory 21, and outputs the waveform data from the waveform read circuit 20. A well-known interpolation operation is performed based on the decimal part data of the read address. The envelope generator 24 generates an amplitude envelope signal by an arbitrary known method according to various information or envelope parameters given from the CPU 1. The multiplier 23 multiplies the waveform signal output from the interpolation circuit 22 by an amplitude envelope signal, and outputs a tone signal. The tone generation circuit 20 performs time-division multiplexing operation and can independently generate a plurality of tone signals. The tone signals of the respective tone generation channels are added by an accumulator (not shown) provided after the multiplier 23. The signals are combined and output to the D / A converter 7.
[0010]
FIG. 1 is a block diagram showing a configuration of the waveform readout circuit 20 of FIG. In the waveform reading circuit 20, a process of converting a key number signal into a frequency number is performed. Therefore, the conversion principle will be described first. Assuming that X = (K−KS) / 12, the key number K and the frequency number F have a relationship of F = (2 × X). Here, KS is the reference key number of the waveform stored in the waveform memory 21 (for example, 96 = C7). Here, the key number K can be expressed as K = 12 * oct + note + Frac. Here, oct is an integer part of a quotient obtained by dividing K by 12, note is a remaining integer part (0 ≦ note ≦ 11), and Frac is a remaining decimal part (0 ≦ Frac <1). Here, if Ks = 96, X is expressed as X = (oct−96 / 12) + note / 12 + Frac / 12, and F is expressed as F = (2 × (oct−8) power) × (2 × 2 (note / 12 power) × (2 Frac / 12 power). The operations of the second and third terms on the right side of this equation are executed by conversion processing by the frequency number table 32 and the exponent conversion table 30, respectively, and the operation of the first term is executed by the converter 31 and the shift circuit 33.
[0011]
The converter 31 is a division circuit or a conversion table that divides the integer part keyno of the input key number signal by 12, and outputs the integer part oct and the remainder of the quotient. The frequency number table 32 is a conversion table storing 12 frequency numbers F corresponding to note numbers, as shown in FIG. 4A. The shift circuit 33 receives the frequency number F and outputs a value obtained by shifting the right bit by a value obtained by subtracting 8 (= Ks / 12) from oct. For example, when 56 (A3) is input as the key number signal, oct becomes 4 and note becomes 9. Therefore, an F value of 1.6817... Is output from the frequency number table 32, and is shifted right by 4-8 = −4 bits, that is, left shifted by 4 bits (divided by 2 to the fourth power) by the shift circuit 33, and becomes 0. . 1051... Are output.
[0012]
The decimal part Frac of the key number signal is represented by, for example, a 7-bit binary number, and the exponent conversion table 30 outputs [2 to the power of Frac / (12 * 128)] using the 7-bit binary number as an address. FIG. 4B is an explanatory diagram showing the storage contents of the exponent conversion table 30. In this figure, Frac is represented as a 7-bit integer value representing an address. The center cent indicates the corresponding value when Frac is expressed as a percentage, for example, a Frac value of 3 (0000011) corresponds to 2.34 cents, or 0.0234. The multiplier 34 multiplies the output of the shift circuit 33 by the output value of the exponent conversion table 30 and outputs and holds a frequency number value F corresponding to the input key number signal.
[0013]
The accumulator 35 includes an adder 36, a holding circuit 37, an AND circuit 38, and the like. The accumulator 35 accumulates the input frequency number value for each sampling cycle and outputs the lower bits of the read address for the waveform memory 21. The edge detector 39 normally outputs 1, and when detecting the rising edge of the key-on signal (change point from 0 to 1), the output is set to 0 for only one cycle. The CPU 1 sets the key number signal in the tone generating circuit 6 and activates the key-on signal after the frequency number value is input from the multiplier 34 to the adder 36. Then, the input from the AND circuit 38 to the adder 36 becomes 0, the output of the adder 36 becomes a frequency number value, and the value is held in the holding circuit 37. Thereafter, since the output of the edge detector 39 becomes 1, the stored value of the holding circuit 37 is input to the adder 36, added to the frequency number value input from the multiplier 34, and stored again in the holding circuit 37. You. Therefore, the accumulator 35 sequentially outputs address signals that increase by the frequency number value in each read cycle. Note that a +1 adder (not shown) is provided at the output of the accumulator 35 to actually perform the interpolation operation, and the address value output from the adder 36 and the value obtained by adding 1 to the address value Are output in a time-sharing manner.
[0014]
The waveform selector 40 is composed of, for example, a conversion table, inputs tone color information, upper bits of a key number signal, touch information, and the like, and outputs upper bits of a waveform read address for selecting a waveform in a waveform memory. The output of the waveform selector may be set in the holding circuit 37 at the time of key-on. The waveform memory 21 sequentially outputs waveform data based on the integer part of the read address output from the accumulator 35 and the upper address output from the waveform selector 40, and the interpolation circuit 22 outputs the waveform data 2 output from the waveform memory 21. A well-known interpolation operation is performed based on the two waveform data and the decimal part data of the read address output from the accumulator 35. With the above configuration, conversion to a highly accurate frequency number can be performed using a conversion table having a small capacity.
[0015]
In the upper clean, it represents the key number K with K = 12 * oct + note + Frac, note the remainder of the integer part (0 ≦ note ≦ 11), was a Frac is less of the fraction (0 ≦ Frac <1), The method of dividing note and Frac is arbitrary. For example, note may be up to 2 bits after the decimal point of K represented by a binary number, and Frac may be 3 to 7 bits after the decimal point. In this case, the number of frequency numbers stored in the frequency number table 32 is quadrupled to 48, but the number of exponent data in the exponent conversion table 30 is 1/4, 32, and the total number of data in both tables is The number is 80, which is smaller than in the first embodiment. Further, since the shift circuit 33 performs the multiplication function, the result is the same even if it is disposed after the multiplier 34.
[0016]
5, the exponential conversion table 30 were also combined in other applications, a block diagram illustrating a real施例waveform readout circuit 20 in the present invention. The blocks having the same functions as those in FIGS. 1 and 3 are given the same numbers. The difference from the above example is that the selector 50 is provided before the exponent conversion table, and the holding circuits 51 and 52 are provided behind. At the time of frequency number conversion processing, the selector 50 selects and outputs a key number signal, and the value subjected to exponential conversion by the exponential conversion table 30 is held in the holding circuit 51.
[0017]
On the other hand, when the signal is used for calculating the envelope signal, the signal of the envelope generator 24 is output from the selector 50, the exponentially converted value is held in the holding circuit 52, and -1 is added by the adder 53. The data is shifted to the left by 4 bits by the 4-bit shifter 54 and output to the selector 55. The exponent conversion table 30 stores exponent data as shown in FIG. 4B. In this state, for example, an input value taking a value between 0 and 1 also takes a value between 0 and 1 It cannot be converted to an exponential value. Therefore, by subtracting 1 from the output value of the exponent conversion table 30 and shifting the value by 4 bits (16 times), the output range becomes approximately between 0 and 1 (actually, 0 to 0.94376... Is). Here, a triangular wave having a value between 0 and 1, for example, is generated from the envelope generator 24, and when the selector 55 controls the selector 55 to output the signal from the 4-bit shifter 54 at the time of the down-link, the selection is performed. The output of the detector 55 has an envelope waveform that rises linearly when rising and decreases exponentially when falling.
[0018]
Figure 6 is a block diagram showing the main parts of a modification of the embodiment of FIG. In the above-described embodiment, an example in which the output of the envelope generator 24 is converted using the exponential conversion table is disclosed. In this modification, the exponent conversion table 30 is used to convert the speed parameter inside the envelope generator 24. This is an example used. The terminals A and B of FIG. 6 are connected to the terminals A and B of the read address generation circuit, respectively, instead of the envelope generator 24 and the selector 55 of FIG. The envelope generating circuit of FIG. 6 performs an operation of En + 1 = En + S * (T-En) by a configuration as shown, and generates an envelope signal En asymptotically approaching the target value T. Since the speed parameter is subjected to exponential conversion, a portion having a low speed can be set in more detail, and the accuracy is improved.
[0019]
The embodiments have been described above, but the following modifications are also conceivable. In the above-described embodiment, an example in which the exponential conversion table is used for other purposes is disclosed. However, as described above, it is also possible to increase the accuracy of the frequency number table. If conversion is performed using a number table, a conversion table that outputs an exponent that takes a value between 1 and 2 corresponding to a value from 0 to 96 (12 × 8) can be obtained. May also be used. In this case, the 4-bit shifter 54 becomes unnecessary.
[0020]
【The invention's effect】
As described above, in the present invention, the frequency number converting means only needs to convert, for example, only 12 frequency numbers corresponding to the integer part of the key number, and the exponential converting means has, for example, 7 bits after the decimal point. Since it is only necessary to convert the data into the corresponding 128 index data, when the conversion means is constituted by a table, there is an effect that the storage capacity of the table becomes very small as compared with the conventional example. Further, by using the conversion means in a time-sharing manner, the conversion means can also be used for other purposes such as an operation for generating an envelope signal, and there is an effect that hardware can be reduced.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a waveform reading circuit 20 of FIG.
FIG. 2 is a block diagram showing an example of a circuit configuration of an electronic musical instrument on which the present invention is based .
FIG. 3 is a block diagram showing a configuration of a tone generator 6 of FIG. 2;
FIG. 4 is an explanatory diagram showing the contents of each table.
FIG. 5 is a block diagram showing an embodiment of the present invention which also serves as an exponential conversion table.
FIG. 6 is a block diagram showing a main part of a modification of the embodiment of FIG . 5 ;

Claims (3)

First conversion means for inputting upper digit information of pitch information and performing exponential conversion;
Second converting means for inputting lower digit information of the pitch information excluding the upper digit information and performing exponential conversion;
Multiplying means for multiplying the output information of the first converting means and the output information of the second converting means to calculate a frequency number for reading a waveform from the waveform memory;
An electronic musical instrument, wherein at least one of the first conversion unit and the second conversion unit is also used as a conversion unit for performing exponential conversion of information other than the pitch information . The upper digit information is key number information in semitone units of the pitch information,
The electronic musical instrument according to claim 1, wherein the lower digit information is information of less than a semitone in pitch information. The first conversion means includes:
Conversion means for converting the upper digit information into difference information representing a difference from a predetermined octave in octave units and note name information within the octave,
Frequency number table means for inputting note name information and outputting a frequency number corresponding to a note name within a predetermined octave,
3. The electronic musical instrument according to claim 1, further comprising data shift means for shifting the frequency number data output from the frequency number table means by the number of bits indicated by the difference information.

Images