JPH0285894A - Sound waveform synthesis device for electronic musical instruments - Google Patents

Abstract

(57) [Abstract] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a musical sound waveform synthesis device for an electronic musical instrument for generating musical sound waveforms in a musical instrument generating circuit of the electronic musical instrument.

[Prior art and problems to be solved by the invention] As a conventional musical sound waveform synthesis device for an electronic musical instrument.

A PCM synthesis method that generates musical tones by reading digitally expressed waveforms such as PCM, which are samples of actual musical sound waveforms for a predetermined period of time, from a storage means; 'A sine wave synthesis method that generates a musical sound waveform by adding sine wave waveforms that are overtones (harmonics) of the i note in arbitrary combinations, or a sine wave waveform that is stored in advance from a storage means. Nonlinear synthesis, such as the FD (phase distortion) method, generates musical waveforms by distorting the readout phase angle during readout, that is, by obtaining a signal waveform with harmonic components by nonlinear readout. Methods are known.

However, with the PCM synthesis method, it is difficult to generate a high-quality musical sound waveform unless the recording time of the musical sound waveform is long and the sampling frequency is high. There is a problem that it is necessary to increase the capacity of the means, which increases the cost. Since the sine wave synthesis method uses overtone addition,
It takes time to process the addition of high-order harmonic waveforms, and in real-time calculation processing, it is only possible to add waveforms of the 7th to 8th order harmonics, and it is difficult to create one musical tone with a complex waveform. The problem is that it is difficult. Furthermore, with the nonlinear synthesis method, it is difficult to generate musical sound waveforms whose harmonic components change over time, and the problem is that the circuit configuration becomes greatly complicated to faithfully reproduce such musical a. There is a point.

[Purpose of the Invention] This invention has been made to solve the above problems, and provides musical sound waveform synthesis for electronic musical instruments that enables generation of realistic musical sound waveforms close to natural musical instrument sounds with a simple circuit configuration. The purpose is to obtain equipment.

[Main Points of Issue 111] In order to achieve objective 1-1 as described above, this invention
A first musical sound waveform is generated based on musical sound waveform data read from the first storage means at a first sampling frequency, and a second sampling frequency lower than the first sampling frequency is generated from the second storage means. A second musical sound waveform is generated based on the musical sound waveform data read out, the sampling frequency of the second musical sound waveform is converted to the first sampling frequency by the sampling frequency converting means, and the sampling frequency of the second musical sound waveform is converted to the first sampling frequency by the combining means. The main point is that a desired tone waveform is obtained by combining the tone waveform and the second tone waveform that has passed through the sampling frequency conversion means.

[Embodiments] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

Configuration FIG. 1 is an overall circuit configuration diagram of an electronic musical instrument according to an embodiment of the present invention. The on/off state of the keyboard l and the switch section 2 is determined by the CPU 3, which is the central processing unit of the microcomputer.
is monitored, and the status of key on/+-off, tone selection, etc. is detected.

A ROM 4 and a RAM 5 are connected to the CPU 3 via a bus B, and the ROM 4 stores a program for the CPU 3 to control the entire circuit of this electronic musical instrument, as well as a program for controlling the entire circuit of this electronic musical instrument. The timbre parameters that define the timbre of the timbre are stored in advance. Also,
This RAM 5 includes various registers necessary for temporarily storing calculation results and the like when the CPU 3 performs the above control.

CPU3 uses this ROM4 and RAM5.

Generate pronunciation control data, timbre control data, etc.
It controls musical tone generation by a musical tf generating circuit 6 consisting of an SI or the like. The generated musical tone signal is converted into an analog signal by a DAC (digital-to-analog converter), amplified by an amplifier 8, and emitted as a musical tone through a speaker 9.

FIG. 2 is a diagram of the circuit 4a showing the specific contents of the music'\f generating circuit 6 in FIG. Arithmetic waveform synthesis circuit 10
generates musical tones by computational synthesis such as the sine wave synthesis method or nonlinear synthesis method mentioned above, and the sampling frequency is assumed to be 2 fs −= 2/T (T
is the period), then the output data f (t) is calculated by reading the data of the sine wave waveform as the basic waveform from the built-in ROM when the arithmetic waveform synthesis circuit 10 uses the sine wave synthesis method. f (t) = ΣAnsin
It is expressed by the formula ωo t (1) (where n is an integer, t is time, and A is the amplitude), and
In the case of a nonlinear synthesis method, in a general nonlinear synthesis method, the output data f(1) is t (t) = F (AsIn ωt)
(2) In the FM modulation method, the output data f (t) is f
(t) = Aosin (ωot+ A15in
(ωst) (3) In other nonlinear synthesis methods, the output data f(1) is f(t)=Aosin(Alsin ω+t+A25i
n ω? t) (4) Here, in the case of the nonlinear synthesis method, as is well known, it is relatively easy to output an output that includes high-order harmonics, so the aliasing distortion is treated as noise and is not audible to the ear. In order to suppress this, it is necessary to set the sampling frequency sufficiently high so that it does not affect the audible frequency range. Therefore, the lower limit of the double-sided band at the sampling frequency of 2 fs is set to approximately 40 KHz to 50 KHz.

The PCM waveform synthesis circuit 11 converts the PCM waveform data stored in the PCM waveform memory 12 (digital representation waveform data obtained by PCM recording of actual musical instrument sounds, etc.) into the keyboard l based on control data from the bus B of the CPU 3. By reading out according to the pitch data changed by the operation of
Musical sound waveform g of frequency a (pitch) based on the pitch data
(1). Here, the sampling frequency is set to a frequency lower than the sampling frequency a 2 f s in the arithmetic waveform synthesis circuit 1O described above, but in this embodiment, since it is in the cylinder, fs = 1/T of 1/5 of 2fs. shall be. In other words, the fs(7) value is 20KH
z to about 50KHz. Therefore, the band of the PCM waveform recorded in advance in the PCM waveform memory 12 is 2.
This is approximately half of 0KHz, l0K)Iz, but the storage capacity as a memory is only half that of recording at a sampling frequency of 40KHz to 50KHz.However, the frequency band of the PCM waveform that can be recorded is insufficient. Therefore, by combining the output data g (t) from the PCM waveform synthesis circuit 11 with the output data f (t) from the arithmetic waveform synthesis circuit IO described above, the deficiency in the frequency fiF is compensated for.

The sampling frequency conversion circuit 13 converts output data g (t
) has a function (interpolation function) of converting the sampling frequency to a higher frequency, and in this embodiment,
The frequency value of the sampling frequency fS is converted to 2fs, that is, the sampling frequency in the arithmetic waveform synthesis circuit 10.

Output data g'(t
) is the arithmetic waveform synthesis circuit lO described above in the arithmetic circuit 14.
is combined with the output data f (t) by addition or multiplication. The musical tone signal waveform h (t) obtained from this arithmetic circuit 14 is expressed by the following formula (5) where f (1) and g' (t) are both functions of time. Since these are simple additions, the attack part of the musical tone is divided using the PCM synthesis method using the PCM waveform stored in the PCM waveform memory 12, and the subsequent steady part is divided by the arithmetic synthesis method. How to use it is possible.

In addition, in the case of equation (8), the output data g' (L) of the musical sound waveform by the PCM synthesis method and the output data g' (
t) and the musical sound waveform f(1) using an arithmetic synthesis method,
In other words, the waveform is the sum of the musical waveform whose frequency has been changed by amplitude modulation, so even if a PCM waveform with a narrow frequency bandwidth, that is, a low sampling frequency, is used,
A complex rt waveform h (B) containing many high-order harmonics is obtained.

Then, the musical sound waveform h (t
) is added to the DAC 7 and converted into an analog signal to become a musical tone.

Written by Kudo Next, the operation of the embodiment having the above configuration will be described.

In addition, Fig. 3 is a waveform spectrum diagram of Raku-a, and Fig. 3 (
In each of a) to FIG. 3(d), the horizontal axis represents the angular frequency and the vertical axis represents the level. FIG. 3(a) shows the original sound spectrum of the waveform stored in the PCM waveform memory 12 Since the sampling frequency is fS, the band is limited to fs/2 to prevent aliasing distortion.

FIG. 3(b) shows the spectrum of the output data g(t) sampled at the sampling frequency fS and output from the PCM waveform synthesis circuit 11.

FIG. 3(C) shows the spectrum of the output data g'(t) mentioned above, and this output data g'(t) is the spectrum of the output data g(t) sampled at the sampling frequency afs. The sampling frequency fS is
Sampling frequency conversion circuit 1, which is an ideal filter
3, the sampling frequency is twice the frequency a 2 f s
It was obtained by converting to Here, Fig. 3(b)
If you compare it with , you can see that the vector has been completely removed.

Further, FIG. 3(d) shows the spectrum of the synthesized output data f(t) from the arithmetic waveform synthesis circuit 10, and the sampling frequency *2fs(=2/T
), the spectrum is PCM
It has twice the bandwidth of the output data g (t) from the waveform synthesis circuit 11.

In this way, in the arithmetic circuit 14, the output data g' (t) having the spectrum shown in FIG.
By calculating and combining the output data f(1) from the arithmetic waveform synthesis circuit 10 having the spectrum shown in FIG. 3(d) at a sampling frequency of 2fs,
An output waveform h(t) (not shown) having a frequency band similar to the spectrum shown in d) is generated as musical waveform data.

FIG. 4 shows an example of a specific configuration of the sampling frequency conversion circuit 13. This sampling frequency conversion circuit 13 converts the sampling frequency of the output data g(1) to a higher frequency (interpolation). (to do)
It functions as the most important first-order interpolation filter. That is, the output data g (t
) is the output data g that has passed through the delay circuit 15 at the adder 16.
(t)-1, and the lli power of the adder 16 is 1
/2 multiplier 17 and output as output data g'(t). Due to this operation, the output data g(t) from the PCM waveform synthesis circuit 11 cannot completely remove the spectrum around Ω=2π/T as shown in FIG. 3(b). However, this output data g′
There is no problem if (t) is used to obtain a complex rising waveform such as an attack portion in order to obtain a desired tone waveform.

In addition, as shown in FIG. 3(C), Ω=2π/T is completely satisfied.
In order to remove the spectrum around , it is sufficient to increase the order of the interpolation filter.

Furthermore, in this embodiment, the frequency ratio between the sampling frequency in the PCM waveform synthesis circuit 11 and the sampling frequency in the arithmetic waveform synthesis circuit 10 is set to 1.
:2, but since realizing an interpolation filter as a sampling frequency conversion circuit is a convolution if the ratio is a rational number, it is not limited to l:2.

Furthermore, in the arithmetic waveform synthesis circuit 10, the tone waveform is not limited to a sine wave, but may be a tone waveform such as a rectangular wave or a sawtooth wave.

Furthermore, the present invention is not limited to reading out and synthesizing the PCM waveform of the musical tone, and the digital representation of the musical tone waveform may be based on various types of encoding such as DPCM, ADPCM, etc. other than PCM.

[Effects of the Invention] As described above, according to the present invention, the circuit configuration of the first musical sound waveform synthesizer that generates based on repeatedly read musical sound waveforms can be made small, and the second sampling Since the frequency is lowered, the memory capacity of the second memory means can be small, and moreover, this second memory ti! PCM in f-stage
By storing the waveform data, it is possible to obtain waveforms with complex overtone compositions, and musical tones containing many harmonic components in the attack part can be created. This has the effect of providing a musical sound waveform synthesis device for an electronic musical instrument that is capable of generating musical sound waveforms.

[Brief explanation of the drawing]

FIG. 1 is an overall circuit configuration diagram of an electronic musical instrument according to an embodiment of the present invention, FIG. 2 is a configuration diagram of a musical tone generation circuit in FIG. 1,
FIG. 3 is a musical sound waveform spectrum diagram for explaining an embodiment of the present invention, and FIG. 4 is a diagram showing the configuration of a sampling frequency conversion circuit. 3... CPU, 6... musical tone generation circuit,
10...vtr1 waveform synthesis circuit, 11...
... PCM waveform synthesis circuit, 12 ... PCM waveform memory, 13 ... sampling frequency conversion circuit, 14 ... calculation circuit. Ga (j crystal)

Claims (3)

[Claims] (1) a first storage means for storing musical sound waveforms for repeated reading; a first musical sound waveform generation means for generating a musical sound waveform; a second storage means for storing a musical sound waveform of a part or all of a musical tone; and a second musical sound waveform having a frequency lower than the first sampling frequency from the second storage means. a second musical tone waveform generating means for generating a second musical tone waveform based on the musical tone waveform data read out at a sampling frequency; sampling frequency converting means for converting the first musical sound waveform into the second sampling frequency; and combining means for combining the first musical sound waveform and the second musical sound waveform that has passed through the sampling frequency converting means to obtain a desired musical sound waveform. A musical sound waveform synthesis device for an electronic musical instrument, characterized by comprising the following. (2) The musical tone of the electronic musical instrument according to claim 1, wherein the combining means performs an addition operation on the first musical sound waveform and the second musical sound waveform that has passed through the sampling frequency conversion means. Waveform synthesizer. (3) The combining means performs an addition operation on the musical sound waveform obtained by multiplying the first musical sound waveform by the second musical sound waveform that has passed through the sampling frequency converting means, and the first musical sound waveform. 2. A musical sound waveform synthesis device for an electronic musical instrument according to claim 1.