JPH0535278A - Storing method and synthesizing method for waveform - Google Patents

Abstract

PURPOSE:To obtain a sound source which reproduces a source musical sound faithfully with a small memory, facilitates editing operation, and to enable great timbre variation as the sound source of an electronic musical instrument. CONSTITUTION:A fundamental frequency and an amplitude which show a featured sine wave component are extracted from a source waveform by Fourier transformation and stored in a 1st memory M1. A fundamental waveform component is composed of the sine wave having the extracted fundamental waveform and amplitude by addition and this waveform component is subtracted from the source waveform to extract a residue component. The residue component is stored as PCM data in a 2nd memory M2. The waveform is composed by adding the residue component to the fundamental waveform component obtained by adding the sine wave having the fundamental frequency and amplitude.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic musical instrument or the like which stores musical tone waveform information in advance and generates a musical tone based on the waveform information. , A method of synthesizing the waveform.

[0002]

2. Description of the Related Art Conventionally, as a method of generating a waveform of a musical tone by a sound source such as an electronic musical instrument, there are, for example, an FM synthesis method, a PCM method, an addition synthesis method and the like. The principle of the FM synthesis method is
A sine wave output of a sine wave table called a modulator is added to a phase that changes at a predetermined speed to be a read phase, and waveform data is read from a sine wave table called a carrier at this read phase to generate a tone waveform that combines sine waves. I will get it.

This FM synthesizing method allows a great change in tone color by setting the amplitude of the sine wave output of the modulator and the reading speed (frequency of the phase) by various parameters, or by synthesizing in multiple stages, and making a memory. Although there is an advantage that it is not necessary, it is difficult to obtain the relationship between the parameter and the tone color, and it is difficult to produce a realistic sound such as a natural musical instrument.

On the other hand, in the PCM system, as is well known in digital audio technology, data obtained by sampling the amplitude of an actual musical tone is stored as a PCM table,
The tone waveform is obtained by sequentially reading the data in the PCM table. Therefore, since the recorded musical sound is reproduced as it is, a realistic sound can be produced. However, the only method of editing the timbre is to use a filter, and it is difficult to make a significant timbre change. Further, since the tone waveform is stored as it is, there is a problem that a large amount of memory is required.

The additive synthesis method stores the envelope in the frequency spectrum of the actual musical tone and adds many sine waves corresponding to the frequency and the envelope to perform Fourier synthesis of the musical tone waveform. Therefore, it is possible to reproduce a realistic sound and change the timbre. However, in order to Fourier-synthesize a real musical sound, a large number of sine waves are required for synthesis, so that the number of parameters becomes huge and it is difficult to edit the timbre.

[0006]

SUMMARY OF THE INVENTION According to the present invention, an original musical tone can be realistically reproduced with a small amount of memory in a sound source of an electronic musical instrument, and editing is easy, and a large change in tone color is possible. This is an issue.

[0007]

In order to solve the above-mentioned problems, the waveform storing method of the present invention extracts a fundamental frequency and its amplitude indicating a characteristic sinusoidal component of the original waveform from the original waveform. Then, while storing the basic frequency and the amplitude, a basic waveform component is added and synthesized from a sine wave having the basic frequency and amplitude, and the basic waveform component is subtracted from the original waveform to extract a residual component. It is characterized in that the residual component is stored.

The waveform synthesizing method of the present invention corresponds to the fundamental frequency indicating the characteristic sine wave component of the original waveform, the amplitude of the sine wave component, and the difference between the original waveform and the sine wave component. Based on the residual component, a basic waveform component is added and synthesized from a sine wave having the basic frequency and amplitude, and the residual component and the basic waveform component are added to synthesize a waveform. Characterize.

[0009]

In the waveform storing method and the synthesizing method of the present invention, the basic waveform component is added and synthesized from a sine wave having a fundamental frequency and an amplitude thereof, which shows a characteristic sine wave component of the original waveform. The residual component is obtained by subtracting the basic waveform component from the original waveform.

The data to be stored are the fundamental frequency showing the characteristic sine wave component of the original waveform, the amplitude of this sine wave component, and the residual component. Here, since the amplitude of the residual component is smaller than that of the original waveform, the number of bits is smaller than that in the case of storing the original waveform as it is as in the PCM method, and the residual component can be stored in a small memory. Moreover, since the synthesizing method corresponds to the reverse processing until the residual component is extracted, a waveform faithfully reproducing the original waveform can be obtained. In addition, the basic waveform component of the composite waveform can be made different from the basic waveform component at the time of storage simply by changing the basic frequency of the sine wave during synthesis, making it easy to edit and allowing significant changes in timbre. Is.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing a waveform storing method according to a first embodiment of the present invention. For example, musical tone waveform PCM data is used as original waveform data. First, the original waveform data is Fourier-transformed by the fast Fourier transform unit (FFT) 1 to obtain the frequency spectrum of the original waveform, and the fundamental frequency detection processing unit 2 converts the information of the frequency spectrum into the fundamental frequency and the harmonic component frequency, and each frequency. The corresponding amplitude and phase are extracted and stored in the first memory M1.

On the other hand, the sine wave synthesis processing unit 3 generates a sine wave corresponding to the fundamental frequency and the frequency of the overtone component based on the fundamental frequency, the frequency of the overtone component, each amplitude and phase from the fundamental frequency detection processing unit 2. Additive synthesis is performed in the same phase to generate a basic waveform component. Then, the sign of the basic waveform component is inverted and added to the original waveform. That is, the basic waveform component is subtracted from the original waveform, and the subtraction output is stored in the second memory M2 as the residual component.

The waveform will be described below. When the original waveform as shown in FIG. 2A is Fourier transformed, the fundamental frequency showing the characteristic sine wave component of the original waveform as shown in FIG. 2B. f
₁ and its harmonic component frequency f _2, f _3, ... peak appears at. The basic waveform component obtained by adding and combining these sine wave components becomes a waveform close to the original waveform as shown in Fig. 2 (C), and the residual component between the original waveform and this basic waveform component is as shown in Fig. 2 (D). It becomes a noise component.

As described above, the amplitude of the residual component becomes considerably smaller than the amplitude of the original waveform, and even if the residual component is stored as PCM data, the word of the memory is stored more than when the original waveform is stored as it is as PCM data. The number of bits of can be reduced. Moreover, the number of each data of the fundamental frequency, the frequency of the overtone component, each amplitude, and the phase is extremely smaller than the sampling number of the PCM data. Therefore, the memory capacities of the first memory M1 and the second memory M2 may be small.

FIGS. 3, 4 and 5 are block diagrams showing a waveform synthesizing method corresponding to the waveform storing method of the first embodiment. In each figure, the sine wave synthesis processing unit 3'performs the same processing as the sine wave synthesis processing unit 3 of FIG.
A sine wave corresponding to the fundamental frequency and the frequency of the harmonic overtone component is added and synthesized based on the fundamental frequency, the frequency of the harmonic overtone component, the amplitude, and the phase read out from 1 to generate the fundamental waveform component.

In the example of FIG. 3, the residual component read from the second memory M2 and the basic waveform component obtained by the sine wave synthesis processing unit 3'are added and output as a synthesized waveform. That is,
The processing is the reverse of the storage processing, and the original waveform is synthesized.

In the example of FIG. 4, the residual component processing unit 4 processes the residual component and adds it to the basic waveform component to synthesize the waveform. The processing includes filtering the residual component, AM modulation for modulating the amplitude of the residual component, and FM for modulating the frequency of the residual component.
Modulation or the like can be performed, and the timbre can be changed by this.

In the example of FIG. 5, the envelope from the envelope generator 5 is multiplied by the residual component, and this is added to the basic waveform component to synthesize the waveform. For example, the envelope of the musical sound of the piano has a high peak in the attack portion immediately after the key is pressed, so that it is possible to perform processing such as synthesizing the residual component only in this attack portion. By performing processing such as filtering, AM modulation, FM modulation, and envelope multiplication on the residual component in this way, it is possible to continuously change the timbre from the original sound to obtain different sounds.

FIG. 6 is a block diagram showing a waveform storing method according to the second embodiment of the present invention. In this embodiment, a band pass filter (BPF) 6 ₁ , 6 ₂ , having characteristics of extracting the fundamental frequency, the second order, ...
..., 6 _n , using the original waveform as the fundamental frequency, second order, ..., n
The sine wave components having the following frequencies are extracted respectively.

The extracted sine wave components are
Each Envelope Follower (EF) 7₁, 7₂, ..., 7
_nType in this envelope follower 7₁, 7₂，
…, 7 _nCalculate the amplitude value of each sine wave component by
The width value is stored in the first memory M1 together with each frequency.

On the other hand, the bandpass filters 6 ₁ , 6 ₂ ,
..., each sine wave component extracted in 6 _n is added and synthesized, and this is subtracted from the original waveform. As a result, a residual component obtained by removing the characteristic sine wave component from the original waveform is obtained, and this residual component is stored in the second memory M2. Also in the second embodiment, the memory capacities of the first memory M1 and the second memory M2 may be small as in the first embodiment.

FIG. 7 is a block diagram showing a waveform synthesizing method corresponding to the waveform storing method of the second embodiment. This synthesizing method is substantially the same as that in the case of FIG. 3, but since the phase of the sine wave component is not detected in the storage processing in the second embodiment of FIG. 6, the sine wave synthesizing processing unit 3 ″ uses the first memory. A sine wave is added and synthesized to generate a basic waveform component based on the basic frequency read from M1, the frequency and amplitude of each harmonic component, and the phase parameter given as a constant set number from the outside. A combined waveform is obtained by adding the residual component read from the second memory M2. The phase parameter given to the sine wave combination processing unit 3 ″ may be random.

In the above embodiment, the residual component of the original waveform and the basic waveform component is stored in the memory as it is, and the residual component read from the memory is added to the additive synthesis of the sine wave component as it is at the time of synthesis. . Therefore, as shown in FIG. 8 (A), the residual component is converted by the inverse function table F'and stored in the second memory M2, and as shown in FIG. The converted residual component read from the memory M2 is multiplied by a predetermined coefficient (or added,
The multiplication result may be converted into a residual component in the function table F and used for waveform synthesis.

In this way, it is possible to edit the tone color more effectively. For example, when the inverse function table F ′ is an inverse sine function and the function table F is a sine function, PCM modulation is performed with the carrier frequency in the FM sound source (FM synthesis method) being zero. If the multiplication coefficient before this function table F is increased, the number of harmonics increases, and a tone color change peculiar to an FM sound source can be produced.

If the residual component is converted and stored in the function table as described above, it is possible to perform more effective timbre editing by using different kinds of inverse function tables at the time of reading.

In each of the above embodiments, the characteristic sine wave component and the residual component of the original waveform are separated and stored. Therefore, by performing the synthesizing method as shown in FIG. 3, a realistic waveform can be obtained as in the conventional PCM method. In addition, when synthesizing the basic waveform component from the sine wave component, the fundamental frequency and the frequency of the overtone can be changed, and a new frequency component can be added, so that a significant timbre change can be performed. Further, since the basic waveform component and the residual component, which are characteristic sine wave components of the composite waveform, can be processed separately, the timbre editing becomes easy.

[0027]

As described above, according to the present invention, the fundamental frequency and its amplitude indicating the characteristic sine wave component of this original waveform are extracted from the original waveform, and the fundamental frequency and amplitude are stored. , A basic waveform component is added and synthesized from a sine wave having the basic frequency and amplitude, the basic waveform component is subtracted from the original waveform to extract a residual component, and the residual component is stored. The original waveform can be synthesized by adding and synthesizing the basic waveform components based on the fundamental frequency and the amplitude of the sine wave component and adding the residual component to this. Therefore, the data to be stored becomes a residual component whose amplitude is smaller than the fundamental frequency, the amplitude and the original waveform, and can be stored in a small memory. Further, since the original waveform can be synthesized by the processing reverse to the storage processing, the original musical sound can be reproduced realistically as in the conventional PCM method. Furthermore, the basic waveform component of the composite waveform can be made different from the basic waveform component at the time of storage by simply changing the basic frequency of the sine wave during synthesis, making it possible to easily edit and make subtle tone changes. Is.

[Brief description of drawings]

FIG. 1 is a block diagram showing a first embodiment of a waveform storage method of the present invention.

FIG. 2 is a diagram for explaining the processing of the first embodiment using waveforms.

FIG. 3 is a block diagram showing a method of synthesizing a realistic waveform in the first embodiment.

FIG. 4 is a block diagram showing a synthesizing method for processing a residual component according to the first embodiment.

FIG. 5 is a diagram showing a synthesizing method for performing envelope processing on a residual component in the first embodiment.

FIG. 6 is a block diagram showing a second embodiment of the waveform storage method of the present invention. It is a figure.

FIG. 7 is a block diagram showing a waveform synthesizing method in the second embodiment.

FIG. 8 is a block diagram illustrating an example of pre-storage and pre-combination processing for residual components according to each embodiment.

[Explanation of symbols]

1 ... FFT, 2 ... Basic frequency detection processing unit, 3 ... Sine wave synthesis processing unit, M1 ... First memory, M2 ... Second memory.

Claims (2)

[Claims] 1. A sine wave having a fundamental frequency and amplitude while extracting a fundamental frequency and its amplitude indicating a characteristic sine wave component of the original waveform from the original waveform and storing the fundamental frequency and the amplitude. A method of storing a waveform, wherein a basic waveform component is added and synthesized from the above, the basic waveform component is subtracted from the original waveform to extract a residual component, and the residual component is stored. 2. Based on a fundamental frequency indicating a characteristic sine wave component of the original waveform, an amplitude of the sine wave component, and a residual component corresponding to a difference between the original waveform and the sine wave component, A waveform synthesizing method, characterized in that a basic waveform component is added and synthesized from a sine wave having a fundamental frequency and an amplitude, and the residual component and the basic waveform component are added to synthesize a waveform.