JPH0997084A - Method and device for musical sound production - Google Patents

Abstract

PROBLEM TO BE SOLVED: To employ either of a polyphonic-number valued mode and a quality valued mode according to the user's purpose of use. SOLUTION: In note event data stored in a sound source register, 128 waveform samples generated by a sound generation channel to which an arithmetic mode 0 is assigned are accumulated by a buffer 0, 64 waveform samples generated by a sound generation channel to which an arithmetic mode 1 is assigned are accumulated by a buffer 1, and 32 waveform samples generated by a sound generation channel to which an arithmetic mode 2 is assigned are accumulated in a buffer 2. The 64 and 32 waveform samples accumulated by the buffers 1 and 2 are interpolated respectively and stored as 128 waveform samples in buffers 1' and 2'. Then the 128 waveform samples in the buffer 0 and the 128 waveform samples in the buffers 1' and 2' are accumulated by samples SDn (n=1..., 128) and stored in the buffer 0, and the waveform samples stored in this buffer 0 have their reproduction reserved to a DMA(Direct Memory Access) control part.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a musical tone generating method and apparatus for simultaneously producing musical tones of a plurality of channels.

[0002]

2. Description of the Related Art Sound sources for simultaneously producing musical tones of a plurality of channels have been known. In such a conventional sound source, a tone is independently generated for each sound generation channel, and the number of waveform samples calculated and generated per unit time is made constant for each sound generation channel.

[0003]

However, in the above-mentioned conventional sound source, since the musical tone is independently generated for each sound generation channel, the characteristic of the musical sound generated for each channel is different, and the musical sound is generated for each sound generation channel. Although the quality of the required musical tone differs depending on the musical tone
Since the waveform samples are generated with the same number of samples for all the tone generation channels, there may be a waste of calculation processing for tone generation.

For example, when it is desired to generate a musical tone having a frequency component over a wide frequency band (that is, a high quality musical tone), it is necessary to perform a musical tone waveform generation calculation at a high sampling frequency (that is, a large number of samples). On the other hand, when generating a musical tone having only frequency components in the low frequency band, it is sufficient to perform the musical tone waveform generating operation at a low sampling frequency (that is, the number of samples is small). In addition, depending on the musical composition, the quality of each note may be low so that a large number of pronunciations may be secured, or the number of pronunciations may be small and the tone quality may be improved. In addition, it is desirable to generate high-quality musical tones for sound channels that generate conspicuous parts such as the lead part in a performance song, while the quality is slightly degraded in parts that are relatively unnoticeable such as backing. Even if the generated musical sound is generated, there is no problem in hearing.

In general, a sound source generates a musical sound by a musical sound generation calculation in an arithmetic circuit, and the arithmetic circuit has a limit in arithmetic capacity per unit time. Therefore, the number of musical tones that can be simultaneously generated (the number of simultaneous sounds) is limited. . In the above conventional sound source, since the calculation quality of each sound generation channel is made uniform,
Depending on the channel, there is a case where a tone is generated and calculated with a quality higher than necessary, which causes a problem that the number of tones that can be simultaneously pronounced (the number of sounding channels) is reduced.

The present invention has been made in view of the above problems, and provides a tone generation method and apparatus capable of adopting either the number of pronunciations emphasis or the quality emphasis according to the purpose of use of the user. The purpose is to

[0007]

In order to achieve the above object, the musical tone generating method according to claim 1 comprises a step of generating a musical tone based on musical tone waveform samples generated by a plurality of channels. , A step of inputting performance information, a step of inputting control information according to an operation amount of a manipulator operated by a user, and a step of inputting a channel corresponding to the input performance information within the predetermined period every predetermined period. Generating a tone waveform sample for each channel with the number of samples according to the input control information, and generating a tone based on the tone waveform sample thus generated. .

According to a second aspect of the present invention, there is provided a musical tone generating method including a step of generating a musical tone based on musical tone waveform samples generated by a plurality of channels, and inputting performance information of a plurality of parts. Setting the control information of each part, assigning the input performance information to any one of the plurality of channels, and assigning the musical tone waveform sample to which the performance information belongs to the assigned channel to the performance. Generating a musical tone based on the generated musical tone waveform sample, and generating the musical tone waveform at a time density according to the set control information of the part to which the information belongs.

A musical tone generating method according to a third aspect of the present invention is a musical tone generating method including a step of generating a musical tone based on a musical tone waveform sample, and inputting performance information.
A step of generating control information and a musical tone generation operation based on the waveform data stored in the waveform memory are executed according to the input performance information, and musical tone waveform samples are generated at a time density according to the generated control information. In the step of generating the tone waveform sample, the tone generation operation is performed by selectively using different waveform data in the waveform memory according to the generated control information to generate the tone. The step is characterized in that a musical tone is generated based on the musical tone waveform sample generated by the musical tone generating calculation.

According to a fourth aspect of the present invention, there is provided a method of generating a musical tone waveform, wherein the first waveform data for generating a musical tone at a predetermined sampling frequency is stored in a storage means, and the stored first waveform data is stored. Converting to waveform data having a sampling frequency different from a predetermined sampling frequency and storing the waveform data in the storage means as second waveform data; and generating a tone waveform sample based on the waveform data stored in the storage means. In the step of generating the tone waveform sample, the tone waveform samples can be generated at a plurality of different time densities, and the first waveform data and the second waveform data are generated according to the time densities. One of these is selected, and a musical tone waveform sample is generated based on the selected waveform data.

According to a fifth aspect of the present invention, there is provided a musical tone generating apparatus having musical tone generating means for generating musical tones based on musical tone waveform samples generated by a plurality of channels. A control information input means for inputting control information in accordance with an operation amount of an operator; and, for each predetermined period, the musical tone waveform sample of the channel corresponding to the input performance information within the predetermined period. The number of samples according to the control information, and has a musical tone waveform sample generating means for generating for each channel,
The pre-musical tone generating means is characterized by generating a musical tone based on the generated musical tone waveform sample.

According to a sixth aspect of the present invention, there is provided a musical tone generating apparatus having a musical tone generating means for generating a musical tone based on musical tone waveform samples generated by a plurality of channels. Information input means, control information setting means for setting control information of each part, and the input performance information is assigned to any one of the plurality of channels, and the performance information is assigned to the assigned channel. And a musical tone waveform sample generating means for generating a musical tone waveform sample to which the musical tone waveform sample belongs at a time density according to the set control information of the part to which the musical performance information belongs. It is characterized in that a musical sound is generated based on it.

According to a seventh aspect of the present invention, there is provided a musical tone generating apparatus having musical tone generating means for generating musical tones based on musical tone waveform samples, wherein musical performance information input means for inputting musical performance information and control information are generated. Control information generating means,
Music tone waveform sample generation means for performing a tone generation operation based on the waveform data stored in the waveform memory in accordance with the input performance information, and generating a tone waveform sample at a time density according to the generated control information. The musical tone waveform sample generating means selectively performs musical tone generating calculation by selectively using different waveform data in the waveform memory according to the generated control information, and the musical tone generating means performs the musical tone generating calculation by the musical tone generating calculation. A feature is that a musical tone is generated based on the generated musical tone waveform sample.

According to the configuration of the first and fifth aspects of the invention, the musical tone waveform sample of the channel corresponding to the performance information input within the predetermined period corresponds to the control information input every predetermined period. The number of samples is generated for each channel, and the musical tone is generated based on the generated musical tone waveform sample. Therefore, either the number of pronunciations or the quality is important depending on the purpose of use of the user. You can

According to the present invention, the musical tone waveform sample of the channel corresponding to the input performance information corresponds to the set control information of the part to which the performance information belongs. Are generated at different time densities, and musical tones are generated based on the generated musical tone waveform samples, so that musical tones of a part that have a great auditory effect can be generated with high quality, and the limited computing power is utilized to the maximum. be able to.

According to the third and seventh aspects of the present invention, the tone generation operation based on the waveform data stored in the waveform memory is executed according to the input performance information, and the generated control is executed. Musical tone waveform samples are generated at a time density according to information, and in the musical tone generation calculation, musical tone generation calculation is performed by selectively using different waveform data in the waveform memory according to the generated control information. Since a musical tone is generated based on the musical tone waveform sample thus generated, a channel having a high equivalent sampling frequency has a frequency component over a wide band (that is,
Waveform data (of high recording sampling frequency)
Waveform data having a narrow band of frequency components (that is, low recording sampling frequency) can be used in a channel having a low equivalent sampling frequency, which allows the waveform designation in the timbre data to be unchanged.

If the waveform data to be selected is not changed in accordance with the time density of the musical tone waveform sample to be generated, folding noise may occur in the musical tone waveform sample to be generated for the following reason. , It may not be possible to generate musical tones with quality according to the time density. here,
The time density is a sampling frequency at which a musical tone waveform sample is generated, and is referred to as an equivalent sampling frequency in this specification. According to the sampling theorem, a tone waveform sample has a frequency half the sampling frequency at which it is sampled (hereinafter referred to as the “upper limit frequency”
It is possible to reproduce frequency components in the following bands.

When a musical tone is generated using the waveform data stored in the waveform memory, the stored waveform data is converted into a waveform sample having the pitch of the musical tone to be generated under the equivalent sampling frequency ( Hereinafter, it will be referred to as “pitch conversion”), and a musical tone waveform is generated based on the waveform data after the pitch conversion. At this time, if the waveform data obtained as a result of the pitch conversion of the waveform data includes a frequency component higher than the upper limit frequency corresponding to the equivalent sampling frequency, the frequency component is a folding noise and the waveform data after the pitch conversion. Will be mixed in. on the other hand,
When the waveform data after the pitch conversion is considerably lower than the upper limit frequency, for example, includes only a frequency component equal to or lower than one-third, the high-frequency component is generated even though the waveform is generated at an equivalent sampling frequency which is actually high. Only the missing musical tones are produced and the quality of the musical tones is not very good.

In the present invention, waveform data suitable for the generation time density of a musical tone waveform sample is prepared as waveform data used as a material for musical sound generation, and the waveform data is selected and used according to the time density.

Further, according to the configuration of the invention described in claim 4, as in the invention of claim 3, since the waveform data corresponding to the generation time density of the musical tone waveform sample is used, the aliasing noise is reduced. it can. Furthermore, when a musical tone waveform sample is generated at two different time densities, the waveform data obtained by converting the waveform data suitable for one of the time densities is used for the waveform generation of the other time density, so that even if the time densities are different, It is possible to generate musical tone waveform samples that sound the same timbre.

[0021]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of a musical tone generating apparatus according to an embodiment of the present invention. The configuration of this block diagram is exactly a general-purpose computer itself on which an operating system (OS) such as Windows runs. Therefore, the present invention can be implemented as software on a general-purpose computer. It can also be implemented on an electronic musical instrument having the same configuration as this block diagram.

In the figure, the musical tone generating apparatus of this embodiment has a CPU 1 for performing various data processing, a keyboard 2 for a user to input a program execution instruction and data, various image information and A display 3 for displaying character information, a hard disk device 4 for storing programs and data executed by the CPU 1, a ROM 5 for storing data input / output control programs for the keyboard 2, the display 3, the hard disk 4, etc., and execution RAM 6 for storing programs, waveform data, and data being calculated, a timer 7 for measuring time, a MIDI interface 8 connected to a performance device such as a keyboard for inputting performance data, and an instruction from the CPU 1. According to this, the RAM 6 is directly accessed, and the sampling frequency is 48 kHz, for example. Input to the DA converter 10, sample by sample reads musical tone waveform data at the corresponding frequency (hereinafter, referred to as "read reproduction process") to DMA (Direct M
emoryAccess) controller 9, a DA converter (DAC) 10 for converting musical tone waveform data of a digital signal supplied from the DMA controller 9 into an analog tone signal, and a sound system 11 for amplifying the tone signal and outputting it from a speaker. And a bus 12 for interconnecting the above-mentioned components 1 to 9
And

Hereinafter, in the present embodiment, “waveform sample” or “sample” means individual sampled waveform sample data, and “waveform data”.
Of the data collected from the individual waveform sample data, it means the data that is the basis of the tone generation processing, and the "tone waveform data" is the data generated as a result of the tone generation processing of the same data. I will refer to the data that I obtained.

FIG. 2 is a diagram showing a structure of tone color data and waveform data stored in the RAM 6 and a structure of an input buffer and a tone generator register set in the RAM 6.

First, the tone color data PDp (p
= 1, ..., 16) are waveform designating data WN (p) for designating waveforms in the respective musical ranges, and LFO (Low Frequency).
Oscillator) control original data (OD), filter envelope generator (FEG) control original data, amplitude envelope generator (AEG) control original data, and other original data,
Computation mode CM (p). here,
p indicates a part number, and the musical tone generating apparatus according to the present embodiment is configured so that 16 parts of tone color data can be set. Further, in this embodiment, the waveform designation data WN (p) is set by the waveform name, and the waveform name is set by the user (see FIG. 9). Further, the calculation mode CM (p) indicates data (equivalent to the equivalent sampling frequency) that indirectly specifies how many samples are to be generated per second when a musical tone waveform is generated. In the present embodiment, the operation mode CM (p) is configured to take any one integer value from 0 to 2. Here, the sampling frequency equivalent to the operation mode CM (p) is 48 kHz for CM = 0, 24 kHz for CM = 1, and C
M = 2 has a correspondence relationship of 12 kHz.

Next, the waveform data WD1 and W of FIG.
D2, ... Denote waveform data as a source of the tone waveform data generated by the tone generation processing. Of these waveform data, the waveform data WD1 and WD2 are waveform data each including a sample of a predetermined time length. , A one-shot read waveform or a waveform having an attack part and a loop part. Each of the waveform data WD1 and WD2 is a plurality of waveform data that are sampled at a predetermined recording sampling frequency and stored in the hard disk 4, and are read out as needed and stored in the waveform data area of the RAM 6. The waveform data WD1 'and WD2' are respectively subjected to a predetermined band limitation on the waveform data WD1 and WD2, down-sampled (skipping one sample at a time) at half of the original recording sampling frequency, and placed in the waveform data area. It is stored. The waveform data WD2 ″ is
The waveform data WD2 'is subjected to predetermined band limitation, down-sampled at half the sampling frequency of the waveform data WD2', that is, at 1/4 of the original recording sampling frequency, and stored in the waveform data area. In addition,
Details of the process (waveform LPF process) for generating the waveform data WD1 ′, WD2 ′, WD2 ″ will be described later with reference to FIG.

The input buffer shown in FIG. 3C is a buffer for storing performance data input via the MIDI interface 8, and an area for storing data indicating the number of events waiting to be processed and an event buffer for each event. The area for storing corresponding event data ID1, ID2, ID3 ,. Each event data consists of data indicating the content of the event and data indicating the event occurrence time.
The data indicating the time of occurrence is necessary for the CPU 1 to collectively process a plurality of events.

The tone generator register shown in FIG. 3D stores control data for each tone generation channel in the tone generation processing, and in this embodiment, registers for 32 channels are provided. The control data for each tone generation channel is generated by processing various original data (see (a) above) of the tone color data PDp according to performance data such as touch, and determines the pitch of the musical tone to be generated. The indicated pitch SP and F number F indicating the amount of advance of the address per sample when reading the waveform data
N, waveform designation data for designating waveform data to be read, LFO control data generated by processing the LFO control original data, FEG control data generated by processing the FEG control original data, and the AE
AEG control data generated by processing the G control original data, note-on data indicating whether or not sound is being generated, other data, and operation mode CM (i) (i = 1, ...,
32) and.

Here, the F number FN is 1 as described above.
It is a numerical value indicating the amount of advance of the read address per sample, and is specifically calculated and set by the following equation.

FN = 2 ^{(SP-OP) / 1200} × 2 ^(CM-WM) where SP represents the pitch SP of the musical sound to be generated,
OP indicates a unique pitch (original pitch) of the waveform data when the waveform data designated by the waveform designation data is read out one sample at a sampling frequency of 48 kHz, and CM is the operation mode CM.
(I), WM is a waveform mode, that is, a number that is given to each waveform data one by one and indirectly indicates the recording sampling frequency of the waveform data (in the present embodiment, the waveform mode WM is also the calculation mode CM). 0-2 as in (i)
Is configured to take any one of the integer values). For example, the waveform mode WM = 0,
Corresponding to the recording sampling frequency of 40 kHz to 48 kHz, the waveform mode WM = 1 corresponds to 20 kHz to 24 kHz, and the waveform mode WM = 2 corresponds to 10 kHz to 12 kHz. The waveform data in each waveform mode can have a high frequency component with an upper limit of half the recording sampling frequency. Since the values SP and OP are given as cent values, they are divided by "1200".

Here, it is this F number FN that controls the aspect of the pitch conversion described above.

Further, a work area for sound generation processing is secured in the sound source register, and this work area is used, for example, when it is necessary to correct the F number FN by LFO control data or the like.

FIG. 3 is a diagram showing the structure of the output buffer set on the RAM 6.

In the figure, the buffer 0 is a calculation mode 0 (CM) in the control data stored in the tone generator register.
(I) = 0) is a buffer for accumulating 128 musical tone waveform samples generated in the assigned tone generation channel.
(I) = 1) is a buffer for accumulating 64 musical tone waveform samples generated in the assigned tone generation channel.
(I) = 2) is a buffer for accumulating 32 musical tone waveform samples generated in the assigned sounding channels.

The buffer 1'is the same as the buffer 1 64.
Buffer 2 ', which is a buffer for interpolating 128 musical tone waveform samples to generate 128 musical tone waveform samples.
Is a buffer for interpolating 32 musical tone waveform samples of the buffer 2 to generate 128 musical tone waveform samples. The 128 tone waveform samples of buffers 1'and 2'generated in this way are
128 musical sound samples are accumulated for reproduction and 128 for reproduction.
It is stored in buffer 0 as individual waveform samples. That is, the buffer 0 also functions as a buffer for storing the 128 musical tone waveform sample data for reproduction read by the DMA control unit 9.

Buffer 0 is used to interpolate the musical tone waveform samples (the musical tone waveform sample data of buffers 1 and 2 in this embodiment) generated at such a low sample rate into 128 musical tone waveform samples. The 128 generated musical tone waveform samples and the 64 and 32 musical tone waveform samples generated in the buffers 1 and 2, respectively, are accumulated without generating aliasing noise, and the 128 reproduced musical tone samples are finally reproduced. This is to generate a musical tone waveform sample. As this interpolation method, a well-known method such as linear interpolation may be used.

In this embodiment, as described above, 3
Different types of buffers 0 to 2 are provided, and the buffer size of each buffer is as follows: buffer 0: buffer 1: buffer 2 = 4:
Although the ratio is set to 2: 1, it goes without saying that the number of buffers and the buffer size of each buffer are not limited to this. For example, two types of buffers 0 and 1 may be provided, and the buffer size of each buffer may be set so that buffer 0: buffer 1 = 3: 1.

Next, the outline of the tone generation processing performed by the tone generation device of the present embodiment will be described with reference to FIG.

When, for example, a note-on event of part p is input to the input buffer shown in FIG. 2C, the tone color data PDp is designated by this event, and the waveform name WN (p) and the operation are calculated by this tone color data PDp. When the mode CM (p) is designated, the CPU 1 executes the operation mode CM
A waveform sample of a musical tone corresponding to an event input at the time density (equivalent sampling frequency) of the musical tone generating device corresponding to (p) is generated on a buffer corresponding to the operation mode,
A generation operation such as an interpolation operation is executed on this waveform sample, the calculated data is stored in the buffer 0 of the RAM 6, and the completion of the operation is notified to the DMA control unit 9 (FIG. 4).
(B)). The DMA control unit 9 sequentially reads out the data and performs a reproduction process ((c) in the figure). Therefore, MI
Of the performance data input via the DI interface 8, the performance data corresponding to the performance input ((a) in the same figure) between the time tBC at which the clock BC was generated last time and the time tBC at which this clock was generated this time is generated. It is the target of calculation. Then, the waveform data read / reproduction processing based on the generated data that has been calculated at time tCE is performed from the next time tBC to D.
This is executed by the MA control unit 9 and a musical sound is output. The arrow P shown in the figure simply indicates the correspondence between the data that has been generated and calculated and the read / reproduction processing.
It does not indicate that the operation result finished at is transferred at time tBC.

The DAC 10 of the musical tone generating apparatus of this embodiment
Since the sampling frequency at is 48 kHz,
The generation cycle of the clock BC is 2.7 msec (128/4).
8k), and the maximum delay time from the performance input to the actual tone generation is about 5 msec, which is not a problem for human hearing. Of course, in the case of automatic performance, there is no problem even if this delay time is longer, so the size of the buffer 0 may be increased. Further, the read / reproduction processing is not limited to the timing of the clock BC, and may be any timing, for example, after a predetermined time from the time tBC of the clock BC, as long as it does not cause a hearing problem.

C of the musical tone generating apparatus configured as described above
The control process executed by the PU 1 will be described below with reference to FIGS.

FIG. 5 is a flowchart of the main routine. For example, when the user turns on the power of the musical tone generating device (in the case of a general-purpose personal computer, when the program of the software tone generator is started), the process is started. . FIG. 6 is a flowchart of MIDI reception interrupt processing. This interrupt processing is executed with the highest priority when performance data is input via the MIDI interface 8.

First, the processing of FIG. 6 will be described. In step S11, received data is fetched, and in step S12.
Then, the received data is written in the input buffer of the RAM 6 together with the time data indicating the reception time.

In the main routine of FIG. 5, first, all tone generation channels are turned off, registers are cleared, and further initialization processing such as initializing and starting the reading and reproducing processing of the DMA control unit 9 is performed (step S1). Then, it is determined whether or not there is received data in the input buffer (step S2). As a result, if there is no received data, the process immediately proceeds to step S4. If there is received data, a process corresponding to the received data, for example, note-on event process, note-off event process, pedal process, etc. is performed, and then step S4.
Proceed to.

In step S4, it is determined whether or not a switch event such as selection of a tone color has occurred. If not, the process immediately proceeds to step S6. If so, the tone color selection switch is set according to the setting. After performing a panel switch event process such as selecting a tone color for each MIDI channel (step S5), the process proceeds to step S6.

In step S6, a sound source processing subroutine which will be described later with reference to FIG. 13 is executed, and in step S7, other processing is performed and the process returns to step S2. After that, steps S2 to S7 are repeatedly executed.

FIG. 7 is a flow chart showing the procedure of note-on event processing which is one of the received data processing executed in step S3 of FIG.

First, in step S21, the pitch and part number indicated by the received note-on event data are
The areas SP (hereinafter, the content is referred to as “pitch SP”) and p (hereinafter, the content is referred to as “part p”) secured in the RAM 6 are stored, and the time of occurrence (reception time) is stored in the RAM 6. Area TM (hereinafter, this content is called "occurrence time TM")
To be stored.

Next, a tone generation assignment process is performed to determine which channel of the tone generator register (FIG. 2 (d)) is to be written, and the assigned channel number is reserved in the RAM i in the area i (hereinafter, this content is referred to as "assigned channel i"). “”) (Step S22).

In a succeeding step S23, the tone color data PDp (FIG. 2A) of the part p is searched to obtain the waveform name (waveform designation data) WN (p) and the operation mode CM (p). The waveform data corresponding to the operation mode CM (p) is selected from the waveform data having WN (p),
The read address on the RAM 6 (see FIG. 2B) is set as the waveform designation data (FIG. 2D) of the assigned channel i.

At this time, there may be a plurality of waveform data having the same waveform name WN (p). For example, the waveform data WD2 'and WD2 "are obtained by down-sampling the waveform data WD2 as described above, and the waveform name is the same as the waveform name of the waveform data WD2.
When there are a plurality of waveform data having the same waveform name as described above, one waveform data is selected from the waveform data according to the operation mode CM (p).

However, even when a plurality of waveform data having the same waveform name exist, the waveform data corresponding to the operation mode CM (p) may not exist. Figure 2
In (b), the waveform data WD1 ″ corresponding to the calculation mode CM (p) = 2 does not exist.
This is because, for example, the waveform data WD1 ″ was not created by the waveform LPF processing (described later with reference to FIG. 11) or the like. In such a case, the waveform data corresponding to the calculation mode CM (p) cannot be selected. Therefore, the same waveform name W
By selecting waveform data of different waveform modes WM with N (p) and changing the read speed of the waveform data (that is, changing the pitch, specifically, adjusting the F number FN), a desired musical tone signal is obtained. To generate.

In this embodiment, the calculation mode CM
There is a one-to-one correspondence between (p) and the waveform mode WM of the waveform suitable for the tone generation of the calculation mode. That is, for example, when the calculation mode CM (p) = 0, the waveform data whose waveform mode WM is “0” is selected. However, the calculation mode CM (p) and the waveform mode WM
Is a different concept, and does not always match. In the present embodiment, the waveform data is simply generated so that the calculation mode CM (p) and the waveform mode WM match.

In a succeeding step S24, the tone color data PDp of the part p (that is, the tone color data indicated by the tone color number TC (p)) is processed according to the pitch SP and the operation mode CM (p), and together with the event occurrence time TM, The sound source register of the assigned channel i is set in a predetermined area (step S24). Here, the tone color data PDp is processed 1
One purpose is to prevent the time-deformed state of the volume EG control data for controlling the volume envelope generator (not shown) and the cutoff frequency of the tone color filter (not shown) from changing in accordance with the operation mode CM (p). This is because
Another purpose is to change the musical tone characteristics such as the envelope shape in accordance with the performance information such as the pitch SP, as is the case with ordinary electronic musical instruments. The tone color number TC (p) is set by the user as described later with reference to FIG.

Next, note-on data is written in the sound source register of the assigned channel i (step S25), and this processing ends.

FIGS. 8 to 11 are flowcharts showing the procedure of processing various events that occur when the user presses the panel switch.
This is a process of the panel switch event process of FIG. Here, the panel switch may be one assigned in advance to the keyboard 2 or one displayed on the display 3. When the panel switch is the one displayed on the display 3, the cursor may be moved by the up / down key of the keyboard 2 or a mouse (not shown) and the desired panel switch may be pressed.

FIG. 8 is a flow chart showing the procedure of part tone color selection processing when a part tone color selection switch (not shown) is pressed.

In the figure, when the user inputs a part number, the part number is stored in the area p, and when the user inputs a tone color number, the tone color number is stored in the area TC (reserved in the RAM 6). p) (step S31). The tone color number stored in the area TC (p) is the tone color number TC (p).

Next, timbre data is prepared (step S
32). Specifically, the tone color number TC is selected from the tone color data group stored in advance in a predetermined area of the hard disk 4.
The tone color data indicated by (p) is searched and loaded into the tone color data area indicated by part p (see FIG. 2A). Also, the waveform designation data W in the loaded tone color data
Referring to N (p), the data WN (p) and data CM
It is determined whether or not the waveform data designated by (p) exists in the waveform data storage area of FIG. 2 (b), and if it does not exist, the waveform data is automatically stored in the waveform data storage area from the hard disk 4. To load. At that time, the tone color data originally set in the part p is saved in the corresponding storage area of the hard disk 4.

FIG. 9 is a flow chart showing the procedure of the part mode selection process when the part mode selection switch (not shown) is pressed.

First, as in step S31, the part number input by the user is stored in the area p, and when the user inputs the operation mode, the operation mode (0
2 to 2 are stored in the area CM (p) in the tone color data area indicated by part p (step S41).

FIG. 10 is a flow chart showing the procedure of part waveform selection processing when a part waveform selection switch (not shown) is pressed.

First, as in step S31, the part number input by the user is stored in the area p. Then, when the user inputs a waveform name, the waveform name is stored in the part p.
It is stored in the area WN (p) in the tone color data area indicated by (step S51).

Next, the waveform data corresponding to the operation mode CM (p) is prepared (step S52). In particular,
From the waveform data group stored in the predetermined area of the hard disk 4, the waveform name WN considering the operation mode CM (p)
The waveform data shown in FIG.
Waveform data area (see FIG. 2B).

Here, "in consideration of the calculation mode CM (p)" means that there is a plurality of waveform data having the same waveform name WN (p) and the waveform mode WM corresponding to the calculation mode CM (p). If there is waveform data, the waveform mode WM
This means that only the waveform data of is loaded into the waveform data area, and by doing so, the RAM 6 can be used efficiently. However, the present invention is not limited to this, and regardless of whether or not there are a plurality of waveform data having the same waveform name WN (p) without considering the operation mode CM (p), the waveform name WN
You may make it load all the waveform data shown by (p) in a waveform data area.

If the same waveform name is already stored in the waveform data area of the RAM 6, this processing is not necessary, but reloading does not cause any problem.

FIG. 11 is a flowchart showing the procedure of the waveform LPF process when the waveform LPF switch (not shown) is pressed.

First, when the user inputs a waveform to be processed and its processing contents, the waveform to be processed is retrieved from the waveform data area of the RAM 6 (step S61). At this time, if the waveform to be processed is not searched, the waveform data group of the hard disk 4 is searched. Further, the processing content refers to the type of downsampling, the band to be limited, and the like.

Next, the waveform retrieved in this way is
It is determined whether or not the loop portion that is repeatedly read is included (step S62). If the loop portion is not included,
First, band limitation is performed (step S63), then down-sampling is performed (step S64), the waveform is completed, and R
Store in the waveform data area of AM6 (step S6)
5). The reason why the band is limited in step S63 is that if the down-sampled waveform data is stored without being band-limited, aliasing noise is mixed into the stored waveform data. Therefore, it is necessary to limit the band so that aliasing noise does not occur in the downsampling, and the frequency band to be limited is set by the user in step S61.

FIG. 12 shows steps S63 and S6.
4 is a diagram for explaining the process of No. 4, in which the horizontal axis represents frequency and the vertical axis represents level. And (a)
Is an original waveform (waveform name: PIAN) that is sampled at a predetermined recording sampling frequency Fs (that is, in waveform mode WM = 0) before performing the waveform LPF processing, that is, for the tone generation calculation of the equivalent sampling frequency of 48 kHz.
O. 0) shows an example of the frequency spectrum of (0), and (b) is
(A) shows a frequency spectrum after band limitation processing is applied to the waveform of (a), and (c) shows an equivalent sampling frequency of 24 kHz obtained by down-sampling the waveform of (b) at half the original recording sampling frequency. Shows the frequency spectrum of the waveform data (waveform mode WM = 1) for the tone generation calculation of FIG.
9 shows a frequency spectrum of waveform data (waveform mode WM = 2) for tone generation calculation with an equivalent sampling frequency of 12 kHz, which is subjected to PF processing, that is, band limiting processing, and is down-sampled at a frequency half the recording sampling frequency. . Here, the downsampling of (c) is performed by skipping the waveform data of (b) one by one as described above with reference to FIG. Similarly, the downsampling of (d)
The waveform data of (c) is read by skipping the band-limited data one by one. Therefore, in the present embodiment, it is only possible to down-sample at a recording sampling frequency that is 2 ⁻ⁿ (n is a positive integer) times the original waveform, but by performing the processing of steps S63 and S64 at the same time, It is also possible to downsample at a frequency that is an arbitrary multiple of the sampling frequency of the waveform.

Returning to FIG. 11, in the determination of step S62,
If the retrieved waveform data has a loop portion, the waveform is first developed into waveform data in which a loop portion that repeats a plurality of times is connected to the attack portion (step S66), and then the bandwidth is limited (step S67), and then down. Sampling is performed (step S68), new attack parts and loop parts are cut out from the waveform data thus generated (step S69), and the waveform is completed and stored in the same manner as in step S65 (step S70). The processing of steps S67 and S68 is the same as the processing of steps S63 and S64, respectively. Here, the above processing is applied to a waveform having a loop portion in the case of waveform data including an attack portion and a loop portion,
Since the loop part has a small number of waveform data samples, when limiting the band as it is for the original waveform,
This is because a high-order low-pass filter cannot be used and a sufficient band attenuation characteristic cannot be obtained. According to the above method, since even a waveform data having a loop portion can use a high-order low-pass filter, LPF processing with less unnecessary noise can be performed.

In the following step S71, the above-mentioned step S
65 or the waveform data stored in S70 is used as the RA
After the M6 or the predetermined area of the hard disk 4 is stored (registered), the waveform LPF processing is ended.

FIG. 13 is a flow chart showing the detailed procedure of the sound source processing subroutine of step S6 of FIG.

First, in step S81, the sound source register (see FIG.
(D)) is checked, and it is determined in the subsequent step S82 whether or not there is new writing. If there is no new writing, the process immediately proceeds to step S84, while if there is new writing, a sound source control preparation process for the tone generation channel for which the writing has been performed is performed (step S83). The sound source control preparation process is, specifically, channel i (sound channel that has been newly written).
Is a process of converting the data of 1 to various control data for actually performing the waveform calculation, and setting the first read address of the waveform data corresponding to the data of the channel i.

In the following step S84, the calculation time is managed. In other words, the time at which the reading of the waveform data currently being reproduced is started is set as the calculation start time tBC (see FIG. 4) of the next waveform data so that the reading of the reproduced waveform data by the reproducing unit (DMA control unit 9) is not interrupted. specify. Further, in step S85, it is determined whether or not the calculation start time tBC has come, and if not, the present process is immediately ended.

When the calculation start time comes, first, channel control is performed to determine the calculation order and the channels to be muted according to the musical sound to be generated by each sound generation channel (step S86). The calculation order is determined in consideration of the case where the calculation is not finished by the time when the calculation should be finished (the calculation is discontinued at this time), and the calculation with the higher importance is processed first. Next, the data prepared in step S83 is expanded on the time axis to prepare for waveform calculation (step S87), and the tone generation channel number in the calculation order 1 is set as the parameter i (step S8).
8).

Next, in step S89, a calculation mode CM
The value of (i) is discriminated. When CM (i) = 0, 128 samples of tone waveform data of channel i are generated and added to the buffer 0 (step S90), and CM (i)
When = 1, musical tone waveform data for 64 samples of channel i is generated and added to the buffer 1 (step S
91), and when CM (i) = 2, 32 of channel i
The musical tone waveform data for the sample is generated and added to the buffer 2 (step S92). Here, each step S90
The processing of 92 to 92 is not a simple addition of the musical tone data (waveform sample data) in the corresponding buffers, but an update control of the read address according to the F number FN (see FIG. 2D) of each channel i, The waveform sample data is read out from the waveform data storage area shown in FIG. 2B and interpolated according to the read address, and the waveform sample obtained by the interpolation is further processed such as tone color and volume envelope. It is the addition after performing the processing such as making a musical tone waveform sample. The above process is repeated for each sample of the musical tone waveform data generated in one generation process. For example,
In the channel i of the operation mode CM (i) = 0, 128 musical tone waveform samples are repeated until they are completed. Here, the waveform data read out is the waveform data designated by the above-mentioned waveform designation data. The waveform data is selected and designated in correspondence with the waveform name WN (p) of the part p to which the musical sound being generated in the tone generation channel i belongs and the calculation mode CM (i). If the recording sampling frequency of the waveform data used for the tone signal currently being generated and the calculation mode (equivalent sampling frequency) set for channel i are not compatible, the above-mentioned F number FN The value of the F number FN is corrected by the term in the latter half of the calculation formula, in order to generate the musical sound of the instructed pitch even in such a case. For details of the processing in steps S90 to S92, refer to Japanese Patent Application No. 7-
See FIG. 13 of the drawings of 197923 and the description relating to that drawing of the specification.

As described above, even if the optimum waveform mode waveform is not selected for each operation mode, the F number F
By correcting N, a musical tone waveform sample with a specified pitch SP can be generated in any of the calculation modes.
If the numerical value of the F number FN greatly deviates from the reference value "1" due to the correction or the like, there arises a problem in the quality of the generated musical sound. For example, the F
If the number FN becomes very large (for example, FN
> 2), the high frequency component of the waveform data read from the waveform data storage area exceeds a frequency equal to or more than half the equivalent sampling frequency, and the aliasing noise occurs in the waveform sample obtained by interpolation. Conversely, if the correction causes the F number FN to be very small (eg, FN <0.5),
The waveform sample obtained by the interpolation contains only a harmonic component equal to or less than a quarter of the equivalent sampling frequency, and the quality corresponding to the time density of the musical tone waveform sample generation set for that channel cannot be obtained. Therefore,
The waveform data indicated by the waveform designation data, that is, the waveform data read in the tone generation of each channel i, is the waveform data of the waveform mode suitable for the calculation mode CM (i) of each channel (for example, if CM = 1, the waveform of WM = 1). It is desirable to select and set (data).

Next, it is judged whether or not the channel i is the final channel, that is, whether or not the calculation of all the waveform calculation channels is completed (step S93). The channel number in the calculation order of is set to channel i (step S9
4) After that, it returns to step S89 and repeats the above-mentioned processing. On the other hand, if it is determined in step S93 that all the channels to be calculated are completed, the process proceeds to step S95.

In step S95, as described with reference to FIG. 3, the musical tone waveform data of 64 samples generated in the buffer 1 and 32 samples generated in the buffer 2 are interpolated (oversampled), respectively.
The musical tone waveform data of 128 samples is stored in the buffer 1'and the buffer 2 ', and in step S96, the musical tone waveform of 128 samples obtained by adding the 128 musical tone waveform data of each of the buffers 0, 1', and 2 '. Store the data in buffer 0.

In a succeeding step S97, the musical tone waveform data thus stored in the buffer 0 is subjected to reverb processing to give a reverberation effect, and in a step S98, 128 samples of the buffer 0 are reserved for reproduction in the reproducing section. ,
This sound source processing ends.

As described above, in the present embodiment, a plurality of calculation modes of the calculation process for generating a musical sound are provided, and the user can arbitrarily select from among them, so that the user can select the desired mode according to the purpose of use of the user. It is possible to take either form of emphasis on the number of pronunciations or quality.

Further, since the calculation mode is set for each part, it is possible to generate a high-quality musical sound of a part having a great auditory effect, and to make maximum use of the limited calculation ability.

Furthermore, from the waveform data of the high recording sampling frequency for the calculation mode (mode 0) having a high quality of the generated tone, the waveform of the low recording sampling frequency corresponding to the calculation mode (mode 1 or 2) having a low quality of the generated tone is generated. The data is created by the LPF process, and the waveform data is created from the waveform data created in this way according to the operation mode selected by the user to generate a musical sound. It is possible to generate a waveform having the same tone color as a musical tone calculated at a high equivalent sampling frequency without causing aliasing noise in the channel for which the calculation is performed, although the quality is deteriorated.

Further, since the waveform data is automatically selected according to the calculation mode selected by the user, the waveform data having the frequency component over a wide band of the high recording sampling frequency can be obtained in the channel of the high equivalent sampling frequency. Waveform data having a narrow band frequency component with a low recording sampling frequency can be used in a channel with a low equivalent sampling frequency, and thus the waveform designation in the tone color data does not have to be changed.

In this embodiment, when the operation mode CM (p) having a smaller number of samples is selected and there is no waveform data having the waveform mode WM corresponding to the operation mode CM (p), Is the calculation mode CM
Although the waveform data of the waveform mode WM different from (p) is selected and the musical tone signal is generated based on this waveform data, the configuration is not limited to this, and the selected operation mode CM is used.
It is automatically detected whether or not the waveform data of the waveform mode WM corresponding to (p) is present.
It is also possible to automatically shift to the waveform LPF processing of No. 1 to create the waveform data of the operation mode CM (p), and to generate the tone signal based on the created waveform data.

Further, in this embodiment, the waveform data WD
Waveform LPF processing is performed for n (n = 1, 2 ...) And waveform data WDn ′, which has a recording sampling frequency lower than that,
WDn ″ was obtained, but this waveform data WD
n ′ and WDn ″ may be prepared by a method other than the waveform LPF processing. For example, if the recording is performed at a low recording sampling frequency from the beginning, the waveform data WDn ′ and WDn ″ can be directly obtained. The hard disk 4 stores not only the waveform data WDn but also the waveform data W
Dn ′ and WDn ″ are also stored. When the waveforms are stored in the hard disk 4, the waveform data WD
Even if the waveform LPF process is not performed on n, the waveform data WDn ′,
WDn ″ can be read from the hard disk to the RAM 6.

In this embodiment, the calculation mode CM
Only one (p) can be selected for each part,
The present invention is not limited to this, and even within a plurality of channels that sound the same part, the calculation mode may be automatically changed according to the volume or pitch of each channel.

Further, in the present embodiment, the invention is realized as software executed by the CPU, but the invention is not limited to this. For example, a DSP (digital signal processor)
It can be realized by a general arithmetic device that operates with a program such as. Further, the musical tone generating apparatus according to the present embodiment may be realized by a general-purpose personal computer or a dedicated device.

[0091]

As described above, claims 1 and 5
According to the present invention described above, a musical tone waveform sample of a channel corresponding to performance information input within the predetermined period is generated for each predetermined period at a number of samples corresponding to the input control information. Since a musical tone is generated based on the generated musical tone waveform sample, there is an effect that it is possible to adopt either the number of pronunciations emphasis or the quality emphasis according to the purpose of use of the user.

According to the inventions of claims 2 and 6, the musical tone waveform sample of the channel corresponding to the input performance information corresponds to the set control information of the part to which the performance information belongs. Are generated at different time densities, and musical tones are generated based on the generated musical tone waveform samples, so that musical tones of a part that have a great auditory effect can be generated with high quality, and the limited computing power is utilized to the maximum. be able to.

According to the third and seventh aspects of the invention, the musical tone generation operation based on the waveform data stored in the waveform memory is executed according to the input performance information, and the generated control is executed. Musical tone waveform samples are generated at a time density according to information, and in the musical tone generation calculation, musical tone generation calculation is performed by selectively using different waveform data in the waveform memory according to the generated control information. Since a musical tone is generated based on the musical tone waveform sample thus generated, waveform data having a frequency component over a wide band is used for a channel with a high sampling frequency, and a narrow band frequency is used for a channel with a low sampling frequency. Waveform data having components can be used so that the designation of the waveform in the timbre data does not have to be changed.

Further, according to the configuration of the invention described in claim 4, since the waveform data corresponding to the generation time density of the musical tone waveform sample is used, the aliasing noise can be reduced. Furthermore, when a musical tone waveform sample is generated at two different time densities, the waveform data obtained by converting the waveform data suitable for one of the time densities is used for the waveform generation of the other time density, so that even if the time densities are different, It is possible to generate musical tone waveform samples that sound the same timbre.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a musical sound generating apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of tone color data and waveform data stored in a RAM of FIG. 1 and configurations of an input buffer and a tone generator register set on the RAM.

FIG. 3 is a diagram showing a configuration of an output buffer set on the RAM of FIG.

FIG. 4 is a diagram for explaining an outline of a musical sound generation process performed by the musical sound generation device of FIG.

5 is a flowchart of a main routine executed by the CPU of FIG.

FIG. 6 is a flowchart of MIDI reception interrupt processing.

7 is a flowchart of note-on event processing when data is received via the MIDI interface of FIG. 1. FIG.

FIG. 8 is a flowchart showing a procedure of part tone color selection processing when a part tone color selection switch is pressed.

FIG. 9 is a flowchart showing a procedure of part mode selection processing when a part mode selection switch is pressed.

FIG. 10 is a flowchart showing a procedure of part waveform selection processing when a part waveform selection switch is pressed.

FIG. 11 is a waveform L when the waveform LPF switch is pressed.
It is a flow chart which shows the procedure of PF processing.

12 is a diagram showing an example of frequency characteristics of waveform data generated by the waveform LPF processing of FIG.

FIG. 13 is a flowchart showing a detailed procedure of a sound source processing subroutine of step S6 of FIG.

[Explanation of symbols]

1 CPU (control information input means, musical tone waveform sample generation means, control information generation means) 2 Keyboard (performance information input means, operator, control information setting means) 5 ROM 6 RAM (waveform memory, storage means) 8 MIDI interface ( Performance information input means) 9 Direct memory access control section (musical sound generation means) 10 DA converter (musical sound generation means)

Claims (7)

[Claims] 1. A musical tone generating method including a step of generating a musical tone based on a musical tone waveform sample generated by a plurality of channels, the method comprising inputting performance information and an operation amount of an operator operated by a user. The step of inputting control information, and for each predetermined period, the musical tone waveform samples of the channels corresponding to the input performance information within the predetermined period are sampled according to the number of samples according to the input control information. And a step of generating for each channel, wherein a musical sound is generated based on the generated musical tone waveform sample. 2. A musical tone generating method including a step of generating a musical tone based on musical tone waveform samples generated by a plurality of channels, wherein the performance information of a plurality of parts is input, and the control information of each part is set. Assigning the input performance information to any one of the plurality of channels, and setting a musical tone waveform sample to which the performance information belongs to the assigned channel to the part to which the performance information belongs. Generating a musical tone based on the generated musical tone waveform sample, the musical tone generating method comprising: generating a musical tone based on the generated musical tone waveform sample. 3. A musical tone generating method comprising a step of generating a musical tone based on a musical tone waveform sample, a step of inputting performance information, a step of generating control information, and a waveform according to the input performance information. Performing a tone generation calculation based on the waveform data stored in the memory, and generating tone waveform samples at a time density according to the generated control information. According to the generated control information, different waveform data in the waveform memory is selectively used to perform a musical tone generation operation, and in the step of generating the musical tone, a musical tone is generated based on the musical tone waveform sample generated by the musical tone generation operation. A method for generating a musical sound characterized by generating a. 4. A step of storing, in a storage means, first waveform data for generating a musical sound at a predetermined sampling frequency, and a sampling frequency different from the predetermined sampling frequency of the stored first waveform data. Of the waveform data stored in the storage means and the step of generating a musical tone waveform sample based on the waveform data stored in the storage means. In the step of generating, the musical tone waveform sample can be generated at a plurality of different time densities, and one of the first waveform data and the second waveform data is selected according to the time density. A method for generating a musical tone waveform, which comprises generating a musical tone waveform sample based on the selected waveform data. 5. A musical tone generating apparatus having musical tone generating means for generating musical tones based on musical tone waveform samples generated by a plurality of channels, wherein the musical performance information input means for inputting musical performance information and the operation amount of the manipulator are used. Control information input means for inputting control information in accordance with the number of samples of the musical tone waveform sample of the channel corresponding to the input performance information within the predetermined period for each predetermined period. And a musical tone waveform sample generating means for generating each channel, wherein the pre-musical tone generating means generates a musical tone based on the generated musical tone waveform sample. 6. A musical tone generating apparatus comprising musical tone generating means for generating musical tones based on musical tone waveform samples generated by a plurality of channels, and musical performance information input means for inputting musical performance information of a plurality of parts, and each of said parts. Control information setting means for setting the control information, and assigning the input performance information to any one of the plurality of channels, and assigning the musical tone waveform sample to which the performance information belongs to the assigned channel to the performance. And a musical tone waveform sample generating means for generating at a time density according to the set control information of the part to which the information belongs, and the musical tone generating means generates a musical tone based on the generated musical tone waveform sample. Characteristic tone generator. 7. A musical tone generating apparatus having musical tone generating means for generating musical tones based on musical tone waveform samples, performance information inputting means for inputting performance information, control information generating means for generating control information, and said inputting. A musical tone waveform sample generation means for performing musical tone generation calculation based on the waveform data stored in the waveform memory according to the performance information, and generating musical tone waveform samples at a time density according to the generated control information. The musical tone waveform sample generating means selectively uses different waveform data in the waveform memory in accordance with the generated control information to perform a musical tone generating operation, and the musical tone generating means generates the musical tone generating operation. A musical tone generating apparatus which generates a musical tone based on a musical tone waveform sample.