JP3141789B2 - Sound source system using computer software - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sound source system and a sound generator which combine musical tone waveform generating blocks described in software to generate musical tone waveform data based on musical tone waveform generation operations performed by the respective tone waveform generating blocks. The present invention relates to a waveform data generation method.

[0002]

2. Description of the Related Art Conventionally, for example, a waveform memory sound source system or F
When a tone is generated in accordance with various tone generation methods such as the M tone generator method, a circuit for implementing the tone generation method is provided by a dedicated hardware (for example, a DSP (Digital) which operates based on a tone generator dedicated LSI or a fixed microprogram.
Signal Processor) etc.). in this way,
The tone generator and the tone generation method constituted by dedicated hardware are hereinafter referred to as “hard sound source”.

However, a hardware tone generator requires dedicated hardware, so that it is difficult to further reduce manufacturing costs, and it is not possible to flexibly cope with a specification change after design. Was.

[0004] On the other hand, in recent years, with the improvement of the computational power of CPUs, a general-purpose computer or a CPU mounted on a dedicated tone generator has to execute a software program describing a predetermined tone generation processing procedure. As a result, musical tone waveform data has been developed. The tone generator and the tone generation method mainly composed of software programs are hereinafter referred to as “soft sound source”.

[0005]

As described in the "prior art" described above, the use of a hard tone generator in a computer system or a computer-applied device causes an increase in cost and a lack of flexibility in changing specifications. There were problems such as becoming.

[0006] In addition, also entered the soft sound source, the only of the software sound source was directly simply soft c E A the function of the dedicated hardware and devices, such as a sound source LSI that has been conventionally used, the specification change after design Although it can respond more flexibly than the hardware sound source, it can flexibly respond to various requests generated during sound generation, that is, during operation of the sound source, for example, requests from the capabilities of the CPU and the system environment, and requests for user preferences and settings. could not. Specifically, a change in the accuracy of the output musical tone waveform (including a change to a high accuracy as well as a change to a low accuracy) and a change in the degree of the timbre change (for example, a timbre that is more subtle than a normal timbre change) Change, or change the tone color in the opposite direction).

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and reduces the cost by generating a musical tone by using a software program without adding special dedicated hardware. Sound source system using computer software capable of changing the load on the arithmetic means for the arithmetic processing of musical tone generation and the quality of the output musical tone waveform in accordance with the setting and the like
The purpose is to provide a system.

[0008]

According to a first aspect of the present invention, there is provided a tone generator system comprising: a plurality of software-generated tone waveform generation blocks for performing a tone waveform generation operation; And an algorithm setting means for setting an algorithm for determining the number of musical tone waveform generating blocks to be combined and the combination mode for each tone color in a tone generator system that generates musical tone waveform data of different timbres for each of the sound channels. , Set the tone for each performance part of the song
Tone setting means, and the combination
Block to set the number of musical tone waveform generation blocks to be used
The tone color is set by the number setting means and the tone color setting means.
When the performance part is
When the number of musical tone waveform generation blocks is set
The tone set for the performance part
To a tone with a tone waveform generation block within the range of
Tone changing means for changing, sounding channel selecting means for selecting a channel for generating musical tone waveform data from the plurality of sounding channels, and
A tone waveform generation block indicated by the set algorithm is assigned according to the tone color of the tone waveform data to be generated, and the assigned tone waveform generation block is combined in a combination mode indicated by the algorithm to perform a tone waveform generation operation. And a musical sound waveform data generating means for generating musical sound waveform data.

Preferably, in said sounding channel
During the tone waveform generation calculation, the sound
Number of musical tone waveform generation blocks assigned to channels
It has a block number changing means for changing the number .

[0010] Preferably, each of the tone waveform generation Bed
Tone generated by each tone waveform generation block for each lock
Set the sampling frequency that is the reference for the waveform generation calculation
Sampling frequency setting means,
The data generating means is a music data generating block for each musical tone waveform.
The waveform generation operation is set in each tone waveform generation block.
The sampling is performed with reference to the sampling frequency .

[0011] To achieve the above object, the present invention is described in claim 4.
Is a software that performs tone waveform generation calculation.
Combination of multiple tone waveform generation blocks
This allows you to select each channel of multiple sound channels.
In a sound source system that generates musical sound waveform data for each file
The number of musical tone waveform generating blocks to be combined
And the algorithm that determines the combination
Algorithm setting means, and the plurality of sound channels
To select a channel for generating musical sound waveform data from
Sound channel selecting means, and the selected sounding channel
The tone waveform generation block indicated by the set algorithm
Assigns a lock and assigns the assigned tone waveform generation block
In the combination mode indicated by the algorithm.
Performs tone waveform generation calculation to generate tone waveform data
Musical tone waveform data generating means,
During the musical sound waveform generation calculation in accordance with predetermined conditions.
Musical sound waveform generation block assigned to the sound channel
Block number changing means for changing the number of blocks .

Preferably, each of the musical tone waveform generating blocks is provided.
Tone generated by each tone waveform generation block for each lock
Set the sampling frequency that is the reference for the waveform generation calculation
Sampling frequency setting means,
The data generating means is a music data generating block for each musical tone waveform.
The waveform generation operation is set in each tone waveform generation block.
The sampling is performed with reference to the sampling frequency .

[0013] In order to achieve the above object, a sixth aspect of the present invention is provided.
Is a software that performs tone waveform generation calculation.
Combination of multiple tone waveform generation blocks
This allows you to select each channel of multiple sound channels.
In a sound source system that generates musical sound waveform data for each file
The number of musical tone waveform generating blocks to be combined
And the algorithm that determines the combination
Algorithm setting means, and the tone waveform generation block.
The tone waveform generated by each tone waveform generating block
A sample frequency setting reference for generation calculation
Sampling frequency setting means, and the plurality of sounding channels.
Select a channel to generate musical waveform data from
Sounding channel selecting means, and the selected sounding channel
The sound waveform generation shown by the set algorithm
Allocate a block and assign the assigned musical tone waveform generation block.
Locks are combined in the combination indicated by the algorithm.
The sump set in each musical tone waveform generation block
Performs a tone waveform generation operation based on the ring frequency and
Music sound waveform data generating means for generating sound waveform data;
Characterized in that it.

[0014] In addition, good Mashiku, the algorithm setting
Means for setting the algorithm for each timbre,
The waveform data generating means includes:
Contains the sound of the tone waveform data to be generated in the channel.
Music sound wave indicated by the set algorithm corresponding to the color
The method is characterized in that a shape generation block is allocated .

Further, for each performance part of the music,
The algorithm corresponding to the tone set for each
Reduction mode for setting the reduction mode of the musical tone waveform generation block shown
Setting means, wherein the number-of-blocks changing means includes
Tone waveform according to the reduced mode and the predetermined condition
The number of generated blocks is changed .

Preferably, the predetermined condition is:
Volume of the tone waveform generated in the sound channel
The envelope was attenuated, and the number of blocks was changed.
Means for controlling the volume envelope of the musical sound waveform to be equal to or less than a predetermined value.
When the decay, according to the decay, the algorithm
Musical sound waveform indicated by the algorithm set by the setting means
It is characterized in that the number of generated blocks is changed so as to decrease .

Further, the predetermined condition is that
The volume of the tone waveform to be generated in the
The number of blocks changing means,
When the volume of the tone waveform is set low,
The algorithm set by the algorithm setting means
In the direction to decrease the number of
Characterized in that it further.

Further, the predetermined condition is that
During the tone waveform generation block assigned to the sound
There is a block whose output level is lower than the specified value
Wherein the block number changing means has a predetermined output level.
The tone waveform generation block whose value is less than or equal to the other tone waveform generator
If it does not affect the building block,
By stopping the operation of the tone waveform generation block,
Reduce the number of musical tone waveform generation blocks indicated by the algorithm
It is characterized in that it is changed in a direction to decrease .

[0019]

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

FIG. 1 is a diagram showing an example of a system configuration of software provided in a sound source system according to an embodiment of the present invention. As shown in FIG. That is, the tone waveform data is generated on the OS side. In the figure,
DAC (Digital to Tone) that converts digital tone signals (tone waveform data) generated on the OS side into analog tone signals
Codec hardware including an Analog Converter) is also shown.

In FIG. 1, APS1 is application software (hereinafter referred to as "sequence software") for generating MIDI messages, that is, performance information on which musical tone waveform data is generated, in real time and sequentially. The sequence software APS1 is
A plurality of MIDI files (for example, a file including various event data and MIDI data such as timing data for generating the event) prepared in advance for each music are provided.
When an IDI file is selected, M
The IDI data is read, and a MIDI message (a MIDI message generated corresponding to the event data) is generated in real time and sequentially. And
The sequence software APS1 converts the MIDI message into a first interface IF1 (MIDI APplication I) provided on the OS side for inputting the MIDI message.
nterface; MIDI API).

The OS has a first interface IF1
A software sound source module SSM, which is a program (software) for generating musical tone waveform data based on a MIDI message input via the PC, is installed as a driver, and the musical tone waveform data generated by the software tone source module SSM is converted to a software tone generator. Second interface IF2 (WAVE
out APplication Interface; WAVE out API). Further, the OS has a second interface I
An output device OUD for outputting the musical tone waveform data input via F2 to the outside is also installed as a driver. The output device OUD is generated by, for example, processing of the software sound source module SSM and stored in a hard disk or the like. Direct access to the tone waveform data stored in the buffer
MA) It is configured by software that performs processing to read out via a controller and output to predetermined hardware or a device (for example, the codec hardware).

The MI output by the sequence software APS1
The DI message is supplied to the input interface of the software tone generator module SSM via the first interface IF1 and the OS, and the software tone generator module SSM generates a musical sound waveform data based on the received MIDI message (in this embodiment, the musical tone waveform data is generated). The waveform data is generated by the FM sound source method), and the generated musical sound waveform data is output to the output device OUD via the second interface IF2 and the OS.
Supplied to In the output device OUD, the supplied generated digital musical tone waveform data is output to the codec hardware and converted into an analog musical tone signal.

As described above, in this embodiment, the software sound source module S for generating musical tone waveform data is used.
Since the SM and the sequence software APS1, which is application software for outputting a MIDI message, can be easily combined at the OS level, there is no need to add dedicated hardware for generating musical tone waveform data. Thus, the cost can be reduced.

FIG. 2 is a diagram showing a schematic configuration of hardware for realizing the tone generator system of the present embodiment. This system is realized by a general-purpose personal computer, and a CPU 3 is used as a main control unit. Under the control of the CPU 3, a tone waveform data generation process by a software sound source program and other processes other than the tone waveform data generation process are performed. Is executed in parallel with the program processing.

In FIG. 1, a CPU 3 has a MIDI interface MIDI for inputting an external MIDI message and outputting a MIDI message to the outside.
DII / F1, timer 2 for counting timer interrupt time and various times, ROM (read only memory) 4 for storing various control programs and table data, etc., and the selected MIDI file, various input information, calculation results, etc. RAM (random access memory) 5 for temporarily storing,
Mouse 7, which is a pointing device, keyboard 8, mainly for inputting character information, for example, large LC
A display 9 for displaying various information such as D and CRT,
A hard disk device 10 for driving a hard disk for storing various application programs including various control programs executed by the CPU 3 and various data;
An MA (Direct Memory Access) controller 11 and a communication interface (I / F) 14 for transmitting / receiving data to / from a server computer 102 via a communication network 101 are connected via a data / address bus 6. .

The DMA controller 11 converts the tone data generated by executing the tone generating process and written into the output buffer in the RAM 5 directly according to the empty state of a data buffer (not shown) included in the DAC 12. A process (reproduction process) of directly reading from the output buffer and transferring the data to the internal data buffer of the DAC 12 by the memory access method is executed. The tone data analog-converted by the DAC 12 is sent to the sound system 13 and converted into sound by the system 13.

The hard disk in the hard disk device 10 includes software for realizing a software sound source (ie, the software sound source module SS) in addition to the software such as the OS and the utility software.
M) and other application software (that is, the sequence software APS1).

The output device OUD converts the tone data supplied from the software tone generator module SSM via the second interface IF2 at the OS level into D.
This corresponds to a module to be sent to AC12. As aforementioned,
Since the DMA controller 11 sends the tone data to the DAC 12 by the direct memory access method, the output device OUD is controlled by the CPU 3 under the control of the CPU 3.
This is executed as an interrupt process by the MA controller 11.

The communication I / F 14 is connected to a communication network 101 such as a LAN (local area network), the Internet, and a telephone line, and is connected to the server computer 102 via the communication network 101. If the above programs and various parameters are not stored in the hard disk in the hard disk device 10, the communication I / F 14
2 to download programs and parameters. The computer serving as the client (sound source system of the present embodiment) is connected to the server computer 102 via the communication I / F 14 and the communication network 101.
Send a command to request the download of programs and parameters. The server computer 102 receives the command, distributes the requested program or parameter to the computer via the communication network 101, and the computer receives the program or parameter via the communication I / F 101, and The download is completed by accumulating the data in the hard disk in 10.

In addition, an interface for directly exchanging data with an external computer or the like may be provided.

Next, the software sound source module SSM
An outline of a musical sound generation process using the FM sound source method according to the first embodiment will be described with reference to FIGS.

When the sequence software APS1 is started, a MIDI message starts to be supplied to the software sound source module SSM. This supply is made to the input interface of the software tone generator module SSM (hereinafter, referred to as “soft tone generator interface”) via the first interface IF1 and the OS as described above. The SSM generates a tone parameter VOICEj (the details of which will be described later) based on the voice data (tone color number in the present embodiment) assigned to the MIDI channel (performance part) of the supplied MIDI message, The musical tone parameter VOICEj is stored in a tone color register corresponding to a tone generation channel (channel to which tone generation is assigned).

FIG. 6 is a diagram showing a tone color register group provided corresponding to a sounding channel. When 32 sounding channels are secured, for example,
This tone register group consists of 32 tone registers TONEP
ARk (k = 1,..., 32). In addition,
The number of sound channels need not be limited to this.
Can be set arbitrarily in accordance with the arithmetic processing capability of.

In the figure, if the tone generation channel is channel n, for example, the tone parameter VO
ICEj is stored in the tone color register TONEPARn in an area for storing the tone parameter VOICEj.
In other words, the tone color register group constituted by these tone color registers TONEPARk forms a part of the software tone generator interface of the software tone generator module SSM.

As described later, each tone color register T
ONEPARk includes not only the tone parameter VOICEk but also, for example, the software tone generator module SS
M is the MID corresponding to the musical tone parameter VOICEk
Data TM indicating the time at which the I message was received is also stored, and serves as information for determining the key-on / key-off of a musical tone in a frame or the start and end time positions of various controls.

Returning to FIG. 3, under the control of the CPU 3, the software sound source module SSM basically responds to an activation opportunity set for each section of a predetermined time length (hereinafter, referred to as “frame”). The tone generation processing based on the MIDI message that is activated and supplied in the immediately preceding frame is executed in accordance with the tone parameter VOICEn stored in the tone color register TONEPARn. For example, in FIG. 3, the tone generation process based on the MIDI message supplied in the frame from time t1 to t2 is executed in the frame from time t2 to t3.

As described above, when the musical sound waveform data for one frame is generated by the musical sound generation processing (the processing method will be described later), the musical sound waveform data
M5 is written to the output buffer, and reproduction of the written data is reserved in the output device OUD.
Reserving this output device OUD means that the generated tone waveform data is transmitted from the software sound source module SSM to the second interface (“WAVE out”) at the OS level.
API ”).

The output device OUD is, for each frame,
The tone waveform data is read out one sample at a time from the output buffer reserved for reproduction in the immediately preceding frame, and read from the DAC.
12 is output. For example, as shown in FIG.
The tone waveform data generated in the frames from 2 to t3, written in the output buffer, and reserved for reproduction is stored at time t
It is read out and reproduced in the frames from 3 to t4.

Next, referring to FIG.
An outline of a tone generation process according to ICEn, that is, a tone generation process using the FM sound source system will be described. The details of the tone generation process will be described later with reference to FIGS.

FIGS. 4A to 4C are diagrams for explaining the outline of the tone generation processing by the FM sound source system.
3 shows three different tone generation processing methods.

As shown in FIG. 4, the tone generation by the FM tone generator system is performed by combining two types of operators, an operator called a "carrier" and an operator called a "modulator". The number of operators to be combined and the order of connection (connection mode) differ according to the quality and quality. The way of connecting these operators is called an "algorithm".

Here, the operator refers to a block which is a unit of sound creation, that is, a tone generation process. Specifically, for example, various waveform data (hereinafter, referred to as "basic waveforms") shown in FIG. Data), one basic waveform data selected according to a wave select parameter WSEL, which will be described later, is converted into two types of input data (for example, pitch data and modulation data) based on input data (eg, pitch data and modulation data). In the case of the data, it refers to a processing block that reads out (for example, based on an addition result obtained by adding the respective data), adjusts its amplitude, and outputs it. The “carrier” refers to an operator that generates a tone waveform that is a reference of the tone waveform to be generated, and the “modulator” is an operator for modulating the “carrier”, that is, input to the “carrier”. An operator who generates the modulation data.

It goes without saying that the algorithm is not limited to the three types shown in FIG.

Next, the data format of the tone parameter VOICEj will be described.

FIG. 7 shows this musical tone parameter VOICEj.
FIG. 7A is a diagram showing a data format of a musical tone parameter VOICEj;
(B) shows each operator data OPmDATA of (a).
j shows the data format of (j), and (c) shows the data format of each operator buffer OPBUFm of (b).

As shown in (a), the tone parameter VO
ICEj sets key on / off to “1” / “0”, respectively.
Key-on data KEYONj, a frequency number (specifically, represented by a phase rate) FNOj determined according to pitch information included in the MIDI message (that is, a corresponding note-on event), Algorithm specification data ALGO that specifies the algorithm
Rj, volume data VOLj determined according to the volume set for the MIDI channel (for example, set by the “Control Change # 7 Event” of the MIDI message), and touch velocity information in the MIDI message. Velocity data VELj determined by the above, operator data OPkDA composed of a buffer for storing data required for performing tone generation calculation in each component operator and a result of the calculation.
TAj (k = 1,..., M).

The tone parameter VOICE of FIG.
j is the data obtained by reading data stored in advance in the ROM 4, RAM 5, hard disk, or the like, and MID.
2 of data determined according to the data in the I message
Types of data are mixed. As described above, the data determined according to the data in the MIDI message includes the key-on data KEYONj, the frequency number FNOj, the volume data VOLj, and the touch velocity data VEL.
j, and the data read from the ROM 4 or the like is the algorithm designation data ALGORj and the operator data OPkDATAj.

Each operator data OPkDATAj is
As shown in (b), the sampling frequency designation data FS for designating the sampling frequency in the operator m
AMPm and a parameter that is set corresponding to the tone color and is a parameter for setting a frequency ratio between operators substantially, specifically, a frequency multiple that is a parameter for specifying a multiple for changing the frequency number FNOj Data MULT
m and a feedback level, that is, feedback level data FBL indicating the degree of modulation of feedback modulation.
m, wave select data WSELm for selecting basic waveform data to be used by the operator m from various basic waveform data (the various basic waveform data described with reference to FIG. 5) stored in the ROM 4 or the like, and generated by the operator m. Output level of the musical tone waveform to be performed (this level changes according to the touch velocity data VELj)
And an envelope parameter EGPARm including various data (for example, various data such as an attack time, a decay time, a sustain level, and a release time) for determining an envelope of a musical sound waveform generated by the operator m.
And other parameters (eg, speed and depth of vibrato and tremolo, various key scaling factors, etc.)
MSCm, operator priority data OPPRIm indicating the priority of the operator m (for example, the priority of the waveform generation operation and its stop in each operator), and a buffer OPBUFm for storing the operation result of the tone waveform generation operation in the operator m. Be composed.

Here, the sampling frequency designation data F
SAMPm stores an integer value f equal to or greater than "0", and can change the sampling frequency FSMAX (for example, 44.1 kHz) in the standard mode to 2- ^f times. For example, when f = 0, a tone waveform at the operator m is generated at the sampling frequency FSMAX in the standard mode, and when f = 1, a tone waveform at the operator m is generated at a sampling frequency of FSMAX / 2.

The operator priority data OPPRI
In Om, data (for example, a number indicating the order in which the waveform arithmetic processing is performed) indicating the priority of the waveform arithmetic processing of the operator m in all the operators k (k = 1,..., M) constituting the musical tone parameter VOICEj is stored. The calculation order of each operator is determined according to the priority, and the waveform calculation process is performed. However, the present invention is not limited to this.
3 so that the ability and load condition can be confirmed.
If the CPU 3 cannot afford to perform the sound source processing, the arithmetic processing of the operator with a low priority may be omitted. Further, in the present embodiment, the priority of the arithmetic processing is
Although the setting is made according to the tone color, the setting is not limited to this, and the setting may be made according to, for example, a MIDI channel. In short, the priority may be selected and set according to some criteria, and the priority may be used at the time of sound generation. For example, when the priority is not set according to the tone, the operator priority data OPPRIm of the tone parameters developed in the tone register TONEPARn may be determined according to the priority. . Further, the operator priority data OPPRIm may be handled as a setting of using / not using the operator m.

In this embodiment, the sampling frequency is the sampling frequency designation data FSAMP.
m, the setting can be made for each operator m. However, the present invention is not limited to this, and it may be set for each of two categories of “carrier” and “modulator”. For example, “carrier” is the frequency FSMAX and “modulator” is 1
/ 2 times (= FSMAX / 2) frequency,
The contents of the algorithm of the tone color parameter may be checked, and the sampling frequency may be set for the operator to be combined. CPU3
May be checked, and the sampling frequency may be increased or decreased as appropriate according to the load condition.

As shown in (c), the buffer OPBUFm executes a waveform operation by the operator m, that is, the operator ON parameter OPONm indicating "1" indicating that the operator m is ON, and the operator m executes the operation. A phase value buffer PHBUFm for storing a phase value calculated by a phase calculation process of a waveform calculation process (described later with reference to FIGS. 14 and 15), and a feedback output value calculated by a feedback sample calculation process of the waveform calculation process. A feedback output value buffer FBm for storing, a modulation data input buffer MODINm for storing the modulation data (this data is used in the phase calculation process), and an operator for storing a tone waveform generated by the operator m, that is, an output value Output value buffer OPOU Processing for calculating and m, the amplitude control EG of the waveform calculation process (hereinafter, "AEG processing"
EG state buffer EGSTATEm for storing each EG parameter calculated by the above-mentioned method.

FIG. 8 shows a MIDI-CH voice table for storing voice (tone color) data (tone color number of the tone parameter VOICEn in this embodiment) selected and set for each MIDI channel, that is, for each part. FIG.

As shown in the figure, in this embodiment, M
The IDI channel is composed of 16 channels, and different timbres can be set for each channel, that is, for each part. Therefore, the tone generator system of the present embodiment generates musical tones of up to 16 types of timbres. be able to. The MIDI-CH voice table stores a tone number corresponding to a tone selected for each channel, that is, a number assigned to the tone parameter VOICEn.

Note that the MIDI-CH voice table is
The table data, that is, the tone numbers are stored in a predetermined area of the RAM 5 in advance in a hard disk or the like corresponding to the MIDI files, and the MIDI file selected by the user is stored in a predetermined position in the RAM 5. At the same time as loading into the data storage area,
The table data corresponding to the IDI file is also MIDI-
Loaded to CH voice table. Of course, the present invention is not limited to this, and the user may be able to set arbitrarily from the beginning, or after a standard timbre (number) is set in advance for a song and this table data is loaded into the MIDI-CH voice table, May be arbitrarily changed.

As described above, MIDI messages are sequentially generated one by one, and when the MIDI message is recognized by the software sound source module SSM, the software sound source module SSM
The tone number assigned to the MIDI channel of the DI message is searched from the MIDI-CH voice table. For example, when the MIDI channel of the MIDI message is “2CH”, the tone number stored in the second position VOICENO2 of the MIDI-CH voice table is searched.

When the timbre number j is retrieved, the software tone generator SSM generates the tone parameter VOICEj as described above. That is, the basic data is read out from the ROM 4 and the other parameters are determined from the MIDI message, and the musical tone parameters VOICEj shown in FIG. 7 are created. Then, the software tone generator module SSM stores the tone parameter VOICEj created in this way in the tone register corresponding to the tone generation channel (for example, tone register TONE) in the tone register group shown in FIG.
PARn).

The control processing executed by the sound source system configured as described above will be described below with reference to FIGS.

FIG. 9 shows the procedure of the tone generator system of the present embodiment, particularly the initial program executed by the CPU 3, that is, the program executed when the user turns on the power of the tone generator system or presses the reset switch. It is a flowchart.

In the figure, first, resetting of various ports and VR5 (not shown) in the RAM 5 and the display 9 are performed.
System initialization such as clearing of an AM (video RAM) is performed (step S1).

Next, the program of the OS is read out from, for example, the hard disk of the hard disk device 10, loaded into a predetermined area of the RAM 5, and the OS program is started (step S2).

FIG. 10 shows that after the initial program, C
It is a flowchart which shows the procedure of the main program which PU3 executes, and this main program is a main routine of the program of the software sound source module SSM.

In the figure, first, in the RAM 5, the area used by the software tone generator module SSM (including the tone register group in FIG. 6) is cleared, and various basic waveform data ( For example, various basic waveform data shown in FIG.
Initial settings such as loading into a predetermined area of No. 5 are performed (step S11).

Next, a basic display is executed on the display 9 to display information according to the progress of the process and to display various icons selected mainly by the mouse 7 (step S12).

Then, the occurrence of the following activation factors is checked (step S13).

Trigger 1: The sequence software APS
1 is activated and the MIDI message is supplied to the software sound source module SSM.

Activation factor 2: The generation of an internal interrupt signal (activation signal) for activating the execution of the waveform calculation processing (performed by the software tone generator module SSM) by the soft timer.

Activation factor 3: A request from the codec hardware to transfer the tone waveform data stored in the output buffer to a buffer (not shown) in the codec hardware.

Activation factor 4: The user operates an input operator such as the mouse 7 or the keyboard 8, and the operation event is detected, or no other activation factor has occurred.

Activation factor 5: The user has performed an operation process for terminating the main routine, and this operation event has been detected.

In the following step S14, the above-mentioned activation factor 1
It is determined whether or not any one of .about.5 has occurred.

As a result of the determination in step S14, when "activation factor 1" has occurred, the process proceeds to step S16, and FIG.
A MIDI processing subroutine described later is executed by using steps 12 and 12. When "activation factor 2" occurs, the process proceeds to step S17, and a waveform calculation processing subroutine described later with reference to FIGS. Proceeds to step S18, the tone waveform data stored in the output buffer is transferred to the buffer of the codec hardware, and when "activation factor 4" occurs, the process proceeds to step S19, where a tone setting event occurs. In this case, a tone color setting processing subroutine described later with reference to FIG. 19 is executed, and if another event occurs, other processing corresponding to the event is executed.
Then, end processing such as returning the display on the display 9 to the state before the main program is started is executed.

After completing any one of the steps S16 to S21, the process returns to the step S12 to repeat the above-mentioned processing.

FIGS. 11 and 12 show the operation in step S16.
5 is a flowchart showing a detailed procedure of a MIDI processing subroutine of FIG.

In the figure, first, the software sound source interface AP of the software sound source module SSM is described.
It is checked whether or not a MIDI event (MIDI message) has been input via I (step S31).

When a MIDI message is output from the sequence software APS1, the MIDI message is converted by the first interface IF1 and the OS into an RA after being converted by the OS via the software tone generator API.
The data is transferred to the MIDI event buffer secured at a predetermined position of M5. In response to this transfer, the software sound source module SSM determines that “activation factor 1” has occurred, and shifts the control of the CPU 3 from step S15 to S16. The processing up to this point is performed in the other processing in step S20.
The occurrence of the event is checked by checking the event stored in the DI event buffer.

Next, in step S32, it is determined whether or not the MIDI event is a note-on event. If the MIDI event is a note-on event, the process proceeds to step S33.
If it is not a note-on event, step S in FIG.
Proceed to 40.

In step S33, the note-on event data is analyzed, and the note number, velocity value, and part number (ie, MIDI channel number) are analyzed.
Are stored in registers NN, VEL, and p secured at predetermined positions in the RAM 5, respectively, and data indicating the time at which the note-on event has occurred is stored in RA
The data is stored in the register TM secured at a predetermined position of M5.
Hereinafter, the contents of the registers NN, VEL, p, and TM are referred to as a note number NN, a velocity VEL, a part p, and a time TM, respectively.

In the following step S34, velocity VE
It is determined whether or not L is equal to or less than a predetermined value VEL1 and whether (&) volume data VOLp is equal to or less than a predetermined value VOL1. Here, VOLp indicates the volume data of the part p stored in the area VOLp secured at a predetermined position in the RAM 5, and this value is, as described with reference to FIG.
It is changed by the event control change # 7 event. This change processing is performed using control change # 7.
When an event occurs, it is performed in the other processing of step S20.

In step S34, VEL ≦ VEL1 & V
When OL ≦ VOL1, the tone color close to the tone color of the part p is replaced by the tone number of the operator, that is, the tone color of the algorithm having a small total number of the carrier and the modulator. A tone parameter VOICE having a tone number and its alternative algorithm
(Step S35), while VE
If L> VEL1 or VOL> VOL1, step S35 is skipped and the process proceeds to step S36.

In the present embodiment, step S35
Is determined according to the values of the velocity VEL and the volume VOL, but is not limited to this.
For example, the load state of the CPU 3 may be detected, and the determination may be made according to the detection result.

In step S36, a tone assignment process based on the note-on event is performed, and the channel number of the assigned tone channel is stored in a register n secured at a predetermined position in the RAM 5. Hereinafter, the content stored in the register n is referred to as a sounding channel n.

In the following step S37, the MI of FIG.
Search the DI-CH voice table and find the position V of part p
The tone color data (tone number) of OICENOp is converted into a tone tone parameter (tone parameter) according to the note number NN and the velocity VEL. For example,
If the tone number j is stored at the position VOICENOp, the tone parameter VOICE described with reference to FIG.
j is generated. Then, the tone parameter VOICEj
Buffer O in each operator data OPmDATAj
Initialize, ie clear, PBUFm.

Next, in step S38, step S3
7, the tone parameter VOICEj together with the time TM and the tone color register T corresponding to the tone generation channel n.
The key-on data KEYONn and each operator-on parameter OPONm in the tone color register TONEPARn are set to “1” “ON”, respectively.

Further, in step S39, the calculation order is determined between the tone generation channels to which the tone generation is performed so that the tone generation calculation is performed in the order in which the note-on event occurrence time is later. Specifically, the channel numbers are rearranged in accordance with the determined operation order and stored in the CH sequence register CHSEQ shown in FIG. Thereafter, the present MIDI processing ends.

In step S40 of FIG. 12, it is determined whether or not the MIDI event is a note-off event. If the MIDI event is a note-off event, the process proceeds to step S41.
Proceed to.

In step S41, the note-off event data is analyzed, the note number is stored in the register NN, and data indicating the time at which the note-off event occurred is stored in the register TM.

In the following step S42, note number N
N is searched for a sounding channel to which sounding is assigned, and the channel number is stored in a register i (hereinafter referred to as a sounding channel i) secured at a predetermined position in the RAM 5.

Then, in step S43, the key-off is designated in the tone color register TONEPARi corresponding to the tone generation channel i, that is, the note-off is reserved at a timing corresponding to the time TM, and then the present MIDI processing is ended.

In step S44, it is determined whether or not the MIDI event is a program change event (an event for changing a tone). If the MIDI event is a program change event, the received program change is stored in the MIDI-CH voice table. The data of the position VOICENOp corresponding to the part p specified by the event (this part p is not necessarily the part number stored in step S33) is changed to the change value PCHNG specified by the received program change event. When the MIDI event is not a program change event while the MIDI event is not a program change event after the change, the MIDI process is terminated after executing other event corresponding processes corresponding to the MIDI event.

In the present MIDI processing, MIDI-C
When a tone corresponding to a plurality of parts is designated in the H voice table and a note-on event of the designated part occurs, a tone of the tone of the part is generated and pronounced. But
The present invention is not limited to this, and it may be possible to select a so-called single tone mode in which only a note-on event of a specific part is received and a tone of the corresponding tone is generated and pronounced.

FIGS. 14 and 15 are flowcharts showing the detailed procedure of the waveform calculation processing subroutine of step S17 in FIG.

In the figure, first, the tone waveform buffer is initialized (step S51). Here, the musical tone waveform buffer refers to an area (buffer) for one frame time generated this time in an area excluding the area (buffer) reserved for reproduction from the output buffer area. And
The initialization of the tone waveform buffer means that the area is secured on the output buffer and the area is cleared.

Next, the load state of the CPU 3 is checked (step S52), and the maximum number of channels CHmax capable of executing the waveform calculation processing is determined (step S53).
Here, the load state of the CPU 3 is checked when the OS
If the load state of U3 is constantly checked, this information may be used. On the other hand, if the OS does not check the load state of CPU3, the main program of FIG. What is necessary is just to provide a routine for measuring the turning time, and use the value calculated according to the measured value.

Instead of the processing in step S53, for example, the same processing as in step S35 in FIG. 11, that is, the tone color assigned to the part is changed to a smaller number of constituent operators (a case where the number is larger depending on the capacity of the CPU 3). (A) A process for changing to an alternative tone may be performed.

Next, an index (index) i indicating a channel number is initialized (i ← 1) (step S5).
4).

Then, in step S55, FIG.
Channel number SEQCHN stored in the i-th position SEQCHi in the CH sequence register CHSEQ
Oi is stored in a variable n (this value is called “channel n” in the present waveform calculation processing subroutine) (n ←←
SEQCHNOi), In step S56, the operator (OP) used in the FM calculation processing for the channel n (the details of which will be described later with reference to FIG. 16) by referring to the algorithm specification data ALGORn of the tone register TONEPARn corresponding to the channel n. The number and the connection mode of each operator are determined.

Further, the calculation amount in the current frame is determined according to a note event or the like (step S57). Here, the determination of the calculation amount specifically refers to determining from which area in the musical tone waveform buffer the waveform arithmetic processing for the channel n is to be performed. That is, as described above, the tone waveform buffer is an area for one frame to be calculated this time, while tone waveform data for each channel is not generated over the entire area for one frame. The tone generation timing and silence timing of each channel are different for each channel,
Since the tone of a certain channel may start or be muted in the middle of the tone waveform buffer, it is meaningful to determine the calculation amount for each channel.

Next, in step S58 of FIG. 15, an FM operation processing subroutine for generating one sample of tone waveform data for channel n, which will be described later with reference to FIG. 16, is executed. It is determined whether or not the tone generation processing has been completed. The determination in step S59 is based on step S5.
Needless to say, the calculation is performed in consideration of the calculation amount determined in step S7.

If it is determined in step S59 that the tone generation processing for one frame has not been completed for channel n, the flow returns to step S58 to generate the next one sample of tone waveform data. On the other hand, when the tone generation processing for one frame has been completed for channel n in step S59, the process proceeds to step S60.

In step S60, the tone waveform data for one frame generated and calculated in steps S58 and S59 is written to the tone waveform buffer. At this time, if the musical tone waveform data is already stored in the musical tone waveform buffer, the calculated data is added to the data, and the result of the addition is written.

Then, the value of the index i is incremented by "1" and updated (step S61), and it is determined whether or not the value of the index i is larger than the maximum channel number CHmax (step S62).

In step S62, when i ≦ CHmax, that is, when processing for a channel for which a waveform is to be generated remains, the flow returns to step S55 in FIG. 14 to repeat the above-described processing. On the other hand, when i> CHmax, that is, when i> CHmax, When the process for the channel for which the waveform is to be generated is completed, a mute channel process for gradually reducing the size of the volume envelope for the sound channel for which the note-off has occurred this time is executed (step S6).
3).

In the following step S64, the tone waveform data generated in this manner is separated from the tone waveform buffer, passed to codec hardware as an output device, and a reproduction instruction (reproduction reservation) is issued. To end.

When the velocity value of the channel n becomes smaller than the predetermined value, the FM calculation for the channel n may not be performed. In order to realize this, as shown in FIG.
5, a step S71 is provided, and the touch velocity data V in the tone register TONEPARn of the channel n is provided.
It is determined whether or not ELn is equal to or greater than a predetermined value VELn1.
When n ≧ VELn1, the process proceeds to step S56, while when VELn <VELn1, the process proceeds to step S72.
Then, after the key-off designation of the channel n is performed in the same manner as in step S43 of FIG. 12, the process may proceed to step S61.

FIG. 16 is a flowchart showing a detailed procedure of the FM calculation subroutine for the channel n in the step S57.

In the figure, first, a variable m for storing the operator number of an operator to be operated (hereinafter, this value is referred to as “operator m to be operated”) is initialized (m ←).
1) Yes.

Next, in the same manner as in step S51,
The load state of the CPU 3 is checked, and the operator priority data OPPRIm of the operator m to be calculated is checked.
Is checked (step S82), and it is determined whether or not to execute the operator calculation process on the calculation target operator m according to the both check results (step S8).
3).

In step S83, the calculation target operator m
Is performed, it is determined whether or not the channel n is continuously sounded from the previous frame (step S84). If the channel n is continuously sounded from the previous frame, the tone color register TONEPARn is determined. Operator data OPmDATA
Based on each data stored in the buffer OPBUFm in n, the operator data OPmDATAn is returned to the state of the operator m at the end of the previous frame calculation (step S85). This is because each operator data OPmDA
In the buffer OPBUFm in TAn, the operation result calculated immediately before is stored, and by using the storage result, the immediately preceding operator data OPm
This is because it is possible to return to the state of DATAn. Then, as described above, the operator data OPmDAT
An is returned to the state at the end of the calculation of the previous frame because the tone waveform data of channel n in the current frame needs to be generated as continuing from the previous frame.

On the other hand, if it is determined in step S84 that the channel n has not been sounded continuously from the previous frame, step S85 is skipped and the flow advances to step S86.

In step S86, an operator calculation processing subroutine for the calculation target operator m, which will be described later with reference to FIGS. 17 and 18, is executed.

In the following step S87, the value of the variable m is updated by incrementing it by "1".
Then, it is determined whether or not the operator calculation has been completed for all the related operators, that is, all the operator data OPmDATAn indicated by the algorithm designation data ALGORn (step S88).

In step S88, when the calculation for the related operator remains, the flow returns to step S82 to repeat the above-described processing. On the other hand, when the calculation has been completed for all the related operators, the FM calculation processing for the current channel n ends. I do.

Note that in steps S82 and S83,
The load state of the CPU 3 is checked to determine whether or not to execute the operator calculation for the operator m to be calculated. However, the present invention is not limited to this. May not be executed. This allows the CPU
Even if the ability of the third is not so high, the number of sounds can be increased.

FIG. 17 and FIG.
FIG. 19 is a flowchart showing a detailed procedure of an operator calculation processing subroutine for the calculation target operator m. FIG. 19 is a diagram showing a basic flow of operator calculation performed in this operator calculation processing. Hereinafter, FIG.
The operator calculation process for the calculation target operator m will be described with reference to FIG.

In FIG. 17, first, it is determined whether or not the operator-on parameter OPONm in the operator data OPmDATAn of the operator m to be operated is on ("1") (step S91). When OPONm = 0, that is, when OPONm = 0 If the operator m does not need to perform the operator operation, the operator operation process is immediately terminated. On the other hand, if OPONm = 1, that is, if the operator m is to perform the operator operation, step S92 is executed. Proceed to.

In the step S92, the operator data O
It is determined whether or not the sampling frequency designation data FSAMPm in PmDATAn is "0", that is, whether or not a tone waveform is generated at the sampling frequency FSMAX in the standard mode. When FSAMPm = 0, the calculation by each operator is performed. The process is based on generating a tone waveform at the sampling frequency of the standard mode.
AEGm calculation is performed according to the set value of the envelope parameter EGPARm in the operator data OPmDATAn, and the calculation result is stored in the EG state buffer EGS.
It is stored in TATEm (step S93).

On the other hand, in step S92, FSAMPm ≠
0, for example, when FSAMPm = f, that is, when changing the sampling frequency FSMAX of the standard mode to 2- ^f times and generating a musical tone waveform at the frequency after the change, a parameter (hereinafter, referred to as a rate changing) in the envelope parameter EGPARm. The AEG calculation is performed by multiplying the “change rate related parameter” by 2 ^f and the calculation result is stored in the EG state buffer EGSTATEm. The reason why the rate of the parameter related to the change rate is increased by 2 ^f to perform the envelope generation operation is that at this time, the sampling frequency is reduced to FSMAX × 2 ^−f. In the embodiment, the rate of the parameter related to the change rate of the envelope parameter EGPARm, that is, the tone waveform generation at the sampling frequency is performed by increasing the time change, and then the generated waveform samples are stored in 2 ^f corresponding continuous buffers. This is because, by writing, a method of adjusting the tone so that the tone has the original pitch is adopted.

As described above, step S93 or S94
Calculates the envelope data AEGm in FIG.

In the following step S95, operator data OP is added to data AEGm calculated by AEGm calculation.
the total level parameter TLm in mDATAn
And the output level AMPm (= AEGm × TLm) of the operator m to be calculated is calculated as shown in FIG.

Then, the amplitude control envelope data AEGm calculated in step S93 or S94 and the output level AMPm of the calculation target operator m calculated in step S95 are checked (step S9).
6), each data value AEGm and AMPm is
For example, it is determined whether or not the output is below the predetermined level for a predetermined time and the output of the calculation target operator m may be completed, that is, whether or not the tone waveform calculation in the calculation target operator m may be ended (step S97), and the answer is affirmative. (YE
In the case of S), the process proceeds to step S98, while if the answer is negative (NO), the process proceeds to step S101 in FIG.

In step S98, it is determined whether or not the calculation target operator m is the "carrier". If it is "carrier", the flow advances to step S99 to modulate only the calculation target operator m and the calculation target operator m. Clear each buffer OPBUF of "Modulator",
After stopping the waveform calculation, the operator calculation process ends. As described above, when the calculation target operator m is “carrier”, not only the calculation target operator m,
The reason for stopping the waveform calculation of the modulator modulating only the calculation target operator m is that the carrier is
As shown in FIG. 4, since the operator is the operator who finally outputs the musical tone waveform data, when the output from the carrier may be omitted, that is, when the process proceeds from step S97 to step S98 to step S99, the modulator at the preceding stage is used. Is not required. However, when the modulator is modulating another carrier, the waveform calculation of the modulator cannot be stopped, so this is secured by “only”.

On the other hand, if the operation target operator m is not a carrier in step S98, that is, if it is a modulator, the buffer OPBUF of the operation target operator m
After clearing only m and stopping the waveform calculation (step S100), the operator calculation process ends.

In step S101 of FIG. 18, the algorithm designation data ALGORn is checked. Then, in step S102, it is determined whether or not the calculation target operator m is modulated by another operator.

In step S102, when the operation target operator m is modulated by another operator, each operator data OPkDATA performing the modulation is operated.
n, and adds the operator output data stored in the operator output value buffer OPOUTk in n.
Modulated data input buffer M of operator m to be operated
If it is stored in ODINm (step S103), on the other hand, if the calculation target operator m has not been modulated by another operator, step S103 is skipped and the process proceeds to step S104.

At step S104, at step S9
Similarly to 2, it is determined whether or not the sampling frequency designation data FSAMPm in the operator data OPmDATAn is “0”. .

In step S105, a phase value update calculation for updating the phase value is performed, and the calculation result is stored in a phase value buffer PHBUFm (hereinafter referred to as “phase value P”) in the operator data OPmDATAn of the operator m to be calculated.
HBUFm ”). Here, the phase value update operation is an operation surrounded by a broken line A in FIG. 19, that is, an operation represented by the following equation.

MODINm + FBm + FNOn × MUL
Tm + PHBUFm where MODINm and FBm are modulation data input buffers M in operator data OPmDATAn, respectively.
ODINm and feedback output value buffer FBm
, And FNOn is a tone parameter V
Indicates the frequency number FNOn in OICEn, and indicates MULT.
m indicates the frequency multiple data MULTm in the operator data OPmDATAn, and PHBUFm indicates the phase value buffer PH in the operator data OPmDATAn.
The previous value of the value stored in BUFm is shown.

In the following step S106, step S1
A table address is calculated based on the phase value PHBUFm calculated in step 05, and a basic waveform selected according to the wave select data WSELm of the operator m to be calculated (for example, selected from the eight basic waveforms shown in FIG. 5). From the calculated waveform) data (hereinafter, referred to as a “basic waveform table”).
VEm (PHBUFm) is read, and the data WAV
Em (PHBUFm) is multiplied by the output level AMPm calculated in step S95, and an operator output value buffer OPOUTm (= WAV) of the operator m to be calculated is obtained.
Em (PHBUFm) × AMPm).

In step S107, a feedback sample calculation is performed by the following equation, and the calculation result is stored in the feedback output value buffer FBm of the calculation target operator m.

0.5 × (FBm + OPOUTm × FBLm) where OPOUTm indicates the waveform sample data generated in step S106, and FBLm is the feedback level data FBLm of the operator m to be calculated.
Is shown. Here, the feedback sample calculation is performed to prevent parasitic oscillation (the same applies to step S112 described later).

In step S108, similarly to step S98, it is determined whether or not the operator m to be operated is a "carrier". If the operator m is a "modulator", the operator operation processing is immediately terminated.
In the case of “carrier”, the waveform sample data OPOUTm generated in step S106 is multiplied by the volume data VOLn of the musical tone parameter VOICEn, and the multiplication result (= OPOUTm × VOLn) is stored in the waveform buffer (corresponding buffer) in the current time. Is added to the position indicated by the pointer indicating the write position of the pointer, and the value of the pointer is updated by being advanced by “1” (step S
109) Thereafter, this operator calculation process ends.

In step S110, a phase value update operation is performed, and the operation result is stored in the phase value buffer PHBUFm. The arithmetic processing in step S110 is different from the arithmetic processing in step S105 only in that the processing shown in block B of FIG. 19 is added. This is FS
Since AMPm = f (≠ 0), the phase value is shifted up by f bits, that is, the phase value buffer PHBU
This is because it is necessary to multiply the value of Fm by 2 ^f to change the read address of the basic waveform table to an address at the sampling frequency FSMAX × 2 ^−f , that is, it is necessary to skip reading.

Next, in the same manner as in step S106, a waveform sample is generated by the following equation and stored in the operator output value buffer OPOUTm.

WAVEm (2 ^f × PHBUFm) × AMPm Then, a feedback sample calculation is executed in the same manner as in step S107 (step S112).

Next, in step S113, similarly to step S108, it is determined whether or not the operator m to be operated is a "carrier". If the operator m is a "modulator", the operator operation processing is immediately terminated.
In the case of "carrier", the waveform sample data OPOUTm generated in step S111 is multiplied by the volume data VOLn of the musical tone parameter VOICEn in the same manner as in step S109, and the multiplication result (= OP
The OUTm × VOLn), in the waveform buffer, as well as added to 2 ^f-number of regions (buffer) that is continuous from the position where the pointer indicates the value of the pointer "2 ^f"
After the update (step S109), the operator calculation process ends. Step S109
Then, when writing sample data of the same value, an interpolation operation between samples may be performed as necessary, and the interpolated value may be written.

In the present embodiment, step S10
As described in 6 and S111, the values stored in the basic waveform table are used as the basic waveform data. However, the present invention is not limited to this, and the basic waveform data may be generated by calculation. Of course, the basic waveform data may be generated by a combination of the table data and the calculation.

Also, steps S106 and S110
As the address for reading out the basic waveform table, the address calculated based on the phase value PHBUFm calculated in steps S105 and S110, respectively, is used. However, the present invention is not limited to this. You may use what was distorted by the table etc.

FIG. 20 is a flowchart showing a detailed procedure of the tone color setting processing subroutine of step S19 in FIG.

In the figure, first, a MIDI channel and a tone color corresponding to each channel are set (step S121). As described above, in the present embodiment,
MIDI channels and their corresponding tones are determined by a MIDI-CH voice table. Table data to be loaded into the MIDI-CH voice table is stored on a hard disk or the like. Table data to be
Since the data is loaded into the IDI-CH voice table, the processing in step S121 is editing of the currently loaded table data or loading (reading) of new table data.

Note that the user can set the desired number of operators for each MIDI channel.
-When changing the tone number in the CH voice table,
If the desired number of operators is set for the channel, a list of tone numbers corresponding to the tone parameters VOICE corresponding to or within the number of operators is displayed, and the user selects a desired tone number from the list. You may make it select and set. At this time, CP
The number of desired operators set for the channel may be automatically changed in accordance with the load state of U3, and a list of tone numbers within the changed number of operators may be displayed. Further, when the user changes the tone number of a certain channel in the MIDI-CH voice table, the total number of operators constituting the tone parameter VOICE corresponding to the tone number is checked. A warning that the tone cannot be assigned may be displayed. In addition to this warning,
The timbre number of the channel may be automatically changed to the timbre number of the substitute timbre having a small number of operators.

As described above, in the present embodiment,
The number of operators used in the FM calculation processing is flexibly changed according to the capability, operating environment, purpose, setting, etc. of the CPU 3, so that the load on the CPU 3 and the quality of the output musical sound waveform can be freely changed. Therefore, the degree of freedom of the entire sound source system can be improved.

Although the present embodiment has been described using the FM sound source system as the musical sound generation process, the present invention is not limited to this.
For example, by combining a tone waveform generation block
The present invention can be easily applied to a sound source that performs predetermined signal processing such as (amplitude modulation) and PM (phase modulation). Further, the method of reducing the load on the CPU included in the present invention can also be applied to a case where a sound source of a waveform memory reading method and a physical model sound source are realized by software.

In the present embodiment, application to a personal computer has been described as a main example, but application to amusement equipment such as games and karaoke, electronic musical instruments, and general electronic equipment is also easily possible. Also, the present invention can be applied to a sound source board or a sound source unit as an option of a personal computer.

The software according to the present invention can be supplied on a disk medium such as a floppy disk, a magneto-optical disk, a CD-ROM, or a memory card. Further, a semiconductor memory chip (typically, a ROM) storing software data may be replaced with a computer device and added. Further, the sound source software according to the present invention may be distributed through the communication I / F 14.

Also, the sound source software according to the present invention
In the computer system, what kind of position (application software or, for example, device driver software) to start and operate may be determined as appropriate according to the system configuration and the OS.

The sound source software according to the present invention or its functions may be incorporated into other software, for example, amusement software such as games, karaoke software, automatic performance / accompaniment software, and the like. Further, it may be directly incorporated in the computer OS.

[0149]

As described above, according to the present invention,
A tone waveform generation block indicated by the set algorithm is assigned to the selected sounding channel, and the assigned tone waveform generation blocks are combined in a combination manner indicated by the algorithm to perform a tone waveform generation operation, thereby generating tone waveform data. Is generated, the number of musical tone waveform generating blocks for each tone generating channel can be arbitrarily changed before tone generation is assigned. Thereby, the tone waveform data can be changed according to the capability of the tone waveform data generating means. There is an effect that it is possible to flexibly change the load state on the generating means and the quality of generated musical sound waveform data.

A tone waveform generation block indicated by an algorithm set corresponding to the tone color of the tone waveform data to be generated is assigned to the selected tone generation channel, and the assigned tone waveform generation block is assigned to the algorithm. Since the musical sound waveform generation operation is performed in combination in the combination mode shown by, and the musical sound waveform data is generated,
The same effect as the above effect can be obtained.

Preferably, when the tone color is set by the tone color setting means, and the musical tone waveform generating block number is set by the block number setting means in the performance part, the tone color set in the performance part is set. Since the tone color is changed to a tone having the tone waveform generation block within the range of the set number of blocks, the above effect can be further enhanced.

Preferably, the number of musical tone waveform generation blocks assigned to the tone generation channel is changed according to a predetermined condition during the tone waveform generation calculation for the tone generation channel. According to the capability of the waveform data generating means, the load state on the musical sound waveform data generating means and the quality of the generated musical sound waveform data can be flexibly changed.

Further, according to the present invention, a personal computer
In computer equipment that often executes a plurality of arbitrary task processes (eg, word processing, network communication, etc.) in addition to music performance, even if CPU power is applied to tasks other than the sound source during the execution of the software sound source process, the sound is interrupted. And the like can be reduced. From a different point of view, it can be said that more tasks can be processed even while the sound source processing is being executed.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an example of a system configuration of software provided in a sound source system according to an embodiment of the present invention.

FIG. 2 is a diagram showing a schematic configuration of a sound source system according to an embodiment of the present invention.

FIG. 3 is a diagram for explaining an outline of a musical sound generation process performed by the sound source system of FIG. 1;

FIG. 4 is a diagram for explaining an outline of a musical sound generation process using the FM sound source system.

FIG. 5 is a diagram showing an example of basic waveform data from which a basic waveform table is selected.

FIG. 6 is a diagram showing a timbre register group that develops timbre parameters of musical tones to be produced by assigned tone generation channels.

FIG. 7 is a diagram showing a data format of a musical tone parameter VOICEj.

FIG. 8 is a diagram showing a MIDI-CH voice table for storing tone color numbers of musical tone parameters VOICEEn selected and set for each MIDI channel (CH).

FIG. 9 is a flowchart showing a procedure of an initial program executed by a CPU of the sound source system of FIG. 1;

FIG. 10 is a flowchart showing a procedure of a main program executed by a CPU subsequent to the initial program of FIG. 9;

FIG. 11 is a flowchart showing a detailed procedure of a MIDI processing subroutine of FIG. 10;

FIG. 12 is a flowchart showing a continuation of the MIDI processing subroutine of FIG. 11;

FIG. 13 is a diagram illustrating an example of a format of a CH sequence register.

FIG. 14 is a flowchart showing a detailed procedure of a waveform calculation processing subroutine of FIG. 10;

FIG. 15 is a flowchart showing a continuation of the waveform calculation processing subroutine of FIG. 14;

16 is a flowchart illustrating a detailed procedure of an FM calculation processing subroutine for a channel n in FIG. 15;

17 is a flowchart illustrating a detailed procedure of an operator calculation processing subroutine for a calculation target operator m in FIG. 16;

18 is a flowchart showing a continuation of the operator calculation processing subroutine for the calculation target operator m in FIG. 17;

FIG. 19 is a diagram showing a basic flow of operator calculation performed in the operator calculation processing of FIGS. 17 and 18.

FIG. 20 is a flowchart showing a detailed procedure of a tone color setting processing subroutine of FIG. 10;

[Explanation of symbols]

3 CPU (algorithm setting means, tone generation channel selection means, musical tone waveform data generation means, tone color setting means, block number setting means, tone color changing means, block number changing means,
OPmDATAn operator data (tone waveform generation block) SSM software tone generator module (tone waveform generation block, algorithm setting means, sound channel selection means, tone waveform data generation means, tone color setting means, number of blocks) Setting means, tone color changing means, block number changing means, reduction mode setting means, sampling frequency setting means)

Continuation of the front page (56) References JP-A-7-271378 (JP, A) JP-A-7-319471 (JP, A) JP-A 2-179697 (JP, A) (58) Fields investigated (Int .Cl. ⁷ , DB name) G10H 7/08 G10H 1/02

Claims (11)

(57) [Claims] 1. A sound source system for generating musical tone waveform data having different timbres for each of a plurality of tone generation channels by combining a plurality of tone waveform generating blocks formed of software for performing a tone waveform generating operation. , and algorithms setting means for setting an algorithm which determines the number and the combination aspect of the combination is tone waveform generation block for each of the tone color, a tone color setting means for setting a tone color for each performance part of the song, respective performance parts Musical sound waveform generation combined with each other
Block number setting means for setting the number of blocks, and when a timbre is set by the timbre setting means,
A musical sound waveform is generated by the block number setting means for the performance part.
If the number of composed blocks is set, the performance
The tone set to the preset
Tone changer that changes to a tone with a sound waveform generation block
Step, sound channel selection means for selecting a channel for generating tone waveform data from the plurality of tone channels, and the selected tone channel is set in correspondence with the tone color of the tone waveform data to be generated. Allocate the musical tone waveform generation block indicated by the algorithm, perform the musical tone waveform generation operation by combining the assigned musical tone waveform generation block in a combination manner indicated by the algorithm,
A sound waveform data generating means for generating musical sound waveform data. 2. The method according to claim 1, further comprising a block number changing means for changing the number of musical tone waveform generating blocks assigned to the tone generating channel according to a predetermined condition during the musical tone waveform generating operation on the tone generating channel. Claim
2. The sound source system according to 1 . 3. A tone waveform data generating means for each tone waveform generating block, comprising: a sampling frequency setting means for setting a sampling frequency serving as a reference for tone waveform generating calculation executed by each tone waveform generating block. 3. The sound source system according to claim 1, wherein the tone waveform generating operation of each tone waveform generating block is performed based on a sampling frequency set in each tone waveform generating block. . 4. A sound source system for generating musical tone waveform data for each of a plurality of tone generation channels by combining a plurality of musical tone waveform generating blocks formed of software for performing a musical tone waveform generating operation. Algorithm setting means for setting an algorithm for determining the number of musical tone waveform generation blocks and the combination thereof; tone generation channel selection means for selecting a channel for generating tone waveform data from the plurality of tone generation channels; Musical tone waveform generating block indicated by the set algorithm, and performing the musical tone waveform generating operation by combining the assigned musical tone waveform generating blocks in a combination manner indicated by the algorithm, thereby generating musical tone waveform data. Generate And means, in tone waveform generation operation in the sound channel, where
Assigned to the sound channel according to the specified conditions.
Number of blocks to change the number of generated tone waveform generation blocks
Tone generator system, characterized in that it comprises a further means. 5. A tone waveform data generating means, comprising: for each tone waveform generating block, a sampling frequency setting means for setting a sampling frequency serving as a reference for tone waveform generating calculation executed by each tone waveform generating block. 5. The sound source system according to claim 4 , wherein the tone waveform generating operation of each tone waveform generating block is performed based on a sampling frequency set in each tone waveform generating block. 6. A sound source system for generating musical tone waveform data for each of a plurality of tone generation channels by combining a plurality of musical tone waveform generating blocks configured by software for performing a musical tone waveform generating operation. Algorithm setting means for setting an algorithm for determining the number of musical tone waveform generation blocks and the combination thereof; and for each musical tone waveform generation block,
Samp that is the basis for the tone waveform generation operation performed by Lock
Sampling frequency setting means for setting the ring frequency
When a sound channel selection means for selecting a channel for generating musical tone waveform data from said plurality of tone generation channels, the sound channel is the selected, assigned tone waveform generation block indicated by the algorithm set, assigned the The tone waveform generation blocks are combined in the combination mode indicated by the algorithm, and each tone waveform generation block is combined .
A tone waveform data generating means for generating tone waveform data by performing tone waveform generation calculation on the basis of the sampling frequency set to the lock . 7. The algorithm setting means, wherein:
Set the algorithm for each tone, The musical sound waveform data generating means is configured to select the selected one of the sound generation channels.
The channel contains the sound waveform data to be generated in the channel.
The algorithm set above is indicated according to the data tone.
Characteristic sound waveform generation block is assigned
The sound source system according to claim 4. 8. A method for reducing the number of musical tone waveform generating blocks indicated by an algorithm corresponding to a tone color set for each part for each performance part of a song, wherein the number of blocks is changed. 3. The sound source system according to claim 2 , wherein the means changes the number of musical tone waveform generation blocks according to the set reduction mode and the predetermined condition. 9. The predetermined condition is that a volume envelope of a musical tone waveform generated in the sound channel is attenuated, and the block number changing unit attenuates the volume envelope of the musical tone waveform to a predetermined value or less. Occasionally, depending on the attenuation, claim 2 or 4 Neu and changes in the direction of decreasing the number of tone waveform generation block indicated algorithm set by the algorithm setting means
Sound source system of Zurekani described. 10. The predetermined condition is that the volume of a musical sound waveform to be generated in the sound channel is set to be low, and the block number changing means sets the volume of the musical sound waveform to be low. 5. The sound source system according to claim 2 , wherein the number of musical tone waveform generation blocks indicated by the algorithm set by the algorithm setting means is changed in a direction of decreasing the number. 11. The predetermined condition is that, among the tone waveform generation blocks assigned to the sound channel, there is a block whose output level is smaller than a predetermined value. When the tone waveform generation block that has become equal to or less than the predetermined value does not affect other tone waveform generation blocks, the operation of the tone waveform generation block is stopped, and the number of tone waveform generation blocks indicated by the algorithm is reduced. 5. The method according to claim 2, wherein the value is changed in a direction of decreasing the value.
Sound source system of Zurekani described.