JP3735172B2 - Performance information processing method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a performance information processing method, and more particularly to a performance information processing method suitable for use in constructing a synthesizer by software on a general-purpose computer such as a personal computer or a workstation.
[0002]
BACKGROUND OF THE INVENTION AND PROBLEMS TO BE SOLVED
Conventionally, in order to generate a musical sound, it has been necessary to use dedicated hardware for generating a musical sound, a so-called electronic musical instrument.
[0003]
However, in recent years, general-purpose computers have been equipped with high-performance CPUs and D / A converters (digital-analog converters) for outputting data processed by the CPU as analog signals to the outside. Accordingly, the possibility that musical sounds can be generated in real time on a general-purpose computer without using dedicated hardware has been studied.
[0004]
Specifically, since the musical sound data must be output to the D / A converter accurately at every sampling period, it is considered to perform processing for generating musical sound data at every sampling period according to a timer. Yes.
[0005]
In order to realize this processing, it is conceivable to perform processing for generating musical tone data in the interrupt routine according to the timer interrupt generated at each sampling period, but the processing for generating musical tone data is complicated. It has been pointed out that it is difficult to perform such complicated processing in the interrupt routine.
[0006]
Alternatively, in order to realize the above-described processing, it is conceivable to perform processing for generating musical tone data by a task activated at each sampling cycle, but according to this, the task is performed at a complicated timing at every sampling cycle. It is necessary to perform the process of switching the task, and the overhead for task switching increases, and the time that the CPU can actually process is shortened.
[0007]
By the way, in general, it is possible to use one software on various general-purpose computers. However, the capabilities of the CPUs mounted on these various general-purpose computers vary widely, and are adjusted to match those of low-performance CPUs. When software is created, there is a problem that a computer equipped with a high-capacity CPU cannot fully use the CPU power.
[0008]
In general-purpose computers, a method of using a plurality of software programs at the same time is generally performed. However, when processing other than the processing for generating a musical tone is performed simultaneously with the processing for generating a musical tone, The load on the CPU changes every moment according to the processing status of other processes. For this reason, when software is created in accordance with the heaviest CPU load, there is a problem in that the CPU capacity cannot be fully used when the load is light.
[0009]
Further, in general, at least a part of the process of generating a musical tone by the CPU is repeatedly performed at regular time intervals. For example, note-on (sounding start) / note-off (sounding end) of MIDI information is indicated. The instruction is generated asynchronously with the repetition interval in accordance with a user (user) desired timing. That is, data is transferred to a process for generating a musical tone in accordance with an instruction indicating note-on / note-off, and a process for generating a musical tone based on the transferred data in the process of generating a musical sound that has received the data. However, since the instruction indicating note-on / note-off is asynchronous with the repetition interval, the instruction indicating note-on / note-off is performed during the process of generating a musical tone. There is a problem that noise is generated in the musical tone being generated if the processing is performed to rewrite the data or generate the musical tone while rewriting the data according to the instruction indicating note on / note off. The need to improve the data transfer method was pointed out.
[0010]
For this reason, the applicant of the present application solves the various problems described above by performing processing suitable for a general-purpose computer, and can generate a musical sound efficiently in real time on the general-purpose computer. A patent application has already been filed as a “musical tone generation processing method” in 1996 Patent Application No. 147820 (filing date: May 20, 1996).
[0011]
In this 1996 Patent Application No. 147820 “Musical Sound Generation Processing Method”, musical sounds are efficiently generated in real time on a general purpose computer without using dedicated hardware for generating musical sounds. It is designed to build a synthesizer by software on a computer.
[0012]
Here, the 1996 Patent Application No. 147820 “Musical Sound Generation Processing Method” will be described. The 1996 Patent Application No. 147820 “Musical Sound Generation Processing Method” includes four wave buffers capable of storing a musical sound signal for 10 ms. The sound source task that is provided more than 10 ms and is activated every 10 ms stores a musical sound signal for 10 ms in the wave buffer, and simultaneously passes the wave buffer storing the musical sound signal for 10 ms to the output task. It is designed to receive new wave buffers.
[0013]
When the output task receives a wave buffer storing a musical tone signal for 10 ms from the sound source task, it returns the free wave buffer to the sound source task and transfers the stored wave buffer to the audio data output device. Is stored in the wave buffer that has received the earliest musical sound signal for the predetermined number of samples required by the audio data output device in response to an interrupt from the audio data output device in chronological order. Only the DMA transfer to the audio data output device is made.
[0014]
The audio data output device has a built-in buffer for temporarily storing the audio data to be output, outputs the audio data temporarily stored for each sampling period for each sampling, and the audio data remaining in the built-in buffer. The output task is interrupted when becomes less than a predetermined amount. Then, the audio data output from the audio data output device is emitted as a musical sound.
[0015]
The interruption to the output task described above usually takes a cycle of about several ms.
[0016]
One of the four or more wave buffers described above is always output by the output task to the audio data output device, one is used by the sound source task for processing, one is free, and one is the audio data output device. The rotation is repeated such that the remaining ones are free or the other ones are free or waiting for output to the audio data output device.
[0017]
As described above, the 1996 Patent Application No. 147820 “Musical Sound Generation Processing Method” is configured to output a musical sound signal via a wave buffer or an internal buffer of an audio data output device. For this reason, the tone generator task can perform the generation process of the tone signal for 10 ms, and the tone signal is interrupted even if the start timing of the tone generator task is slightly delayed due to the processing of tasks other than the tone generator task. This produces an excellent effect of being output without any problems.
[0018]
If you want to increase the allowable delay in the start timing of the sound source task, increase the number of wave buffers, and increase the number of wave buffers waiting to be output to the audio data output device. Good.
[0019]
By the way, the 1996 Patent Application No. 147820 “Musical Sound Generation Processing Method” is configured to output a musical sound signal via a buffer such as a wave buffer or an internal buffer of an audio data output device as described above. Therefore, it takes a time interval of at least several tens of ms from when note-on is received as MIDI information until the musical sound corresponding to the received note-on is actually emitted. For this reason, when MIDI information is supplied by a sequencer (automatic performance), the sound generation timing in the entire performance is delayed by several tens of ms, respectively. The end of the performance is only delayed by several tens of ms, and does not cause any problems. However, when the MIDI information is supplied by a real-time performance such as a keyboard performance of the performer, the musical tone corresponding to the note-on is supplied after the player presses the keyboard to supply the note-on. Since a delay of several tens of ms occurs before the sound is actually emitted, the delay time is too long, and the follower between the performance by the performer and the musical sound to be emitted is inferior, and the performer performs naturally. There was a problem that could not be done. In particular, such problems have become extremely alarming when the performance data played by the performer in real time is input to the sequencer as sequence data.
[0020]
When the number of wave buffers in the output waiting state is increased, this delay amount becomes larger, and such a problem becomes larger.
[0021]
The present invention has been made in view of the above-described problems, and the object of the present invention is to perform a player's synthesizer with a software on a general-purpose computer such as a personal computer or a workstation. It is an object of the present invention to provide a performance information processing method capable of promptly emitting a musical sound to be emitted in correspondence with a real-time performance without requiring a large delay time from a pronunciation instruction.
[0022]
[Means for Solving the Problems]
In order to achieve the above object, the performance information processing method according to the present invention is a method for generating a musical tone signal, for example, as a dedicated hardware sound source independent of a general-purpose computer among recent general-purpose computers. It was made by paying attention to the fact that some sound source boards composed of dedicated hardware are installed.
[0023]
It is known that the above-described sound source board is inexpensive and is used for a sound source of a computer game, so that it is mounted on a large number of general-purpose computers. That is, many general-purpose computers are already equipped with a sound source board when purchased by a user.
[0024]
By the way, as described above, such a sound source board is inexpensive and mainly used for a sound source of a computer game. Compared to the one generated by building a synthesizer by software on a general-purpose computer of No. 147820 “Musical sound generation processing method”, it is not satisfactory in terms of music, but it is composed of dedicated hardware. Therefore, a delay time from when a note-on is supplied until the tone signal corresponding to the note-on is output to the outside, that is, a tone corresponding to the note-on after the note-on is supplied is actually released. The delay time until the sound is played is short, and it can sufficiently follow the real-time performance by the performer. In the performance information processing method according to the present invention, for example, such a sound source board is used as a dedicated hardware sound source independent of a general-purpose computer, and a musical tone signal based on a real-time performance by a performer is generated by the sound source board. It is what.
[0025]
That is, in the performance information processing method according to the first aspect of the present invention, the general-purpose processing device is operated based on software, and the sound source processing constituting the first sound source means and the supplied performance information are automatically performed. A discriminating process for discriminating whether the performance is due to the performance or the performance of the performer, and if the performance information supplied by the discriminating process is determined to be due to an automatic performance, the sound source processing is performed based on the supplied performance information When the musical performance signal supplied by the discrimination process is determined to be due to the performance of the performer, a dedicated signal independent of the general-purpose processing device is generated based on the supplied performance information. And a performance control process for generating a musical sound signal by the second sound source means constituted by the hardware sound source.
[0026]
Therefore, according to the performance information processing method of the first aspect of the present invention, the performance information supplied by the performer's performance is a dedicated hardware sound source that is independent of the general-purpose processing device. Since the tone signal is generated by the two tone generator means, the tone to be emitted in response to the performance of the performer can be emitted promptly without requiring a large delay time from the pronunciation instruction.
[0027]
Here, for example, the discriminating process discriminates that the performance information supplied from the sequencer program of the general-purpose processing device is based on the automatic performance as in the performance information processing method according to claim 2 of the present invention. The performance information supplied from the performance information input interface may be processing for determining that the performance information is based on the performance of the performer.
[0028]
Further, in the performance information processing method according to claim 1 as in the performance information processing method according to claim 3 of the present invention, the first sound source means and the second sound source means according to instructions from the performer. And a process of supplying common control information to the sound source means.
[0029]
Therefore, according to the performance information processing method of the third aspect of the present invention, tuning and timbre can be shared between the first sound source means and the second sound source means.
[0030]
In the performance information processing method according to the fourth aspect of the present invention, the general-purpose processing device is operated based on software, and the sound source processing constituting the first sound source means and the supplied performance information are automatically performed. The first mode and the second mode are discriminating processing for discriminating whether the performance is due to the performance or the performance of the performer, and the processing information when the performance information supplied by the discrimination processing is determined to be due to the automatic performance. And when the first mode is set by the setting process when it is determined that the performance information supplied by the determination process is based on the automatic performance, and the setting process sets the first mode, A musical tone signal is generated by the sound source process based on the supplied performance information, and the performance information supplied by the determination process is determined to be an automatic performance. When the second mode is set by the setting process, a tone signal is generated by the tone generator process based on a part of the supplied performance information, and the performance information supplied by the discrimination process is If it is determined that the performance is performed by the performer, a musical tone signal is generated by the second sound source means composed of a dedicated hardware sound source independent of the general-purpose processing device based on the supplied performance information. Performance control processing to be performed.
[0031]
Therefore, according to the performance information processing method according to claim 4 of the present invention, for example, with respect to a specific part of the performance by automatic performance, the first sound source means does not generate a musical sound, and the specific part is played. The music can be generated by the second sound source means when the person performs.
[0032]
In the performance information processing method according to claim 5 of the present invention, the general-purpose processing device is operated based on software, whereby the sound source processing constituting the first sound source means and the supplied performance information are automatically performed. A determination process for determining whether it is due to a performance or a performance by a performer, and a first mode as a processing mode when it is determined that the performance information supplied by the determination process is due to a performer's performance, When it is determined that the performance information supplied by the setting processing for setting the second mode and the performance determination processing is based on the automatic performance, a musical tone signal is generated by the sound source processing based on the performance performance information supplied. And the first mode is set by the setting process when it is determined that the performance information supplied by the determination process is based on the performance of the performer. Generates a musical tone signal by the second sound source means composed of a dedicated hardware sound source independent of the general-purpose processing device based on the supplied performance information, and the performance information supplied by the discrimination processing is If it is determined that the performance is performed by the performer and the second mode is set by the setting process, a tone signal is generated by the tone generator process based on the supplied performance information. The performance control process is performed.
[0033]
Therefore, according to the performance information processing method according to claim 5 of the present invention, the performance information supplied by the performer's performance is supplied to the first sound source means by switching the mode, and the first sound source means. Thus, a musical tone can be generated.
[0034]
As described above, since the sound source board is already mounted on many general-purpose computers, the sound source is used as the second sound source means constituted by a dedicated hardware sound source independent of the general-purpose processing device. When using a board, there is an effect that the user does not need to make a new investment in hardware.
[0035]
In addition, as described above, the quality of the tone signal generated by the tone generator board is not satisfactory in terms of music. However, when the performance data played by the performer in real time is input to the sequencer as sequence data. Since the purpose is to input performance data to the sequencer, the quality of the tone signal generated by the tone generator board is not musically satisfactory, and no inconvenience arises. That is, when the sequence data inputted while listening to the musical sound generated by the tone generator board is reproduced by the sequencer, the musical tone signal is not generated using the tone generator board, but the 1996 Japanese Patent Application No. 147820. A musical tone signal may be generated by constructing a synthesizer by software on a general-purpose computer of the “musical sound generation processing method”, and a musical tone signal having a quality satisfactory in terms of music can be obtained when sequence data is reproduced.
[0036]
Note that a medium recording a program for causing a computer to execute the performance information processing described in claims 1 to 5 belongs to the scope of the present invention.
[0037]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an example of an embodiment of a performance information processing method of the present invention will be described in detail with reference to the accompanying drawings.
[0038]
FIG. 1 is a block diagram showing a hardware configuration of a typical general-purpose computer on which a program according to the performance information processing method of the present invention can be installed.
[0039]
The general-purpose computer shown in FIG. 1 has a very general configuration, and via a bus 10, a CPU 12, a memory 14, a disk controller 16, a video card 18, an I / O card 20, A sound card 22, a MIDI interface 32, and a sound source board 34 composed of hardware independent of the general-purpose computer are connected. Furthermore, a hard disk 24 is connected to the disk controller 16, and the video card 18. Is connected to a display device 26, a keyboard 28 and a mouse 30 are connected to the I / O card 20, and another electronic musical instrument 36 such as a keyboard is connected to the MIDI interface 32. In this general-purpose computer, musical sound signals are appropriately output from the sound card 22 and the sound source board 34.
[0040]
That is, in this general-purpose computer, as shown in FIG. 2, the program is supplied from a sequencer program that is started simultaneously with a music generation process program (hereinafter referred to as a “synthesizer program”) in a general-purpose computer. The MIDI information and the MIDI information supplied from the other electronic musical instrument 36 are appropriately distributed to the sound card 22 and the sound source board 34.
[0041]
Specifically, the MIDI information supplied from the sequencer program is supplied to the synthesizer program by the distribution program to generate a tone signal, and the tone signal is output via the sound card 22.
[0042]
On the other hand, the MIDI information supplied from the other electronic musical instrument 36 is supplied to the tone generator board 34 by the distribution program, and the tone generator board 34 generates and outputs a tone signal according to the supplied MIDI information. That is, in this general-purpose computer, when MIDI information is input from the other electronic musical instrument 36 to the MIDI interface 32, the MIDI interface 32 interrupts the CPU 12 in response to the input, and the CPU 12 responds to the interrupt by the sound source board. The processing for outputting a musical tone signal from the 34, that is, the processing for generating and outputting the musical tone signal in accordance with the MIDI information supplied to the MIDI interface 32 by the tone generator board 34 is executed.
[0043]
FIG. 3 shows a flowchart showing the processing contents of the distribution program. When MIDI information is supplied from a sequencer program, or MIDI information is supplied from another electronic musical instrument 36 via the MIDI interface 32. In this case, this distribution program is started first.
[0044]
When the distribution program is activated, it is first determined whether the supplied MIDI information is from the sequencer program or from the MIDI interface 32 (step S302).
[0045]
If it is determined in step S302 that the supplied MIDI information is from the MIDI interface 32, the MIDI information is supplied to the tone generator board 34 (step S304), and the distribution program is terminated.
[0046]
Then, the tone generator board 34 generates a tone signal according to the supplied MIDI information, and outputs the generated tone signal.
[0047]
On the other hand, if it is determined in step S302 that the supplied MIDI information is from the sequencer program, the MIDI information is supplied to the synthesizer program (step S306), and the distribution program is terminated.
[0048]
Next, the processing contents of the synthesizer program will be described. FIG. 4 shows the overall configuration of the synthesizer program, which includes a control task, a sound source task, and an output task.
[0049]
In other words, the control task executes the control program, exchanges MIDI information with the sequencer program, and passes the event flag, sound generation parameter, steady processing parameter, and effect parameter to the sound source task side, and the voice task from the sound source task side. Receive release information.
[0050]
The tone generator task performs processing relating to a plurality of voices (musical tone generation channels) and effects, stores the musical tone signal generated in the wave buffer, and delivers it to the output task. In other words, each of the multiple voices is processed, the output of each voice is transferred to the effect according to the effect feed amount of the sound generation parameter, which will be described later, and the output of each voice to which no effect is added and each voice with the effect added Are finally mixed and stored in the wave buffer and delivered to the output task.
[0051]
In the output task, wave output processing is performed using a wave buffer, and a musical sound signal is output to the outside. Then, the wave buffer used in the output task is returned to the sound source task side.
[0052]
As described above, information exchanged between the control task and the sound source task includes an event flag, a sound generation parameter, a steady process parameter, an effect parameter, and voice release information. Hereinafter, such information will be described.
[0053]
(1) Event flag
There are as many event flags as the number of voices, and the tone generator task receives information (flags) of the start and end of each voice by the event flag from the control task. The event flag is stored in a shared memory accessible from both the control task and the sound source task.
[0054]
(2) Pronunciation parameters
As will be described later, each time the sound source task is activated, the contents of the event flag are searched, and when the sound generation start flag is found, the sound generation starts according to the sound generation parameter. The sound generation parameter includes the following (a) for each voice: To (i).
(A) Parameters related to waveform readout
(Waveform start address, loop related address, pitch)
(B) Pitch LFO related parameters
(Waveform, period, delay, etc.)
(C) Filter LFO related parameters
(Waveform, period, delay, etc.)
(D) Amplifier LFO related parameters
(Waveform, period, delay, etc.)
(E) Pitch envelope parameters
(Attack, decay, sustain, release time and level)
(F) TVF (Time Variable Filter) Envelope-related parameters
(Attack, decay, sustain, release time and level)
(G) TVA (Time Variable Amplifier) envelope-related parameters
(Attack, decay, sustain, release time and level)
(H) Bread
(I) Effect feed amount
The sound generation parameters including these pieces of information are stored in a shared memory accessible from both the control task and the sound source task. Two memory spaces for storing the sound generation parameters are provided for each voice. In other words, when the tone generator task finds the sound start flag in the event flag, the tone generator task switches the memory space referenced by the tone generator task from the previously referenced plane to the other plane and refers to it until now. Pass the face that was left to the control task. Therefore, even when the sound parameters for the same voice number are set, the control task uses the new memory space on the other side that is different from the currently processed surface that is passed from the sound source task, so that the sound source task references The sound generation parameters can be transferred to the sound source task without destroying the parameter area of the currently processed surface. Conversely, it is possible to prevent the sound source task from referring to the parameter area being rewritten on the control task side.
[0055]
(3) Steady processing parameters
The steady process parameters include information shown in the following (a) to (c) for each voice.
(A) Pitch offset
(B) Filter offset
(C) Amplifier offset
These steady processing parameters are stored in a shared memory accessible from both the control task and the sound source task, and serve to service information obtained from an operator such as a pitch bender or a wheel.
[0056]
These steady process parameters are read each time the sound source task is activated, and are synthesized with an envelope or LFO generated inside the sound source task during sound generation. The synthesized data is used as a sound service parameter. Called.
[0057]
The sound generation service parameter controls the waveform readout pitch, the TVF cutoff frequency, and the TVA amplitude.
[0058]
(4) Effect parameters
The effect parameter is information for effect control stored in a shared memory accessible from both the control task and the sound source task, and is passed to the effect. This effect parameter is read every time the sound source task is activated, and if there is a parameter different from the previously read content, it is reflected in the effect.
[0059]
The effect parameters are different depending on the specific contents of the effect, and the specific contents of the effect are not included in the gist of the present invention, and thus detailed description thereof is omitted.
[0060]
(5) Voice release information
The voice release information is returned from the sound source task to the control task, and indicates the number of the voice that has been sounded. Note that the end of pronunciation of each voice is determined by the envelope of TVA. This voice release information does not use the memory space but is passed to the control task by an event.
[0061]
Here, the event flag, sound generation parameter, steady processing parameter, and effect parameter are exchanged using a shared memory, while the voice release information is delivered by event, the difference in the communication method between tasks is that the sound source task This is because asynchronous event information cannot be received from the control task side.
By the way, the above-described control task, sound source task, and output task are provided with a priority (priority: priority), and a task whose processing is not preferably interrupted by another task or the like should have a higher priority. Has been made. In this embodiment, the output task has the highest priority, the sound source task has the next highest priority, and the control task has the lowest priority.
[0062]
Hereinafter, the control task, the sound source task, and the output task will be described in detail with reference to FIGS. 4 and 5.
[0063]
(1) Control task
The control task obtains MIDI information and controls the sound source task, and is activated by receiving MIDI information from the sequencer program and voice release information from the sound source task.
[0064]
Here, for each voice, the control task always manages voice information indicating whether the voice is sounding or vacant. When note-on is received as MIDI information, the voice information is referred to as appropriate. To specify a free voice and use that voice to sound the note-on, set the sound parameters necessary for the note-on for that voice and then set the event flag to start sounding. The voice information of the designated voice is updated. Then, in the sound source task, a new pronunciation is started with the designated voice in accordance with the processing of the control task described above.
[0065]
In addition, when the control task receives note-on as MIDI information, if there is no free voice by referring to the voice information, the control task selects a voice with a low sounding level or starts sounding most recently. Select a voice that seems unnecessary according to a predetermined selection criterion such as selecting a voice, and use the voice to sound the note-on. Is set, and thereafter, the event flag is set to start sounding. In the sound source task, the current pronunciation of the quickly selected voice is muted in accordance with the processing of the control task described above, and then a new pronunciation is started by the selected voice.
[0066]
As described above, in the control task, when note-on is received as MIDI information, voice assignment, so-called assignment processing, is performed.
[0067]
On the other hand, when the control task receives note-off as MIDI information, the control task performs a process of setting the end of sounding in the event flag so as to end sounding for the sounding sound corresponding to the note-off. In the sound source task, the sounding voice corresponding to the note-off is muted in accordance with the processing of the control task described above.
[0068]
As will be described later, the processing of voices in the sound source task is not performed for voices that are not sounded. Therefore, the calculation amount of the sound source task increases / decreases according to the number of voiced voices. Therefore, in the present invention, the maximum number of pronunciations that can be generated can be set in advance according to the ability of the computer and can be changed as appropriate.
[0069]
Further, when the control task receives a message for changing the value of the steady process parameter and the effect parameter as MIDI information, the control task performs a process of delivering the corresponding steady process parameter and the effect parameter to the sound source task.
[0070]
Further, when the voice release information is transferred from the sound source task to the control task, the control task updates the voice information of the designated voice according to the received voice release information.
[0071]
(2) Sound source task
The sound source task is started by a timer every 10 ms (milliseconds), and waveform reading for 10 ms, TVF, TVA, pan, envelope, LFO, and effect processing are performed in as short a time as possible. As described above, since the sound source task performs processing for 10 ms, if the sampling frequency is 32 KHz, processing for 320 samples is performed.
[0072]
Each time the sound source task is activated, the contents of the event flag are searched, and when the sound generation start flag is found, sound generation starts according to the sound generation parameter. At this time, two memory spaces for storing the sound generation parameters are provided for each voice, and when the sound source task finds the sound generation start flag in the event flag, the sound source task is a memory to which the sound source task refers thereafter. The space is switched from the previously referenced plane to another plane, and the previously referenced plane is passed to the control task. Therefore, even when the sound parameters for the same voice number are set, the control task uses the new memory space on the other side that is different from the currently processed surface that is passed from the sound source task, so that the sound source task references The sound generation parameters can be transferred to the sound source task without destroying the parameter area of the currently processed surface.
[0073]
Here, the processing in each voice will be described. In the sound source task, the sound generation parameter and the steady processing parameter stored in the shared memory are taken in each time the sound source task is activated, and the calculation of the envelope and LFO and the steady state parameter are performed. The processing parameter is reflected to perform a calculation for generating the pronunciation service parameter. The pitch of waveform readout, the cutoff frequency of TVF, and the amplitude of TVA (amplification degree) are controlled by the sound generation service parameters generated in this way, and the voice output is delivered to the effect according to the effect feed amount of the sound generation parameter. The output of the voice to which no effect is added is output in stereo with the panpot controlled by the sound parameter pan (FIG. 5A).
[0074]
Then, the output of each voice to which the effect controlled by the effect parameter is added and the output of each voice to which the effect is not added are mixed and stereo output (FIG. 5B), and finally the wave buffer. Is stored and transferred to the output task.
[0075]
FIG. 6 is a flowchart showing the process of generating the sound generation service parameter, and the process is sequentially performed for each voice from the first voice.
[0076]
First, it is checked whether the event flag indicates the start of sound generation (step S602). If the event flag does not indicate the start of sound generation, it is checked whether the event flag indicates the end of sound generation (step S604).
[0077]
If the event flag does not indicate the end of sound generation in step S604, it is checked whether “1” is set in voiceActive, which is a flag indicating sound generation, to indicate that sound generation is in progress (step S606). .
[0078]
On the other hand, if the event flag indicates the end of sound generation in step S604, the envelope is advanced to the release process in response to this (step S608), and the calculated value n1 of the sound service parameter in the previous process is set to n0 ( After step S610), an operation reflecting the envelope, LFO and steady process parameters is performed to generate a sound generation service parameter and set it to the calculated value n1 (step S612). That is, the envelope service and LFO values are changed in accordance with the passage of time of 10 ms from the previous process, and the new envelope, LFO and steady process parameters are synthesized (added or multiplied) to generate the sound generation service parameter. Note that the sound generation service parameter is calculated independently for each of the waveform readout pitch, TVF, and TVA.
[0079]
Even if “1” is set in voiceActive in step S606 and sounding is in progress, the calculated value n1 of the sounding service parameter in the previous process is set to n0 (step S610), and then the envelope, LFO Then, an operation reflecting the steady process parameters is performed to generate a sound generation service parameter and set it to the calculated value n1 (step S612). Note that the sound generation service parameter is calculated independently for each of the waveform readout pitch, TVF, and TVA. By the way, if the event flag indicates that the sound generation is started in step S602, the pointer indicating the sound parameter area used by the control task and the sound source task is updated to indicate the other sound parameter area (step S614), and voiceActive is updated. It is checked whether or not “1” is set to indicate that the sound is being generated (step S616).
[0080]
If voiceActive is set to “1” in step S616 and sounding, the currently sounding voice is rapidly muted (step S618), and the envelope, LFO, and steady processing parameters are reflected. Calculation is performed to generate a sound generation service parameter at the start of sound generation and set it to the calculated value n0 (step S620). Note that the sound generation service parameter is calculated independently for each of the waveform readout pitch, TVF, and TVA.
[0081]
On the other hand, if “1” is not set in voiceActive in step S616 and the sound is not being sounded, the process proceeds to step S620 as it is and the sounding service at the start of sounding is performed by performing an operation reflecting the envelope, LFO and steady process parameters. A parameter is generated and set to the calculated value n0.
[0082]
As described above, in step S620, calculation that reflects the envelope, LFO, and steady process parameters is performed to generate a sound generation service parameter at the start of sound generation, set it to the calculated value n0, and then set “1” to voiceActive. After indicating that the sound is being generated (step S622), the sound processing service parameter is generated by performing an operation reflecting the envelope, the LFO, and the steady processing parameter according to the passage of time of 10 ms from the start of the sound generation, and the calculated value n1 (Step S612). Note that the sound generation service parameter is calculated independently for each of the waveform readout pitch, TVF, and TVA.
[0083]
Then, in step S612, calculation that reflects the envelope, LFO, and steady process parameters is performed to generate a sound generation service parameter and set it to the calculated value n1, so that the current sound generation is used for performing interpolation processing as necessary. A difference d between the calculated value n1 of the service parameter and the calculated value n0 of the previous pronunciation service parameter is obtained (step S624), and it is checked whether the envelope of the TVA has reached the end (step S626).
[0084]
If the TVA envelope has not come to the end in step S626, the event flag is cleared as it is (step S628). If the TVA envelope has come to the end in step S626, voice release information is sent to the control task. Returning (step S630), setting “0” in voiceActive to indicate that no sound is being generated (step S632), the event flag is cleared (step S628).
[0085]
If “1” is not set in voiceActive in step S606 and no sound is being generated, the event flag is cleared as it is (step S628).
[0086]
Then, when the event flag is cleared (step S628), it is checked whether all voices have been processed (step S634). If the processing of all voices is completed as a result of the check in step S634, the processing of this flowchart is terminated. If the processing of all voices is not completed as a result of the check in step S634, the processing returns to step S602 and the subsequent processing is performed. Repeat to process the next voice. That is, the processing of this flowchart is repeated until processing of all voices is completed.
[0087]
Note that the waveform readout, TVF, TVA, pan, and effect feed processing performed in response to the sound generation service parameters are performed for 10 ms each time the sound source task is activated, and then the effect The processing is performed collectively for 10 ms, and a flowchart of this processing is shown in FIG. That is, in the flowchart of FIG. 7, processing is performed sequentially for each voice from the first voice.
[0088]
First, it is checked whether “1” is set in voiceActive to indicate that sounding is being performed (step S702). If “0” is set in voiceActive in step S702 and sounding is not being performed, The process jumps to step S714 for checking whether all voices have been processed, and proceeds to the next voice process. That is, in this flowchart, the processing of voices that are not pronounced is not performed.
[0089]
On the other hand, if “1” is set in voiceActive in step S702 and sound is being generated, the waveform is read so that the sound is generated at the pitch indicated by the sound generation service parameter (step S704). Here, the tone generation service parameters are linearly interpolated to the required accuracy using the difference d obtained in step S624 in the flowchart shown in FIG. That is, assuming that the sampling frequency is 32 KHz, the tone generation service parameter set to the calculated value n0 obtained in step S610 or step S620 of the flowchart shown in FIG. 6 and the difference d obtained in step S624 are used.
n0 + d × k / 320
To obtain an interpolated value. Note that 320 is the number of samples corresponding to 10 ms when the sampling frequency is 32 KHz, k indicates what number of the 320 samples, and is “0” in the first process, and “ These values are 1 ',' 2 ',.
[0090]
Next, the waveform readout output in step S704 is filtered by the TVF (step S706). Here, the cut-off frequency of the TVF is controlled by the sound generation service parameter, and the sound generation service parameter is used by linear interpolation to the required accuracy using the difference d obtained in step S624 of the flowchart shown in FIG.
[0091]
Further, the amplitude of the filtered output in step S706 is controlled by the TVA according to the sound generation service parameter (step S708). Here, the sound generation service parameter is used by linearly interpolating to a required accuracy using the difference d obtained in step S624 of the flowchart shown in FIG.
[0092]
Next, the feed amount to the pan and the effect is controlled by the sound generation parameter, and addition and mixing processing with the tone signal of the other voice shown in FIG. 5B is performed (step S710), and then for 10 ms. Is checked (step S712), and if the process for 10 ms has not been completed, the process returns to step S704 to perform the processes after step S704 on the next sample. That is, the processing from step S704 to step S710 is processing for each sample, and processing for the number of samples for 10 ms is performed.
[0093]
On the other hand, if the processing for 10 ms has been completed, it is checked whether all voices have been processed (step S714). When the processing of all voices is completed by this check, the effect processing is performed for 10 ms collectively (step S716), and the wave buffer storing the musical sound signal for 10 ms subjected to the above processing is passed to the output task. (Step S718), the process of this flowchart is terminated. Here, the wave buffer storing the musical tone signal subjected to the above processing is transferred to the output task, and at the same time, a new wave buffer is received, and the new wave buffer is used for the next processing in steps S702 to S716.
[0094]
On the other hand, if the voice processing is not completed, the process returns to step S702 and the subsequent processing is repeated to perform the next voice processing. That is, the processing of this flowchart is repeated until processing of all voices is completed.
[0095]
(3) Output task
The output task DMA-transfers the stored contents of the wave buffer received from the sound source task to an audio data output device (in the computer shown in FIG. 1, it is the sound card 22). This DMA transfer is started by an interrupt from the audio data output device.
[0096]
Here, an audio data output device such as the sound card 22 (hereinafter referred to as “output device”) has a built-in buffer for temporarily storing audio data to be output, and temporarily stores it at every sampling period. Audio data is output for each sampling. When the amount of audio data remaining in the built-in buffer falls below a predetermined amount, the output task is interrupted, and in response to this interrupt, the output task DMA-transfers audio data for a predetermined sample required by the output device to the output device. To do.
[0097]
The interruption to the output task described above usually takes a cycle of about several ms.
[0098]
That is, in the computer shown in FIG. 1, the output task is activated when an interrupt from the sound card 22 and a wave buffer from the sound source task are received.
[0099]
When the output task receives a wave buffer for 10 ms from the sound source task, the output task returns a free wave buffer to the sound source task. There are four or more wave buffers. One is always output to the output device by the output task, one is used by the sound source task for processing, one is free, and one is waiting for output to the output device. The rotation is repeated so that the rest is free or waiting for output to the output device.
[0100]
In addition, when the crystal | crystallization which is a source oscillator which determines the operating speed separately is mounted in the computer main body and the sound card 22 and both operation speed is not synchronized, the error of the oscillation frequency of crystal | crystallization is not carried out. There will be an error in the operating speed of both.
[0101]
In this case, the speed at which the tone generator task generates the musical tone signal and the speed at which the output device outputs the musical tone signal are different. If the sound source task is faster, the sound source task overwrites new data on the wave buffer that is waiting for output, causing noise, and the sound source task is slower. In this case, the output device outputs all the wave buffer data in the output waiting state, and the sound is interrupted.
[0102]
In order to prevent such a situation, the free state of the wave buffer is constantly monitored, and when the number of free wave buffers is smaller than a predetermined amount, the activation cycle of the sound source task is lengthened (for example, 11 ms). ) If the number of free wave buffers is larger than a predetermined amount, the activation cycle of the activation task is shortened (for example, 9 ms). Of course, even in this case, the number of musical tone signal samples generated by one activation of the sound source task is set constant (for example, for 10 ms). By performing such processing, a certain amount of tone signal waiting for output is always stored in the wave buffer, so that it is possible to prevent the occurrence of noise and sound interruption as described above.
[0103]
By the way, in this synthesizer program, communication between the sound source task and the control task for controlling it is realized by sharing the memory space.
[0104]
That is, as shown in FIG. 8, there are two memory spaces (memory space A and memory space B in FIG. 8) shared by the sound source task and the control task, and they are used alternately. .
[0105]
Specifically, the control task side prepares parameters necessary for sound generation in the sound generation parameter area of the memory space A, and notifies the sound source task side by setting a flag that the sound generation parameters have been prepared. On the sound source task side, when the sound generation parameter area in the memory space A is captured, the sound generation parameter area on the other side (memory space B) is transferred to the control task side. The control task side prepares a parameter for the newly generated sound generation parameter area in the memory space B when the sound is newly generated with the same voice.
[0106]
As a result, even if the sound source task side is sounding and using the sound generation parameter, the control task side can set a new sound generation parameter without worrying about it.
[0107]
If only one pointer is provided as a management method for the two memory spaces, the control task stores a new parameter in one memory space pointed to by this pointer, and the tone generator task is stored in the other memory space. Refers to parameters stored in space. Alternatively, if a pointer indicating the memory space in which the control task writes the parameters and a pointer indicating the memory space in which the sound source task refers to the parameters are provided separately, the memory spaces indicated by these pointers are opposite to each other. do it.
[0108]
In this general-purpose computer, the user can give various instructions to the tone generator board 34 and the synthesizer program via the keyboard 28. When an instruction is given from the keyboard 28, a processing program corresponding to the type of the instruction is activated, and control information corresponding to the instruction is supplied to the tone generator board 34 and the synthesizer program. The setting state of the synthesizer program is changed according to the control information.
[0109]
Here, the above-described setting state includes, for example, tuning (setting of the pitch of a musical tone generated when the center C of the keyboard 28 is designated) and generated musical tone (“piano”, “organ”, (Designation of timbres corresponding to natural instruments such as “violin” and “trumpet”).
[0110]
By the way, with respect to the above-described setting state, it is more advantageous in terms of performance if the sound source board 34 and the synthesizer program are in the same setting state. That is, if the tuning is different between the tone generator board 34 and the synthesizer program, even if the center C of the same keyboard 28 is specified, musical tones having different pitches are generated between the tone generator board 34 and the synthesizer program. This will cause inconvenience in performance. For this reason, when a change in tuning is instructed by changing the setting state, the tuning instruction processing program shown in FIG. 9 is activated, and the tone generator board 34 and the synthesizer program correspond to the tuning value that is the same pitch designation value. The pitch control information to be supplied is supplied at the same time.
[0111]
That is, when the tuning instruction processing program is activated, pitch control information corresponding to the tuning value specified by the synthesizer program is supplied (step S902), and further, the sound corresponding to the tuning value specified by the tone generator board 34 is supplied. After the high control information is supplied (step S904), the processing of this tuning instruction processing program is terminated.
[0112]
In this general-purpose computer, the sound source board 34 and the synthesizer program have different sound sources, so the sound source board 34 and the synthesizer program do not generate exactly the same tone, and the sound source board 34 and the synthesizer program have the same tone. Even though the timbre indicated by the name of the timbre is actually a timbre that is similar to the timbre. Therefore, in this general-purpose computer, the timbre is specified by the timbre number, and therefore, the tone generator board 34 and the synthesizer program correspond to different timbre numbers with respect to the timbre indicated by the same timbre name.
[0113]
Therefore, in this general-purpose computer, when an instruction to change the timbre is given to either the tone generator board 34 or the synthesizer program, the timbre instruction processing program shown in FIG. The tone color is also set to the tone generator board 34 or the synthesizer program that is not instructed to change the tone color. That is, the memory 14 stores a correspondence table of similar timbres of the other party for the timbres of the tone generator board 34 and the synthesizer program, and instructs the timbre change to either the tone generator board 34 or the synthesizer program. In the case where the timbre is made, referring to the above correspondence table, the other party is also instructed to change the timbre to a timbre similar to the timbre specified.
[0114]
The timbre instruction processing program shown in FIG. 10 will be described in more detail. First, when the timbre instruction processing program is activated, it is determined which timbre change is instructed between the tone generator board 34 and the synthesizer program ( Step S1002).
[0115]
If it is determined in step S1002 that an instruction to change the tone color of the tone generator board 34 has been given, tone color change control information indicating a change to the indicated tone color is supplied to the tone generator board 34 (step S1004). The timbre change control information indicating the change to the timbre similar to the timbre instructed with reference to the correspondence table is supplied to the synthesizer program (step S1006), and the processing of the timbre instruction processing program is terminated.
[0116]
On the other hand, if it is determined in step S1002 that an instruction to change the timbre of the synthesizer program has been issued, the timbre change control information indicating the change to the instructed timbre is supplied to the synthesizer program (step S1008). The timbre change control information indicating the change to the timbre similar to the timbre instructed with reference to the correspondence table is supplied to the sound source board 34 (step S1010), and the processing of the timbre instruction processing program is terminated.
[0117]
In the above-described embodiment, all MIDI information supplied from the sequencer program is generated by the synthesizer program. However, the synthesizer program receives multiple channels of MIDI information and supports each channel. When a plurality of parts can generate different tones of music at the same time, that is, when a so-called multitimbral sound source operation is realized by a synthesizer program, a specific channel supplied from the sequencer program This MIDI information may not be supplied to the synthesizer program.
[0118]
As described above, when the MIDI information of a specific channel supplied from the sequencer program is not supplied to the synthesizer program, the part corresponding to the specific channel is not played by the other electronic musical instrument 36. A musical tone signal can be generated by the tone generator board 34 by playing. That is, the performer can perform the performance by himself / herself with respect to a specific part while listening to the automatic performance by the sequencer program.
[0119]
In such a case, in all parts, a tone signal is generated by the synthesizer program based on the sequencer program, and a tone signal generated by the performer is generated by the tone generator board 34 for a specific part specified by the performer. As for the other parts, it may be possible to arbitrarily switch the mode in which the tone signal is generated by the synthesizer program based on the sequencer program.
[0120]
Further, in the above-described embodiment, all the MIDI information supplied from the other electronic musical instrument 36 is generated by the tone generator board 34. However, the MIDI information supplied from the other electronic musical instrument 36 is used. A mode in which a tone signal is generated by the tone generator board 34 and a mode in which a tone signal is generated by a synthesizer program based on MIDI information supplied from another electronic musical instrument 36 are set, and these modes can be switched arbitrarily. May be.
[0121]
Furthermore, in the above-described embodiment, which one of the synthesizer program and the tone generator board generates the tone signal is controlled by which one of the MIDI information is supplied. Also good. For example, MIDI information is supplied to both the synthesizer program and the tone generator board, but a volume control signal for setting the volume to 0 may be supplied to those who do not generate a musical tone signal.
[0122]
【The invention's effect】
As described above, the performance information processing method according to the present invention should emit sound corresponding to the real-time performance of the performer when a synthesizer is constructed by software on a general-purpose computer such as a personal computer or a workstation. There is an excellent effect that the musical sound can be emitted promptly without requiring a large delay time from the pronunciation instruction.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a hardware configuration of a typical general-purpose computer on which a program according to a performance information processing method of the present invention can be installed.
FIG. 2 is a conceptual diagram of a performance information processing method according to the present invention.
FIG. 3 is a flowchart showing processing contents of a distribution program.
FIG. 4 is a conceptual diagram showing an overall configuration of a synthesizer program.
5A and 5B are conceptual diagrams showing processing in a sound source task, where FIG. 5A shows processing for one voice, and FIG. 5B shows mixing of voice and effect.
FIG. 6 is a flowchart showing a sound generation service parameter generation process in a sound source task.
FIG. 7 is a flowchart showing processing of waveform readout, TVF, TVA, panning and effect feed amount performed in response to a sound generation service parameter in a sound source task.
FIG. 8 is an explanatory diagram showing a state in which two memory spaces shared by a sound source task and a control task are provided and used alternately.
FIG. 9 is a flowchart showing the processing contents of a tuning instruction processing program.
FIG. 10 is a flowchart showing the processing contents of a timbre instruction processing program.
[Explanation of symbols]
10 buses
12 CPU
14 memory
16 disk controller
18 Video card
20 I / O card
22 Sound card
24 hard disk
26 Display device
28 keyboard
30 mice
32 MIDI interface
34 Sound board
36 Other electronic musical instruments

Claims (5)

Sound source processing that constitutes the first sound source means by operating a general-purpose processing device based on software;
A discrimination process for discriminating whether the supplied performance information is due to an automatic performance or a performer's performance;
When it is determined that the performance information supplied by the determination processing is based on the automatic performance, a musical tone signal is generated by the sound source processing based on the supplied performance information, and the performance information supplied by the determination processing is If it is determined that the performance is performed by the performer, a musical tone signal is generated by the second sound source means composed of a dedicated hardware sound source independent of the general-purpose processing device based on the supplied performance information. A performance information processing method characterized by performing a performance control process. The discriminating process discriminates that the performance information supplied from the sequencer program of the general-purpose processing device is based on automatic performance, and discriminates that the performance information supplied from the performance information input interface is based on the performance of the performer. Item 1. The performance information processing method according to Item 1. The performance information processing method according to claim 1,
Furthermore, the performance information processing method characterized by performing the process which supplies common control information to a said 1st sound source means and a said 2nd sound source means according to a player's instruction | indication. Sound source processing that constitutes the first sound source means by operating a general-purpose processing device based on software;
A determination process for determining whether the supplied performance information is based on an automatic performance or a performance of a performer;
A setting process for setting the first mode and the second mode as processing modes when it is determined that the performance information supplied by the determination process is based on an automatic performance;
When it is determined that the performance information supplied by the determination processing is based on the automatic performance and the first mode is set by the setting processing, the sound source processing is performed based on the supplied performance information. Is generated when the musical performance signal is generated by the determination process and it is determined that the performance information supplied by the determination process is based on the automatic performance and the second mode is set by the setting process. When a musical tone signal is generated by the sound source processing based on a part of the performance information and the performance information supplied by the determination processing is determined to be due to the performance of the performer, the supplied performance information is And a performance control process for generating a musical tone signal by a second sound source means comprising a dedicated hardware sound source independent of the general-purpose processing device. Kanade information processing method. Sound source processing that constitutes the first sound source means by operating a general-purpose processing device based on software;
A determination process for determining whether the supplied performance information is based on an automatic performance or a performance of a performer;
A setting process for setting the first mode and the second mode as processing modes when it is determined that the performance information supplied by the determination process is based on the performance of the performer;
When it is determined that the performance information supplied by the determination processing is based on the automatic performance, a musical tone signal is generated by the sound source processing based on the supplied performance information, and the performance information supplied by the determination processing is When it is determined that the performance is performed by the performer and the first mode is set by the setting process, the dedicated processing device is independent from the general-purpose processing device based on the supplied performance information. When the musical tone signal is generated by the second sound source means constituted by the hardware sound source and the performance information supplied by the determination processing is determined to be due to the performance of the performer, the setting processing performs the first When the mode 2 is set, a performance control process for generating a tone signal by the sound source process based on the supplied performance information is performed. Information processing method.

Images