JP6531432B2 - Program, sound source device and acoustic signal generation device - Google Patents

Description

  The present invention relates to a program for causing a computer to function as a sound generation control device, a sound source device that outputs an acoustic signal based on an instruction from the sound generation control device, and an acoustic signal generation device including a sound generation control unit and a sound source unit.

Conventionally, based on sequence data in which events indicating the tone pitch and tone intensity of a tone are recorded in association with the execution timing, the tone generation instruction of each tone is stored in the tone generator circuit at the execution timing of the event previously recorded. Technologies to supply are known. Then, when the sound source circuit that has received the output outputs an acoustic signal based on the sound generation instruction, automatic performance based on the sequence data can be performed.
Such a technique is described, for example, in Patent Document 1.

Japanese Patent Application Laid-Open No. 8-30269

  However, in the prior art as described in Patent Document 1, the method of controlling the timing of supplying the sound generation instruction to the tone generator circuit or the like is such that the CPU executes interrupt processing at a constant cycle and executes it at the timing of the interrupt processing. If there is an event that has passed the timing, a tone generation instruction according to the event is supplied to the sound source circuit or the like.

FIG. 10 shows an outline of the tone generation timing according to this process.
In FIG. 10, it is assumed that the note-on (sound generation start) event is to be executed in the sequence data 9.95 seconds after the start of reproduction. Further, it is assumed that the timing of the interrupt processing executed by the CPU is every 10 ms (milliseconds).

  In the example of FIG. 10, the CPU performs interrupt processing every 10 ms after the start of sequence data reproduction, and determines whether there is an event that has passed the execution timing. It is determined that there is no corresponding event until the processing with an elapsed time of 9.99 seconds. After that, it is determined that the execution timing of the note-on event has passed in the processing of the elapsed time 10.00 seconds, and the tone generation start instruction corresponding to the note-on event is supplied to the tone generator circuit.

The sound source circuit monitors the sound generation instruction at intervals shorter than the CPU (for example, at intervals of 44.1 kHz ≒ 0.022675 ms which is the sampling frequency of the audio signal to be output), and when the sound generation start instruction is supplied from the CPU Sound generation can be started (within a negligible delay time).
However, in the example of FIG. 10, the supply timing of the tone generation start instruction from the CPU to the tone generator circuit is delayed by 5 ms from 9.995 seconds set in the sequence data. Therefore, this delay also occurs at the timing when the sound generation is actually started.

  As can be seen from FIG. 10, even if the interrupt processing is correctly performed at intervals of 10 ms, a delay of 10 ms, which is the same as the interrupt period, occurs at the maximum at the sound generation start timing of each sound. In addition, when the processing load on the CPU is high, it is conceivable that the start of interrupt processing may be delayed, or that interrupt processing at certain timing may not be executed at all. In such a case, a delay larger than the interrupt cycle may occur. In addition, the amount of delay will change according to the situation at the time of reproduction.

  The present invention has been made in view of such circumstances, and its object is to enable generation of an audio signal according to sequence data with higher timing accuracy.

  A program according to the present invention is a program for causing a computer to function as a sound generation control device for instructing a sound source device to generate a sound in order to achieve the above object, and the computer is arranged at predetermined time intervals. An instruction to supply a sound generation instruction to be executed within a predetermined time from the timing according to sequence data prepared in advance or a predetermined amount of sound generation instruction to be executed after that timing along with the specification of the execution timing of the sound generation instruction It is a program for functioning as a supply means.

In such a program, the designation of the execution timing for the sound source device may be performed in units smaller than the predetermined time interval.
Further, when the instruction supply means detects performance data other than those included in the previously prepared sequence data, the instruction to execute the sound generation instruction corresponding to the performance data is designated immediately. And means for supplying the sound source device.

Further, the sound source device of the present invention corresponds to the sound generation instruction corresponding to the sound generation instruction, storing means for storing the sound generation instruction and the execution timing of the sound generation instruction, and the sound signal output according to the sound generation instruction stored in the storage means. and provided with control means for starting the execution timing stored in the storage means, sound the control means, at a predetermined period based on the sampling period of the output acoustic signal, which is stored in the storage means It is checked whether or not the execution timing of the instruction has arrived, and the function of the control means is realized by hardware different from the processor that realizes the function of supplying the sound generation instruction and the execution timing to be stored in the storage means. Further, a timer and means for providing information on time counted by the timer to the processor are provided .

  According to another sound source device of the present invention, storage means for storing a sound generation instruction and an execution timing of the sound generation instruction, and sound signal output according to the sound generation instruction stored in the storage means. And a control means for starting at an execution timing stored in the storage means, and a processor for realizing a function of supplying a sound generation instruction and an execution timing to be stored in the storage means. And a timer and means for providing the information on the time counted by the timer to the processor.

Further, an acoustic signal processing apparatus according to the present invention is an acoustic signal generation apparatus including a sound generation control unit for instructing a sound source unit to make a sound generation, and the sound source unit, wherein the sound generation control unit has a predetermined time interval timing. Each of the above sound source units is provided with an instruction supply means for supplying a sound generation instruction to the sound source unit together with the specification of the execution timing of the sound generation instruction according to the sequence data prepared in advance. A storage means for storing a sound generation instruction supplied from the sound generation control unit and an execution timing of the sound generation instruction, and an acoustic signal output according to the sound generation instruction stored in the storage means correspond to the sound generation instruction. Control means for starting the execution timing stored in the storage means.
Further, the present invention can be practiced in any mode other than the above-described modes, such as an apparatus, a system, a method, a program, and a recording medium.

  According to the above configuration, it is possible to generate an acoustic signal according to music data with higher timing accuracy.

It is a figure which shows the hardware constitutions of the electronic music apparatus which is one Embodiment of the acoustic signal production | generation apparatus of this invention. It is a figure which shows the example of the waveform data memorize | stored in waveform memory. It is a figure which shows the structure of the acoustic signal output function with which the electronic music apparatus shown in FIG. 1 is provided. It is a figure which shows the example of sequence data. It is a figure which shows the example of the data stored in the register | resistor of a sound source circuit. It is a figure for demonstrating generation of data for pre-reading. It is a flowchart of the main processing which CPU of the electronic music apparatus shown in FIG. 1 performs. Similarly, it is a flowchart of interrupt processing. It is a figure for demonstrating the effect of embodiment of this invention. It is a figure for demonstrating the problem of a prior art.

Hereinafter, a mode for carrying out the present invention will be specifically described based on the drawings.
First, FIG. 1 shows a hardware configuration of an electronic music apparatus which is an embodiment of an audio signal generation apparatus of the present invention.
As shown in FIG. 1, the electronic music apparatus 10 includes a CPU 11, a ROM 12, a RAM 13, a storage device 14, a communication I / F 15, a detection circuit 16, a display circuit 17 and a sound source circuit 18. ing. Further, the operator 21 is connected to the detection circuit 16, the display 22 is connected to the display circuit 17, the DAC (digital analog converter) 23 is connected to the sound source circuit 18, and the sound system 24 is connected to the DAC 23. The waveform memory 25 is connected to the sound source circuit 18 and provided.

Among them, the CPU 11 is a control unit that generally controls the operation of the electronic music apparatus 10, and executes a required program stored in the ROM 12 or the storage device 14 to control each part of the electronic music apparatus 10. It realizes various functions including the function of the sound source control unit 116.
The ROM 12 is a non-rewritable non-volatile storage unit that stores programs to be executed by the CPU 11 and data that should be left even after the power is turned off. Note that it may be configured by a rewritable non-volatile storage unit such as a flash ROM.
The RAM 13 is storage means used as a work memory of the CPU 11.

  The storage unit 14 is a storage means for storing sequence data (such as song data and accompaniment style data) and audio waveform data (such as audio song data and accompaniment style data) in MIDI (Musical Instrument Digital Interface (registered trademark) format). . Here, the song data is data indicating one piece of music. The accompaniment style data is data indicating an accompaniment phrase to be attached to the user's performance. Further, the storage device 14 can be configured by an arbitrary storage medium such as a hard disk, a flexible disk, a compact disk (CD), a digital multipurpose disk (DVD), a flash memory, and a drive thereof. The storage medium may be removable or may be incorporated in the electronic music apparatus 10.

  The communication I / F 15 is an interface for connecting an external device to the electronic music apparatus 10, and can adopt an interface of any standard regardless of wired or wireless. It is conceivable that a MIDI device for generating performance data of MIDI format is connected to the communication I / F 15 automatically or automatically according to the user's operation, and MIDI data is input from this.

The detection circuit 16 is an interface for connecting the operating element 21 for receiving an operation from the user, detecting the operation received by the operating element 21, and supplying the CPU 11 with the operation. In addition to setting operators such as keys, buttons, sliders, and touch panels for setting and instructing the operation of the electronic music apparatus 10, the operators 21 include performance operators such as a keyboard and a pedal for performing a performance operation. May be included.
The display circuit 17 is an interface for connecting the display 22 and causing the display 22 to perform display under the control of the CPU 11. The display 22 is a display means for displaying the setting state and the operation state of the electronic music apparatus 10, and can be configured by arbitrarily combining a lamp, a display, and the like.

  The sound source circuit 18 is a waveform synthesis circuit configured by an LSI that is hardware different from the CPU 11 and functions as a sound source unit of the electronic music apparatus 10. Here, the sound source circuit 18 is configured as an auxiliary circuit for waveform rendering to realize the function of a PCM (Pulse Code Modulation) sound source. More specifically, the sound source circuit 18 generates and outputs audio waveform data, which is an acoustic signal, based on waveform data of an instrument sound or the like stored in the waveform memory 25 and a tone generation instruction supplied from the CPU 11. The tone generator circuit 18 can also reproduce song data and style data in the form of audio waveform. The CPU 11 and the sound source circuit 18 may be configured by an LSI in which the CPU core and the sound source core are incorporated in one package.

The DAC 23 has a function of converting digital audio waveform data output from the sound source circuit 18 into an analog audio signal and outputting the analog audio signal to the sound system 24.
The sound system 24 is a sound generation unit that outputs sound based on an analog sound signal, and can be configured by, for example, an amplifier and a speaker.

The waveform memory 25 is storage means for storing waveform data used by the sound source circuit 18 for generating audio waveform data. The waveform memory 25 can be configured as, for example, a ROM, and may be a PCM waveform obtained by sampling a musical instrument of a natural instrument such as a piano or guitar at a predetermined cycle, an artificially synthesized musical tone, or a PCM waveform such as a non-musical tone such as a sound effect. Waveform data is stored.
More specifically, as shown in FIG. 2, the waveform memory 25 arranges waveform sample values sampled at a predetermined sampling rate (for example, 44.1 kHz) in chronological order according to the progress of the address. Remember to

Further, in the waveform memory 25, an address at which waveform data is stored is managed for each kind of tone color such as a musical tone, and by specifying the address, waveform data such as a musical tone having a desired tone color is read out. be able to. The waveform memory 25 may store not only waveform data of a single tone but also waveform data of a musical phrase consisting of multiple tones or a plurality of tones. Further, the waveform memory 25 may be configured by a RAM so that waveform data can be read from the external storage medium to the waveform memory 25 and stored.
The electronic music apparatus 10 described above can be configured, for example, as an electronic musical instrument provided with a performance operator or a sequencer that reproduces MIDI sequence data.

  Next, FIG. 3 shows a configuration of an audio signal output function provided in the electronic music apparatus 10. Note that FIG. 3 shows a function for the CPU 11 to control the sound source circuit 18 to generate an acoustic signal in accordance with MIDI sequence data or performance data, and the sound source circuit 18 generates the acoustic signal according to the control. Is a function to output. Here, the function on the CPU 11 side is realized by the CPU 11 executing a required program, and the function on the sound source circuit 18 side is realized by dedicated hardware. However, as described later, it is not limited to this configuration.

As shown in FIG. 3, the CPU 11 has functions of a music data acquisition unit 111, a prefetch data generation unit 112, a performance operation detection unit 113, a performance data generation unit 114, a performance data acquisition unit 115, a sound source control unit 116, and a timer 117. Equipped with
Among these, the music data acquisition unit 111 has a function of acquiring sequence data to be reproduced according to a reproduction instruction from the user. The sequence data can be read from the storage device 14 or downloaded from an external device through the communication I / F 15 and acquired.

The configuration of the sequence data is, for example, as shown in FIG. That is, the sequence data is data in which event data is arranged in time series in association with a timing value indicating a timing at which a process corresponding to the event should be performed. This event includes note-on indicating the start of sound generation, note-off indicating the sound generation stop, control change indicating designation of volume etc., program change indicating designation of timbre, and tempo change indicating designation of tempo. Further, the timing value can be designated by, for example, the count value of the clock in units of 1/480 of the quarter note. The timing value may be represented by a relative clock value indicating an elapsed time from a previous event, or may be represented by an absolute clock value indicating an elapsed time from the start of playing.
The above sequence data can represent the contents of performance such as music and accompaniment.

  Next, based on the sequence data acquired by the music data acquisition unit 111, the pre-reading data generation unit 112 writes various instructions such as a sound generation instruction and a setting change instruction based on an event in the sequence data in the register 181 of the sound source circuit 18. It has a function of generating pre-read data for detecting timing. The pre-reading data is obtained by replacing each timing value in the sequence data with a count value indicating the same time in the timer 182 of the sound source circuit 18. The details will be described later.

  In the present specification, the "sound generation instruction" includes not only the sound generation start instruction but also the sound generation stop instruction. Further, the "setting change instruction" includes all instructions such as a volume setting instruction other than the sound generation instruction and the like which are to be written in the sound source circuit 18 regarding the sound generation. In the "setting change instruction", the CPU 11 may perform processing corresponding to an event like tempo setting, and there may be no need to write an instruction corresponding directly to the event in the tone generator circuit 18.

The performance operation detection unit 113 has a function of detecting a performance operation performed on a performance operation element in the operation element 21.
The performance data generation unit 114 has a function of generating performance data indicating a performance operation detected by the performance operation detection unit 113 and supplying the performance data to the sound source control unit 116. This performance data has the same format as an event included in sequence data, but should be executed immediately after generation, and has no information on execution timing.

  In addition, the performance data acquisition unit 115 has a function of acquiring performance data transmitted from an external device through the communication I / F 15 and supplying the data to the sound source control unit 116. This performance data also has the same format as the event included in the sequence data, but should be executed immediately after acquisition, and has no information on execution timing. As a transmission source of performance data, a MIDI sequencer or an external performance operation accepting device (keyboard or the like) can be considered.

The sound source control unit 116 has a function of performing interrupt processing at predetermined time intervals (for example, 2 ms intervals) in accordance with the time counted by the timer 117. The interrupt process includes various instructions such as an event in the sequence data and a sound generation instruction and a setting change instruction based on the performance data supplied from the performance data generation unit 114 or the performance data acquisition unit 115 as well as the execution timing. It is a process for supplying the circuit 18.
The supply of various instructions such as a sound generation instruction and a setting change instruction is performed for an event to be executed within a predetermined time from the execution time of the process. For example, if the execution time of an interrupt process is 9.800 seconds from the start of reproduction and the predetermined time is 5 ms, an event and a corresponding sound generation instruction to be executed between 9.800 seconds and 9.805 seconds Make a write supply.

In addition, the performance data supplied from the performance data generation unit 114 or the performance data acquisition unit 115, that is, the performance data detected other than those included in the sequence data being reproduced, should be immediately instructed. Therefore, when the sound source control unit 116 detects these performance data in the interrupt process, the sound source control unit 116 supplies a corresponding instruction to the sound source circuit 18 along with a designation to immediately execute.
The timer 117 is a clock means for measuring time. It is not necessary for the sound source control unit 116 to be dedicated.

On the other hand, the tone generator circuit 18 includes a register 181, a timer 182, a read circuit 183, and a tone control circuit 184.
Among these, the register 181 is a storage unit that stores instructions and settings relating to sound signal generation performed by the sound source circuit 18. The information to be written in the register 181 includes the information included in the sound generation instruction supplied from the sound source control unit 116 as it is, the sound generation instruction supplied from the sound source control unit 116, and the setting made in the sound source circuit 18. On the basis of the above, there is the one that should determine the information to be written on the sound source circuit 18 side.

Here, FIG. 5 shows the configuration of data to be stored in the register 181.
The tone generator circuit 18 includes a plurality of tone generation channels (channels), but as shown in FIG. 5, the register 181 includes a tone generation control register (A) provided for each tone generation channel and settings common to all tone generation channels. There are a change event register (B) and a setting register (C).

  Among these registers, the tone generation control register is for writing a tone on instruction corresponding to the note on event and the note off event. When the sound source control unit 116 supplies the sound source circuit 18 with a sound generation start instruction corresponding to the note-on event, the sound source control unit 116 searches the sound generation ch not being used for sound generation and selects one of the found sound generation ch Designate as the pronunciation channel used for the corresponding pronunciation. Further, when the sound source control unit 116 supplies the sound source circuit 18 with a sound generation stop instruction corresponding to the note off event, the sound source control unit 116 designates a sound generation ch performing sound generation corresponding to the corresponding note on event as a target of sound generation stop. . Then, the tone generator circuit 18 writes necessary information in the tone generation control register corresponding to the designated tone generation channel, based on the tone generation instruction and the execution timing supplied from the tone generator control unit 116.

Further, the sound generation control register has an area for storing the "timer value", the "sound generation instruction", the "waveform address" and the "other parameters".
Among these, “timer value” is information indicating the timing of executing the written sound generation instruction. This timing may be designated by the count value of the timer 182. However, as long as conversion processing necessary for comparison with the count value of the timer 182 can be performed, it may be specified in another unit such as a microsecond unit. Further, it is preferable that the sound source control unit 116 supply the execution timing of the sound generation instruction by the count value of the timer 182 so that the sound source circuit 18 can set the supplied value to the register 181 as it is. Absent. When the sound generation instruction is to be immediately executed, "0" may be set as the timer value. In this way, when the reading circuit 183 determines the execution timing of the corresponding sound generation instruction, it is determined that the current time is "past from the current time" and the execution timing has come.

The “pronunciation instruction” is information indicating the distinction between the onset of pronunciation and the termination of pronunciation. A value indicating whether the sound generation instruction supplied from the sound source control unit 116 is note-on (sound generation start) or note-off (sound generation stop) may be set in the “sound generation instruction”.
The “waveform address” is information indicating a reading position at the time of reading out waveform data from the waveform memory 25 for generating (sounding) an acoustic signal according to a sound generation instruction. The read position is specified by, for example, a start address, an end address, and a loop start and end address. These addresses are determined by the setting of the tone color used for tone generation, and when the tone generation instruction is supplied from the sound source control unit 116, the tone source circuit 18 sets the value corresponding to the setting of the tone color at that time.

  "Other parameters" are the reading speed for specifying the pitch of the tone to be generated, the expansion ratio of time stretch, the velocity for specifying the sound generation strength of the sound, etc., and the envelope to be applied to the waveform data read from the waveform memory 25. It is information that defines effects and the like.

On the other hand, the setting change event register is for writing setting change contents to be executed in accordance with the setting change instruction when the setting change instruction related to the setting change for the entire tone generator circuit 18 or each tone generation channel is detected. is there. The setting for the entire tone generator circuit 18 is stored in a setting register.
Then, the setting change event register has an area for storing a plurality of sets of “timer value” and “change content”.

Among these, "timer value" is information indicating the timing of executing the corresponding "change content", and has the same meaning as "timer value" in the tone generation control register.
The “change content” is information indicating the change content of the setting of the entire tone generator circuit 18 or MIDIch. The tone generator circuit 18 writes information such as tone color change and signal processing change according to the setting change instruction supplied from the sound source control unit 116 here.

  Further, the setting register has an area for storing setting information for the entire tone generator circuit 18 or each MIDI channel. This setting may be, for example, setting of a tone color used for sound generation, setting of signal processing performed by the tone control circuit 184, or the like. The read circuit 183 stores the “change content” corresponding to the setting change event register in the setting register by writing the “change content” at the time indicated by the “timer value” according to the information stored in the setting change event register. Update existing information.

  Returning to the description of FIG. 3, the timer 182 is clocking means for generating the execution timing of the tone generation instruction and the setting change instruction in the tone generator circuit 18 and supplying the timing to the readout circuit 183. The timer 182 also has a function of counting the elapsed time (counting up in this case, but counting down may be used) in units of the cycle of execution timing. Further, the timer 182 is reset by an instruction from the sound source control unit 116 at the start of reproduction of sequence data.

  The execution timing of the sound generation instruction and the setting change instruction in the tone generator circuit 18 is periodically generated at an interval sufficiently shorter than the interval of the above-described interrupt processing executed by the tone generator control unit 116 in the CPU 11. For example, it is conceivable to set the same cycle (0.022675 ms cycle) as 44.1 kHz which is the cycle (DAC cycle) of the DA conversion processing in the DAC 23.

  In this way, processing of sequence data can be performed at timing completely synchronized with the processing period of the audio waveform output from the sound source circuit 18. However, not limited to this cycle, another cycle may be adopted. However, it is desirable that the cycle with a clear relationship with the DAC cycle, such as a natural number multiple with the DAC cycle, 1 / natural number multiple, for example 2 times, 1/2 times, etc. When using a cycle unrelated to the DAC cycle, specify a time that is finer than an integer as the pronunciation time, and use an interpolator to obtain a finer value than an integer to reproduce, thereby processing sequence data with higher time accuracy. It can be carried out.

  The reading circuit 183 has a function of control means for performing processing relating to the execution of the sound generation instruction and the setting change instruction stored in the register 181 at each execution timing supplied from the timer 182. More specifically, the reading circuit 183 converts the “timer value” stored in the tone generation control register and control event register of each tone generation channel and the count value of the timer 182 (if necessary, conversion processing is performed. Go) and compare. Then, if the stored timer value (or conversion value) is equal to or less than the count value of the timer 182, the "sound generation instruction" or "control content" corresponding to the timer value is executed.

  When the "pronunciation instruction" is the start of sound generation, the read circuit 183 starts reading of the waveform data stored in the waveform memory 25 from the read start address stored in the corresponding "waveform address", When the end address is reached, the reading is ended. As necessary, reading from the loop start address to the loop end address is repeated, the reading address is advanced at a speed corresponding to the pitch included in the "other parameters", interpolation processing is performed, and reading speed Perform time stretching of waveform data according to parameters.

  Once readout circuit 183 starts readout of waveform data, readout of waveform data is performed every DAC cycle until generation of sound is forcibly stopped due to a sound generation stop instruction or a volume decrease. repeat. Accordingly, if the execution timing of the sound generation instruction and the DAC cycle have the same cycle, these timing managements can be shared, which is preferable because the circuit configuration of the sound source circuit 18 can be simplified. The waveform data is not necessarily read out every DAC cycle, but burst read is performed once in several DAC cycles to cache the waveform data, and the cached waveform data is pitched up every DAC cycle. The necessary number of samples may be read out accordingly.

In addition, when the “sound generation instruction” is the sound generation stop, the reading circuit 183 stops reading the waveform data regarding the corresponding sound generation channel, or shifts the envelope applied to the waveform data to the release state, etc. Perform the action required to stop.
Further, when the “change content” is executed, the read circuit 183 updates the setting stored in the setting register in accordance with the content.

The sound generation instruction supplied from the sound source control unit 116 and written in the register 181 based on the sequence data being reproduced by the function of the read circuit 183 is a sound generation instruction in the sound source circuit 18 whose interval is much shorter than interrupt processing in the CPU 11 Is executed at the timing specified in the sequence data.
In addition to the above, the readout circuit 183 also has a function of supplying the current count value of the timer 182 to the sound source control unit 116 in response to a request from the sound source control unit 116.

  The tone control circuit 184 has a function of performing signal processing on the waveform data read out by the read out circuit 183 by applying a predetermined operation to impart a musical effect or controlling the volume according to an envelope. The signal processing performed by the tone control circuit 184 is controlled in accordance with the settings stored in the setting register. Further, the process may be common to all the tone generation channels, or may be different processing for each tone generation channel, or both of them may be included. The tone control circuit 184 may be configured by a general purpose DSP (digital signal processor) in addition to being configured by a dedicated processing circuit. It is also conceivable to omit the tone control circuit 184.

  The DAC 23 has a function of DA converting the audio waveform data processed by the tone control circuit 184 output from the tone source circuit 18 into analog tone signals every DAC cycle.

Next, generation of prefetching data to be executed by the prefetching data generation unit 112 will be described using FIG.
As described above, the pre-read data can be generated by replacing each timing value in the sequence data shown in FIG. 4 with a count value indicating the same time in the timer 182 of the sound source circuit 18.

Here, the timing value in the sequence data is represented by a clock value, and the time (absolute time) of the corresponding timing value with reference to the reproduction start time is determined according to the tempo value specified when reproducing the sequence data. fluctuate.
Here, one clock length is set to 1/480 of a quarter note, and the count cycle of the timer 182 is set to 1/44 second. Also, let T be the designated tempo value.
In this case, since the time length of the quarter note is 60 / T seconds, one clock length is 60 / 480T = 1 / 8T seconds. Therefore, when one clock length is converted to the count value of the timer 182, (1 / 8T) / (1/44100) = 44100 / 8T.

Therefore, as shown in FIG. 6, the corresponding count value can be obtained by multiplying each timing value in the sequence data by 44100 / 8T. For example, the occurrence timing of event B can be represented by a count value of 480 × 44100 / 8T.
When the tempo is 120, this value is 480 × 44100 / (8 × 120) = 22050. If it becomes a decimal value, round it up, round it down, round up and so on, and round it to an integer value.

  The pre-reading data may be prepared to the end of the sequence data at the start of reproduction of the sequence data. When the tempo data change instruction is included in the sequence data, an increment of the count value from the previous event may be calculated using the post-change tempo value after that.

  However, if it is possible to change the tempo at any time according to the operation of the user, whenever there is a change in the tempo, it is necessary to recreate the pre-reading data based on the changed tempo after the part currently being reproduced. If this is taken into consideration, there is a possibility that it will be wasted even if the data for prefetching is created too far ahead. Therefore, it is desirable to sequentially read out the sequence data according to the progress of reproduction, and to create pre-reading data for a short time that is in time for reproduction.

Next, processing corresponding to the function shown in FIG. 3 and executed by the CPU 11 will be described using FIGS. 7 and 8.
FIG. 7 is a flowchart of the main process in which the CPU 11 starts execution at startup.
The CPU 11 starts the process shown in the flowchart of FIG. 7 when the electronic music apparatus 10 is activated, such as when the power is turned on or reset.
In this processing, after performing predetermined initialization processing (S11), CPU 11 repeats steps S12 to S22, detects user operation or data reception from an external device (S12), and also detects the detected operation or Perform processing according to the data.

  First, when the start instruction of the MIDI sequencer (instruction for reproducing sequence data) is detected (Yes in S13), the pre-read data described with reference to FIG. 6 is generated (S14) and the operation status of the MIDI sequencer is turned on. (S15), and the count value of the timer 182 of the sound source circuit 18 is reset to "0" (S16). The process of step S14 corresponds to the functions of the music data acquisition unit 111 and the pre-reading data generation unit 112. The processes of steps S15 and S16 correspond to the function of the sound source control unit 116.

  If an operation of a performance operator is detected (Yes in S17), performance data corresponding to the detected operation is generated (S18), and a tone generation instruction based on the performance data is sent to the tone generator circuit 18 together with the timer value "0". Supply (S19). When reception of performance data from an external device is detected (Yes in S20), a tone generation instruction based on the performance data is similarly supplied to the tone source circuit 18 together with the timer value "0" (S21). The tone generator circuit 18 sets necessary information in the register 181 based on the supplied tone generation instruction and timer value.

When performance data other than those included in the sequence data being reproduced is detected by these processes, a tone generation instruction corresponding to the performance data is supplied to the tone generator circuit along with a designation to immediately execute the tone generation instruction. can do. The processes of steps S18, S19 and S20 correspond to the functions of the musical performance operation detection unit 113, the musical performance data generation unit 114, and the musical performance data acquisition unit 115, respectively.
Step S22 is processing according to operations and received data other than those described above.

FIG. 8 is a flowchart of an interrupt process executed by the CPU 11 at predetermined time intervals.
In this process, if the operation status of the MIDI sequencer is not turned on in step S15 of FIG. 7 (No in S31), the CPU 11 ends the process without doing anything.

However, if it is turned on, the CPU 11 acquires the count value of the timer 182 of the sound source circuit 18 (S32). Then, if there is an event to be executed within the predetermined time after the count value acquired in step S32 in the pre-read data created in step S14 (Yes in S33), the sound generation instruction based on the corresponding event is the corresponding event And the corresponding timer value to the tone generator circuit 18 (S34). Then, the CPU 11 deletes the corresponding event from the pre-read data (S35), and ends the process. The deletion in step S35 is to prevent the tone generation instruction based on the same event from being issued again at the next processing timing. The tone generator circuit 18 sets necessary information in the register 181 based on the tone generation instruction and the timer value supplied in step S34.
In the process of FIG. 8, the event to be executed within a predetermined time after the acquired count value is supplied to the tone generator circuit, but instead, the event to be executed at the timing after the acquired count value is predetermined data An amount (for example, a predetermined number of events) may be supplied to the sound source circuit.

  In the above process, since the count value of the timer 182 is acquired in step S32, the CPU 11 can determine the execution timing using the elapsed time actually in progress in the sound source circuit 18. For this reason, even when the time measured by the timer 117 on the CPU 11 side is deviated due to the overload of the CPU 11 or the like, it is possible to correctly determine the execution timing of the event.

In addition, when there exist multiple events which satisfy | fill the conditions of step S33, the process after step S33 is performed for every event.
Further, the determination of step SX may be additionally performed in the process of FIG. If the required sound generation channel can not be secured, in order to perform the process of step S34, the sound generation being executed on any of the sound generation channels is truncated, and the sound generation of the event to be processed is reserved for the sound generation channel. Will do. However, if there is ample time until the execution timing of the event, it is considered that it is not necessary to make a reservation at this time until the sound generation is forcibly stopped. By this, it is possible to avoid useless stop of pronunciation.

However, the determination in step SX may not be performed. In this case, if the required sound generation channel can not be secured, truncation can be started "predetermined time" before actually starting the sound generation. In some cases, this can reduce the damage to the music due to truncation.
The user may optionally set whether or not to perform the process of step SX in consideration of these.
The process of FIG. 8 described above corresponds to the function of the sound source control unit 116.

Next, the effects of the embodiment described above with reference to FIG. 9 will be described.
FIG. 9 is a view corresponding to FIG. 10, showing the processing timing of the event in the sequence data in the electronic music apparatus 10 described above.
Also in the example shown in FIG. 9, it is assumed that the note-on event is to be performed in the sequence data 9.95 seconds after the start of reproduction. Further, it is assumed that the timing of the interrupt processing executed by the CPU is every 10 ms, and the predetermined time in step S33 of FIG. 8 is 20 ms.

  In the example of FIG. 9, the CPU 11 performs interrupt processing every 10 ms from the start of sequence data reproduction, and determines whether there is an event whose execution timing comes within 20 ms from the processing time (S33). It is judged that there is no corresponding event until the processing of 9.97 seconds. Thereafter, in the 9.98 second process, it is determined that the execution timing of the note-on event comes within 20 ms, and the tone generation start instruction corresponding to the note-on event is accompanied by a count value indicating 9.95 seconds which is the execution timing. The sound source circuit 18 is supplied (S34).

The tone generator circuit 18 sets the tone generation start instruction and the execution timing in the register 181, and the reading circuit 183 determines whether or not the set execution timing has come for every processing timing of 0.022675 ms cycle. Then, at the first processing timing after 9.995 seconds, it is determined that the execution timing has come, and sound generation corresponding to the sound generation start instruction is started.
Therefore, the error of the tone generation start timing is within 0.022675 ms from the time specified in the sequence data, and the tone generation with extremely high time accuracy as compared with the case of FIG. Can be started.
Note that events other than note-on and note-off, such as setting change, can be performed similarly.

  Further, in the example of FIG. 9, since the processing load of the CPU 11 is large at 9.99 seconds, a delay occurs in the timing of the interrupt processing, but at this time, the tone generation start instruction has already been supplied to the tone generator circuit 18 So there is no effect of this delay. If a certain long time is taken as the predetermined time in step S33 of FIG. 8, even if the execution timing of the interrupt processing is delayed in some interrupt cycles or processing can not be executed, the influence is not affected. Each event in the sequence data can be executed at the execution time specified in the sequence data.

  In addition, even when a large number of note-on events are concentrated at a specific timing such as the beginning of a measure, as in the case of automatic accompaniment style data playback, the CPU 11 has enough time to make a corresponding sound generation instruction. The sound source circuit 18 can be set. Therefore, processing delay due to concentration of events can also be avoided. Even if a delay occurs in the processing on the side of the CPU 11, it does not affect the execution timing of the tone generation instruction in the sound source circuit 18, unless it is a very large delay.

  The longer the predetermined time in step S33 is, the larger the effect of preventing the execution timing deviation of the sound generation instruction becomes. However, if it is made too long, many sound generation channels will be occupied by the sound generation start instruction in the execution reservation state. In addition, it is conceivable that the tempo is changed after the sound source circuit 18 has once set the sound generation start, and the setting content may be deviated from the timing when the sound should actually be sounded afterward.

On the other hand, if the predetermined time is too short, it may occur that the setting of the tone generator circuit 18 can not be performed by the timing when the sound generation instruction should be executed if delay or jump of the interrupt processing timing in the CPU 11 occurs frequently. However, even if such a situation occurs, it can be understood that there is only a timing deviation similar to that of the conventional process as shown in FIG.
Then, even if the predetermined time in step S33 is made the same as the execution interval of the interrupt processing, a certain effect can be obtained to prevent the timing shift. This predetermined time may be set in consideration of these.

  The tone generator circuit 18 is realized by hardware different from the CPU 11 which has a function of supplying a tone generation instruction and its execution timing to the tone generator circuit 18 based on sequence data. For this reason, since the sound source circuit 18 can be configured as hardware dedicated to generation and output of waveform data, processing can be performed in a constant cycle without causing a shift in processing timing. By this, it is possible to prevent the occurrence of deviation in the execution timing of the sound generation instruction.

  The sound source circuit 18 may not necessarily include wire logic dedicated to generation of waveform data. If processing relating to waveform data generation can be executed in a necessary cycle without being affected by other processing, for example, a configuration in which one core of a multi-core CPU is used only to realize the function of the sound source circuit 18 Is also employable. It is also conceivable to provide a dedicated CPU used only for realizing the function of the tone generator circuit 18.

This is the end of the description of the embodiment, but it goes without saying that the configurations of devices and functions, specific processing procedures, and the like are not limited to those described in the above-described embodiment.
For example, in the embodiment described above, the CPU 11 acquires the count value of the timer 182 of the sound source circuit 18, and compares the acquired count value with the execution timing of the event in step S33 of FIG. However, if the clocking error between the timer 117 of the CPU 11 and the timer 182 of the sound source circuit 18 is small enough (for example, 0.2 ms in 100 minutes), an event is generated using the time clocked by the timer 117 of the CPU 11. It may be compared with the execution timing of In this case, count values are converted as necessary.

  Further, in steps S19 and S21 of FIG. 7, the CPU 11 supplies the sound source circuit 18 with "0", which is the minimum value of the timer value, as a designation to immediately execute the sound generation instruction. In this way, the sound source circuit 18 can identify the sound generation instruction to be immediately executed simply by comparing the current count value of the timer 182 with the count value set in the register 181. Can be made easy. However, the present invention is not limited to this configuration, and a specific flag or the like may be used to indicate that the sound generation instruction should be immediately executed.

  Further, although the timer 182 is reset at the start of reproduction of sequence data in the above-described embodiment, this is not essential. If the count value of the timer 182 at the start of reproduction of the sequence data is stored and the difference from the current count value is taken, it is possible to obtain substantially the same count value as when reset.

  Further, in the tone generation circuit 18, a plurality of tone generation instructions may be set in the tone generation control register corresponding to one tone generation ch as in the case of the control event register. In this case, at the timing indicated by the timer value corresponding to each sound generation instruction, the process related to the corresponding sound generation instruction is executed. If sound generation according to the next sound generation instruction is to be performed while continuation of sound generation according to the previous sound generation instruction, the previous sound generation may be immediately stopped and the next sound generation may be performed.

  Further, in the example of FIG. 6, in the pre-read data, the timer value is recorded as an absolute value indicating the elapsed time from the beginning of the sequence data, but is recorded as a relative value indicating the elapsed time from the previous event You may do so. In this case, the timer 182 of the sound source circuit 18 may also be reset at each processing of an event (sound generation instruction) so as to obtain a relative elapsed time accordingly, or the count value at the time of processing of the previous event. The relative elapsed time may be determined by the difference from the current count value.

The sound source circuit 18 may not only generate an acoustic signal according to the sequence data, but may simultaneously reproduce audio waveform data (song data, accompaniment style data, short audio phrase data, etc.). In that case, if the generation cycle of the audio waveform data (same as the DAC cycle) and the processing cycle of the sound generation instruction based on the sequence data are matched, the audio signal generated according to the sequence data and the audio waveform data to be played back Can achieve completely synchronized reproduction.
Further, the description format of the sequence data is not limited to the MIDI format. Any type of data can be adopted as long as it is data that can express the sound generation instruction to be executed by the sound source device such as the sound source circuit 18 and the like and the information indicating the execution timing in any form.

  Further, the method of generating an acoustic signal in the sound source circuit 18 is not limited to the waveform memory method, and any method such as an FM (Frequency Modulation) method, a physical model method, an analog modeling method, etc. may be used. In the case other than the waveform memory system, it is not necessary to store the waveform address in the tone generation control register in the register 181, and instead, various parameters for generating a tone may be stored.

  In the case of the waveform memory system, the waveform data stored in the waveform memory 25 may be non-compressed data or data compressed according to a predetermined compression system. In addition, in the case of changing the pitch or performing time-stretching for reproduction, even if waveform data corresponding to the DAC cycle is prepared by interpolating, skipping or repeatedly reading a plurality of waveform samples, etc. Good. That is, the present invention is not limited to the configuration in which waveform sample values originally stored are read out every DAC cycle. In addition, regardless of the presence or absence of pitch change and time stretch, the readout circuit 183 reads out waveform data in a cycle different from the DAC cycle (for example, a cycle of 4 times the DAC cycle), and processing such as interpolation as necessary. After that, the waveform data may be supplied to the buffer of the DAC 23 every other cycle.

  Further, the present invention can be implemented as an acoustic signal generation apparatus having the functions of the CPU 11 and the sound source circuit 18 shown in FIG. 3 as in the electronic music apparatus 10, and a sound generation control apparatus having only the functions of the CPU 11. Alternatively, the present invention can be implemented as a sound source device having only the function of the sound source circuit 18. The program can also be implemented as a program for realizing the function of the CPU 11 and the function of the tone generator circuit 18 in a computer.

  Such a program may be stored in the ROM or other non-volatile storage medium (flash memory, EEPROM, etc.) provided in the computer from the beginning. However, it can also be provided by being recorded on any non-volatile recording medium such as a memory card, CD, DVD, Blu-ray disc and the like. Necessary processes can be executed by the computer by installing and executing the programs recorded in those recording media in the computer.

Furthermore, it is possible to download from an external device connected to a network and provided with a recording medium having a program recorded thereon or a program from the external device stored in the storage means, and install it on a computer for execution.
In addition, the configurations and modifications described above can be combined appropriately and applied as long as no contradiction arises.

As apparent from the above description, according to the present invention, generation of an acoustic signal according to sequence data can be performed with higher timing accuracy.
Therefore, by applying the present invention, it is possible to improve the quality of the acoustic signal generated according to the sequence data.

DESCRIPTION OF SYMBOLS 10 ... Electronic music apparatus, 11 ... CPU, 12 ... ROM, 13 ... RAM, 14 ... Storage device, 15 ... Communication I / F, 16 ... Detection circuit, 17 ... Display circuit, 18 ... Sound source circuit, 19 ... System bus, 21: Operator, 22: Display, 23: DAC, 24: Sound system, 25: Waveform memory, 111: Music data acquisition unit, 112: Data for prefetching generation unit, 113: Performance operation detection unit, 114: Performance data Generation unit 115 Performance data acquisition unit 116 Sound source control unit 117 Timer 181 register 182 Timer 183 read circuit 184 Tone control circuit

Claims (6)

A program for causing a computer to function as a sound generation control device for instructing a sound source device to generate sound,
The computer,
A sound generation instruction to be executed within a predetermined time from the timing according to sequence data prepared at a predetermined time interval timing or a predetermined amount of sound generation instruction to be executed after the timing is A program for functioning as instruction supply means for supplying the sound source device together with specification. The program according to claim 1, wherein
The program according to the present invention, wherein the specification of the execution timing for the sound source device is performed in units smaller than the predetermined time interval. The program according to claim 1 or 2,
When the instruction supply means detects performance data other than those included in the previously prepared sequence data, the instruction supply means may specify a tone generation instruction corresponding to the performance data, together with a designation to immediately execute the sound generation instruction. A program comprising means for supplying a sound source device. A sound source device,
Storage means for storing a pronunciation instruction and an execution timing of the pronunciation instruction;
And control means for starting an acoustic signal output according to the sound generation instruction stored in the storage means at an execution timing stored in the storage means in correspondence with the sound generation instruction.
The control means confirms whether or not the execution timing of the sound generation instruction stored in the storage means has arrived at a predetermined cycle based on the sampling cycle of the sound signal to be output .
The function of the control means is realized by hardware different from a processor that realizes a function of supplying a tone generation instruction and an execution timing to be stored in the storage means,
further,
Timer and
Means for providing information on time counted by the timer to the processor . A sound source device,
Storage means for storing a pronunciation instruction and an execution timing of the pronunciation instruction;
And control means for starting an acoustic signal output according to the sound generation instruction stored in the storage means at an execution timing stored in the storage means in correspondence with the sound generation instruction.
The function of the control means is realized by hardware different from a processor that realizes a function of supplying a sound generation instruction and an execution timing to be stored in the storage means,
further,
Timer and
Means for providing information on time counted by the timer to the processor. An acoustic signal generation device comprising: a sound generation control unit that instructs a sound source unit to sound the sound generation unit; and the sound source unit,
The sound generation control unit
The sound source unit includes instruction supply means for supplying a sound generation instruction to be executed within a predetermined time from the timing according to sequence data prepared in advance at a predetermined time interval timing and specifying the execution timing of the sound generation instruction. ,
The sound source unit is
Storage means for storing a sound generation instruction supplied from the sound generation control unit and an execution timing of the sound generation instruction;
Audio signal output according to the sound generation instruction stored in the storage means, and control means for starting the execution timing stored in the storage means in correspondence with the sound generation instruction Generator.

Images