JP2765469B2 - Music signal playback device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tone signal reproducing apparatus suitable for use in electronic musical instruments.

[0002]

2. Description of the Related Art Conventionally, there has been known an apparatus (hard disk recorder) for recording a tone signal on a hard disk and appropriately reproducing the signal. The hard disk recorder is used, for example, with an electronic musical instrument, and an ensemble is performed by a sound reproduced from a hard disk and a manual / automatic performance on the electronic musical instrument. At that time, in the electronic musical instrument, various special effects are given to the reproduced musical tones of the hard disk recorder. Examples of these special effects include volume control (fade-in, fade-out, muting), control of frequency characteristics (filter characteristics) of reproduced sound, amplitude modulation, frequency modulation, sound image localization position (pan), and the like.

[0003] When these special effects should be given,
Data such as the contents of the special effects are stored in advance in a storage device inside the electronic musical instrument, and the reproduced musical tone is controlled based on the data under the control of the CPU of the electronic musical instrument. For example, if the tone reproduction time by the hard disk recorder is "10 minutes" and muting is performed "10 msec" before the end of the tone reproduction, "9:59:99 from the start of reproduction"
It is preferable to set data indicating "muting after 0 msec" in the electronic musical instrument. There has also been an automatic performance device in which a waveform is loaded into a readable and writable waveform memory (waveform RAM) and an automatic performance is performed using the waveform.

[0004]

Some hard disk recorders can fine-tune the reproduction speed in order to fine-tune the pitch of the reproduced sound. However, if only the playback speed of the hard disk recorder is simply changed, there is a problem that the timing at which a special effect is applied to the playback sound is lost. For example, in the above example, "1
"9 minutes from the start of playback, 5 minutes to 5 minutes"
The content of the control is "muting after 9 seconds 990 msec". However, the reproduction time may be "9 minutes 59 seconds" due to the fine adjustment of the reproduction speed. In this case, the reproduced sound is cut off before muting is applied, and a huge click sound is output from the speaker. Further, in the above automatic performance apparatus, it is necessary to load all the waveforms read out during the reproduction of the performance data into the waveform RAM before starting the reproduction of the performance data. If an attempt is made to perform an automatic performance using a waveform stored in a large-capacity external storage means such as a hard disk, the capacity of the waveform to be reproduced is too large to fit in the waveform RAM, or all the waveforms are A large amount of waveform RAM has to be prepared for loading.

The present invention has been made in view of the above circumstances, and has as its object to provide a tone signal reproducing apparatus which satisfies the following requirements. (1) The effect can be given or the tone signal can be reproduced at an accurate timing even when the reproduction speed of the music is changed. (2) Automatic performance can be performed without loading all waveforms in advance into a large-capacity waveform RAM.

[0006]

The invention according to claim 1 is
Storage means for storing waveform sample data, and event
Data, which indicates when to generate
Type data indicating the type of target and parameters
Cue data storage means for storing cue data to be read, an address generation means for generating a read address for sequentially reading the sample data and supplying the cue data to the storage means, and the cue data read in accordance with the read address. a waveform generating means for generating a waveform signal based on said address
Based on the read address generated by the
-Included in the queue data stored in the data storage means
At the timing indicated by the timing data, the type data and
And queue generating means for generating a fine parameter, generated
The control target of the type indicated by the type data is
Characterized in that it comprises a control means for controlling Te. According to a second aspect of the present invention, in the tone signal reproducing apparatus according to the first aspect, the control means performs at least one control of tone generation control, modulation control, tone color control, localization control, and reverberation control. Features . The invention according to claim 3 is a performance
Performance data storage means for storing data;
Automatic performance means for performing an automatic performance based on the performance data stored in the storage means and generating a waveform generation event included in the performance data, a recording medium for storing waveform sample data, and reading and writing of the sample data. storage means that, the sample data of the head portion of the waveform that Ru is generated in the waveform generating events, stored in the performance data storage means
Means for reading from the recording medium prior to the waveform generation event based on the performance data and writing the data into the storage means; and sequentially transmitting the sample data from the recording medium in accordance with the progress of the generation of the waveform corresponding to the waveform generation event. Supply means for reading and supplying the storage means; and in response to the occurrence of the waveform generation event, first reading the sample data of the head portion from the storage means and then successively supplying the sample data continuously. is a reading reading means the sample data stored in said storage means, characterized in that it comprises a means for generating a waveform signal based on the sample data read in the reading means. The invention according to claim 4 is the invention according to claim 3.
In tone signal reproducing apparatus according, characterized in that it stores the data for specifying the waveform to be reproduced in the performance data in the header of the performance data. The invention according to claim 5, in the tone signal reproducing apparatus according to claim 3, said preparation unit may prepare to sample data tip head during automatic performance starting of the automatic performance means. According to a sixth aspect of the present invention, in the tone signal reproducing apparatus according to the third aspect, the performance data further includes a waveform preparation event, and when the automatic performance means reproduces the waveform preparation event, the preparation means is characterized in that to prepare the sample data tip head. According to a seventh aspect of the present invention, in the tone signal reproducing apparatus of the third aspect, the automatic performance means further generates a tone generating event, and the tone signal reproducing apparatus further generates a tone according to the tone generating event. It is characterized by including a musical tone generating means for generating. The invention according to claim 8 is a storage means for storing sample data of a waveform, and a tag indicating a timing at which an event should be generated.
Imaging data, type data indicating the type of control target,
Queue that stores queue data consisting of parameters and parameters
Data storage means, read rate generation means, accumulation means for accumulating the read rate generated by the read rate generation means, and sample data at a position corresponding to the accumulated value of the accumulation means, from the storage means Reading means for reading, waveform generating means for generating a waveform signal based on the sample data read by the reading means, and an arbitrary
Key for writing cue data to the cue data storage means.
And Yudeta setting means, based on the accumulated value of the accumulating means
The cue data stored in the cue data storage means.
Type data at the timing indicated by the included timing data
And queue generating means for generating the over data and parameters, outgoing
The type of control target indicated by the generated type data is
And control means for controlling according to data.
You . According to a ninth aspect of the present invention, in the musical tone signal reproducing apparatus according to the eighth aspect, the control means includes at least one of a musical tone generation control, a modulation control, a tone color control, a localization control, a reverberation control, and a generation of a tempo pulse . Characterized by performing control
You .

[0007]

[Action] claim 1 or 2 According to the invention according, sample data is stored in the storage means, Dasa sequentially read in accordance with the read address generated by the address generator, the waveform generating means is read out Wave based on sample data
A shape signal is generated . The queue generating means includes a read address.
Tie indicating when an event should occur based on
Data, type data indicating the type of control target, and
Queue data that stores queue data consisting of
Data included in the queue data stored in the data storage means.
At the timing indicated by the
Generating the parameter and the species generated by the control means.
The control target of the type indicated by the class data is
Is controlled . Therefore, the control timing of the control object corresponds with the read position of the sample data, as the reproduction speed of the sample data is changed, the wave to be reproduced
Control can be performed at timing completely synchronized with the shape signal. According to the invention of any one of claims 3 to 7, prior to the performance based on the performance data,
Of the waveforms stored in the large-capacity storage medium, the sample data at the beginning of the waveform specified to be generated in the performance data is converted to the performance data stored in the performance data storage means.
Based on the information, it is transferred to the storage means in advance. When the reproduction of the performance data is started, the waveform generation events included in the performance data are sequentially reproduced. The reading means starts reading the head sample data of the waveform designated by the waveform generation event from the storage means in response to the waveform generation event. The supply means supplies the sample data to be read following the sample data to the storage means from the storage medium while the reading means is reading the sample data from the storage means. Therefore, each waveform can be automatically played and reproduced simply by preparing in advance only the sample data at the head of the waveform to be reproduced in the storage medium in the storage means. According to the invention according to claim 8 or 9,
The sample data stored in the storage means is sequentially read according to the accumulated value of the rate generated by the accumulation means, and the waveform generation means forms a waveform signal based on the read sample data. I do. The queue data storage means stores events
Data, which indicates when to generate
Type data indicating the type of target and parameters
Queue data is stored by the queue data setting means.
Queue data is queued according to the operation of the operator.
Written to the data storage means. Queue generating means, cumulative
Stored in the queue data storage means based on the accumulated value of the calculation means
Indicated by the timing data included in the queued data
Type data and parameters are generated at the timing ,
The control means is the type of control object indicated by the generated type data.
Is controlled according to the generated parameters. Therefore, control
The execution timing of the control by the control means corresponds to the reading position of the sample data. Even if the reproduction speed of the sample data is changed, the control is performed at a timing completely synchronized with the reproduced waveform signal. Can be done.

[0008]

Embodiment A. Configuration of an embodiment An electronic musical instrument according to an embodiment of the present invention will be described below with reference to the drawings. In FIG. 1, reference numeral 2 denotes a keyboard, on which a plurality of keys to be played by a player are provided, and operation information for these keys is output via a bus 10. Also, 4
Is a CPU configured to control other components based on a processing program stored in the ROM 6.

Reference numeral 8 denotes a RAM which can be read and written by the CPU 4. 12 is a display,
Various data are displayed to the player. Reference numeral 14 denotes a panel switch provided with various control operators and the like. Among them, an automatic performance start / stop switch 14 for instructing start / stop of automatic performance.
a, tempo setting switch 14 for setting tempo
b, a waveform file sounding preparation instruction switch 14c, and a waveform file sounding start instruction switch 14d. The functions of these switches will be described later.

Reference numeral 18 denotes a hard disk, which is controlled by the CPU 4 via the SCSI input / output control circuit 16 and the bus 10. Reference numeral 20 denotes a DMA controller, and CP
The read control for the hard disk 18 is performed independently of U4. Reference numeral 26 denotes a sound source, which generates and outputs a tone signal based on the supplied waveform data.
The output tone signal is generated through the sound systems 28 and 30.

Reference numeral 24 denotes a time slot control circuit for writing waveform data to the waveform RAM 22 by the DMA controller 20, reading the waveform for reproduction by the sound source 26, and performing processing such as mixing and processing of waveforms by the CPU 4. The total number of accesses per sampling time of the waveform RAM 22 is allocated for each access so that reading and writing can be performed simultaneously. For example, the waveform RAM 22 stores 100 per sampling time.
Assuming that access is possible twice, 32 times are used for accessing the DMA controller 20, 64 times are used for accessing the sound source 26, and the remaining 4 times are used for accessing the CPU 4.

In the time slot control circuit 24, of the accesses from the DMA controller 20, the sound source 26, and the CPU 4, in the write access, the write address and the read address are pushed into the allocated access timings to perform the write. In the read access, the read address is pushed at the assigned timing and read is performed, and the read data is extended to the original access time length. By the operation of the time slot control circuit 24, the DMA
Waveform RA from controller 20, sound source 26, and CPU 4
Access to M22 can be performed in parallel without collision, as if it were a multiport RAM.

Next, the detailed configuration of the sound source 26 will be described with reference to FIG. In the figure, 42 and 46 are registers,
Various data are written by the CPU 4 via the bus 10. That is, the register 42 contains the F number FNO
, A start / stop signal SS, an attack start address AS and a loop start address LS are written, while the register 46 has an interrupt address X for long-time waveform reproduction, an interrupt timing address Y, an attack end address AE and a loop. End address LE is written. Reference numeral 34 denotes an address counter,
A count operation is performed based on the data written in the second data.

Here, the contents of various data written in the registers 42 and 46 will be described. First, the start / stop signal SS is a signal for turning on / off the count operation of the address counter 34. If the value of this signal is "1", the count operation of the address counter 34 is executed and "0" If so, the counting operation is stopped.

Next, the attack start address AS is:
In the waveform RAM 22, an area in which the waveform data of the attack portion of the musical tone waveform (the portion at the beginning of the generation of the musical tone) is stored, that is, the head address of the attack area, and the attack end address AE indicates the last address of the attack area. Things. The waveform data stored in the attack area is read only once at the beginning of generation of a musical tone.

The loop start address LS is
This is the area where waveforms other than the attack part are stored, that is, the start address of the loop area, and the loop end address L
E is the last address of the loop area. This loop area is repeatedly read while a musical tone is generated. However,
The waveform data stored in this area is not constant. That is, as will be described in detail later, new waveform data is sequentially read from the hard disk 18 as the music progresses,
The contents of this area are updated sequentially.

Here, the reason why the storage area is divided into an attack area and a loop area will be described. First, in the present embodiment, all the waveform data used for the music are stored in the hard disk 18, and the necessary waveform data is read out from the hard disk 18 as appropriate, and is read via the DMA controller 20 and the time slot control circuit 24. The waveform R
Stored in AM22. However, if the waveform at the beginning of the musical tone is read from the hard disk 18, a time lag necessarily occurs.

Therefore, for a plurality of musical tone waveforms used in the music, only these attack portions are previously subjected to the waveform R.
It is to be stored in each attack area in the AM 22. That is, at the beginning of the tone generation, the waveform R
The waveform data of the attack part is read out from the attack area of the AM 22 and is immediately sounded. Then, the hard disk 18 is accessed while the waveform data of the attack part is sounded, and data to be sounded next is written in the loop area. In addition, since the attack portion is prepared in the waveform RAM 22 for a predetermined number of musical sound waveforms used for performance out of all the musical sound waveforms in the hard disk 18, which of the predetermined number of musical sound waveforms is instructed to sound. Even though it can be pronounced immediately.

The loop area is divided into a first half and a second half, and the waveform data is updated for each part. That is, when the first half is read and sounded by the sound source 26, the second half is updated with data to be sounded next to the first half, and conversely, when the second half is read, The data in the first half is updated with data to be sounded next to the second half. Thus, new waveform data is stored in the waveform RAM 2 from the hard disk 18.
2, the sound source 26 may simply repeatedly read the loop area.

In order to enable such control, it is detected whether the address read by the sound source 26 is the first half or the second half, and the DMA controller 20 is operated in accordance with the result.
Needs to be changed. Therefore, the interrupt command XS is output from the interrupt register 44 (details will be described later), and the interrupt address X is set as an address at which the interrupt command XS is generated. That is, the interrupt address X is the head address of the latter half of the loop area.

Next, the interrupt timing address Y is
The timing at which various tone controls (for example, volume control, filter control, amplitude modulation, and frequency modulation) are performed is represented as address information. That is, the interrupt command YS is generated at the moment when the address counted by the address counter 34 becomes equal to or more than the interrupt timing address Y. When this interrupt occurs, various special effects are given to the tone signal under the control of the CPU 4.

Next, the F number FNO is a value set in accordance with the pitch of a musical tone to be generated, and the F number FNO increases as the pitch increases. This F number FNO is
The address is supplied to the address counter 34 as an increment value. Therefore, the larger the F number FNO, the faster the address change speed and the faster the speed of reading the waveform stored in the waveform RAM 22, so that the pitch of the generated musical tone also increases.

The address counter 34 sets the attack start address AS as an initial value and sets the F number as described above.
A count operation is performed at predetermined clock cycles using FNO as an increment value. Then, the count value is output as the address signal Sad1. Reference numeral 36 denotes a comparator, which includes an address signal Sad1 and an attack end address AE.
When the former is larger than the latter, the control signal C1 of "1" is output.

When the control signal C1 of "1" is supplied, the address counter 34 sets the address signal Sad1 to the loop start address LS and restarts the counting operation. The address counter 34 is designed to respond to the first rise of the control signal C1 after the rise of the signal SS, and is not affected by changes in the control signal C1 thereafter. The comparator 36 compares the address signal Sad1 with the loop end address LE. If the former is larger than the latter, the comparator 36 outputs a control signal C2 of "1".

When the control signal C1 becomes "1" and the address becomes LS
When the control signal C2 of "1" is supplied after the setting, the address counter 34 returns the address signal Sad1 to the loop start address LS. Since the address signal Sad1 becomes smaller than the loop end address LE, the control signal C2 becomes "0" and the counting operation is continued. Such an address counter 34
By the operation described above, the address signal Sad1 changes as shown by the solid line in FIG. In the figure, the slope of the slope of the address signal Sad1 is determined according to the F number FNO.

The comparator 36 outputs the address signal Sad1
At the moment when the value exceeds the attack end address AE, an attack end interrupt command AES is output. That is, in the previous clock timing, “Sad1 <AE
”, And“ S
ad1 ≥ AE, the attack end interrupt command AE
S is supplied to the CPU 4 via the interrupt register 44.

Similarly, at the moment when the address signal Sad1 exceeds the interrupt address X, the interrupt command XS changes, and when the address signal Sad1 exceeds the loop end address LE, the loop end interrupt command LES changes at the interrupt timing address Y or more. At the moment, the interrupt command YS is issued via the interrupt register 44
It is supplied to PU4. When receiving these interrupt commands, the CPU 4 performs corresponding processes (details will be described later).

Therefore, the address signal Sa shown in FIG.
Based on the change in d1, an interrupt as shown in FIG. That is, the address signal Sa at time t ₁
Since d1 is equal to or higher than the attack end address AE, the attack end interrupt command AES is output at the same time.
Then, since the address signal Sad1 becomes more interrupt address X in time t _2, the same time interrupt instruction XS
Is output. Further, at time t ₃ for the address signal Sad1 becomes more loop end address LE, loop end interrupt command LES is output.

The interrupt timing address Y does not always need to be provided, and may be provided as appropriate when tone control is required. In the illustrated example, the interrupt timing address Y immediately after time t ₃ is assumed to have been set in the register 46. As a result, the address signal
Time when Sad1 matches interrupt timing address Y,
That interrupt command YS is generated at time t _31.

In this embodiment, in order to control the pitch with high precision, the F number FNO has a decimal part as well as an integer part. As a result, the address signal output from the address counter 34 is output. Sad1 also has a decimal part.
The integer part of the address signal Sad1 is supplied to the time slot control circuit 24 to access the waveform RAM 22, while the decimal part of the address signal Sad1 is
2 is supplied. In the interpolation circuit 32, interpolation of the waveform data is performed based on the latest waveform data supplied through the time slot control circuit 24, the past waveform data, and the decimal part of the address signal Sad1. Waveform data is output.

Reference numeral 38 denotes a timbre control filter whose pass characteristics, that is, a cutoff frequency and a Q factor, can be appropriately set. 48 is FC
When a tone color control command is supplied from the CPU 4, the register is latched and supplied to the tone color control circuit 50. The timbre control circuit 50 performs a calculation based on the supplied command, and controls a time-varying parameter Fti for controlling the filter.
1 is generated to control the pass characteristic of the tone color control filter 38.

Reference numeral 40 denotes a sound volume control mixer for controlling the sound volume balance, envelope and pan for the sound systems 28 and 30. 52 is VCD
When a volume control command is supplied from the CPU 4, the register is latched and supplied to the volume control circuit 54. The volume control circuit 54 performs a calculation based on the supplied command, generates a time-varying parameter Fvol for controlling the volume, and controls the volume control mixer 40.

Here, the tone volume control performed in the volume control circuit 54 includes, for example, fade-in performed at the beginning of a music piece, fade-out performed at the end of a music piece, and muting. When generating a musical tone based on the performance on the keyboard 2, the CPU 4
, Envelope information is supplied to the VCD register 52, and the volume control mixer 40 functions as an envelope control circuit. Then, the tone signal output from the volume control mixer 40 is converted into an analog signal via the DA conversion circuit 56 and is emitted via the sound systems 28 and 30.

The tone generator 26 does not generate only the tone signal of the "1" channel, but generates "8"
It is configured to generate a tone signal of a channel. That is, the above-described registers 42 and 44, the address counter 34, and the like all perform the time-division operation of the “8” channel, and the comparator 36 also outputs an interruption signal for the “8” channel by the time-division processing, The control mixer 40 is configured to mix and output signals for the "8" channels by weighting them with the time-varying parameter Fvol.

Here, the first to fourth channels are assigned to tone signals reproduced using the hard disk 18, and the fifth to seventh channels are assigned for playing on the keyboard 2. The eighth channel is allocated for generating a rhythm (tempo pulse). Hereinafter, in this specification, especially when a channel number is described, the channel number is displayed with parentheses after the signal name. For example, the start / stop signal SS supplied to the channel i is expressed as “start / stop signal SS (i)”.

Next, a memory map of the waveform RAM 22 will be described with reference to FIG. In the figure, AW1 to AWn
Is an attack area for storing an attack portion of each musical tone used for the music. That is, at the beginning of the generation of a musical tone, any one of the attack areas AW1 to AWn is selected, and a musical tone is generated based on the selected content. Also, buf
(1,1) to buf (4,2) are loop areas, and the area buf (i,
A set consisting of 1) and buf (i, 2) (where i is any one of "1" to "4") is assigned as a loop area of one sounding channel.

In other words, in FIG. 4B, “LS” to
The area corresponding to the address signal Sad1 of "X-1" is the area b
uf (i, 1), and the area corresponding to "X" to "LE" is buf
(i, 2). That is, the loop areas buf (i, 1) and buf (i, 2) are alternately accessed by the address signal Sad1. Therefore, in the present embodiment, the area not accessed by the address signal Sad1 is accessed by the DMA controller 20, whereby the waveform data in the hard disk 18 is sequentially transferred to the waveform RAM 22.

Hereinafter, an address signal supplied from the DMA controller 20 to the waveform RAM 22 via the time slot control circuit 24 is referred to as an address signal Sad2.
This is indicated by a broken line in FIG. In the figure, times t _{0 to} t ₁
, The attack area of the sound channel (that is, any of the attack areas AW1 to AWn)
And the waveform of the attack portion is reproduced.
During this period, the data to be reproduced is read from the hard disk 18 by the DMA controller 20 at the next times t _{1 to} t ₂ and t _{2 to} t ₃ , and the read data is stored in the loop area buf (i, 1). ) And buf (i, 2).

Next, the loop region buf (i, 1) is in the period of time t ₁ ~t _2, the loop region buf in the period of time t ₂ ~t3 (i, ₂₎ is read by the respective sound source 26. Here, in the latter period, the next time t
Data to be reproduced ₃ ~t ₄ is written in the loop region buf (i, 1). Similarly, reading and writing of the loop areas buf (i, 1) and buf (i, 2) are performed alternately, and the waveform data recorded on the hard disk 18
6 is continuously reproduced.

The data to be read from the hard disk 18 is determined by the variable CNT (i). This variable CNT (i) indicates the block number of the waveform file in the hard disk 18 (details will be described later). The initial value of the variable CNT (i) is “0” and is incremented each time the attack end interrupt command AES, the interrupt command XS, or the loop end interrupt command LES is generated (see FIG. 4A). ). In order to associate a waveform number and a block number with each channel, the RAM 8 stores recorder channel management data as shown in FIG.

B. Operation of Embodiment Overall Operation of Embodiment Next, the operation of this embodiment will be described. First, when the power of the electronic musical instrument of this embodiment is turned on, a main routine shown in FIG. 5A is started. When the process is started in the figure, first, in step SP1, predetermined initial settings are executed. Next, when the process proceeds to step SP2, the operation state of the keyboard 2 is detected, and a predetermined key process is performed based on this.

Next, when the processing proceeds to step SP3, the operation state of the panel switch 14 is detected, and processing such as state setting based on the operation state is performed. Next, the process proceeds to step SP4, where various other processes are performed. Thereafter, the processing of steps SP2 to SP4 is repeated. Here, various processes performed in step SP3 will be described in detail.

(I) Processing for event of automatic performance start / stop switch 14a When an event of automatic performance start / stop switch 14a is detected, a subroutine shown in FIG. 7A is called. When the process is started in the figure, step SP4
At 01, the value of the flag RUN is inverted. Here, the flag RUN is a flag indicating that the automatic performance is being performed if "1", and that the automatic performance is not being performed if "0". Note that the automatic performance here is an automatic performance of a rhythm sound performed on the eighth channel.
Next, when the process proceeds to step SP402, it is determined whether or not the inverted flag RUN is “1”.

Here, if the inverted flag RUN is "0", the process proceeds to step SP403, and a start / stop signal SS (8) of "0" is output to the eighth channel. Accordingly, the counting operation of the address counter 34 (FIG. 2) for the eighth channel is stopped, and the generation of the rhythm (tempo pulse) is stopped. Next, when the processing proceeds to step SP404, the CPU 4 supplies a note-off signal to the sound source 26 for the fifth to seventh channels, that is, all the channels assigned for playing the keyboard 2. As a result, the performance on the keyboard is forcibly terminated, and the process returns to the main routine.

On the other hand, if the flag RUN after the inversion is "1", the process proceeds to step SP405, and the player specifies sequencer data to be reproduced. Here, the sequencer data refers to the settings of a single song (song) such as tempo and tone settings at the start of the song, as well as the song duration, capacity, song name, input date, etc. A header that stores various data to be performed, various events that are the main body of performance data, and a time (duration) until these events are defined. Note that the RAM 8 stores sequencer data corresponding to a plurality of songs (songs) SG1 to SGk as shown in FIG. Hard disk 1
8 stores a plurality of sequencer data with one song as one file.
It is loaded on the RAM 8 as shown in FIG. Further, in each song, data designating a musical tone to be reproduced in the song is stored, and the waveform data of the attack portion of the musical tone used in accordance with the data when trying to play the song is stored in the waveform RAM 22. You may make it prepare automatically in an attack area.

Next, when the process proceeds to step SP406, the predetermined time initial value (for example, “0”) is set in the variable APCNT.
Is assigned to The variable APCNT is incremented at predetermined intervals in a tempo interrupt process (FIG. 7B) described later, and corresponds to the elapsed time after the start of the automatic performance. Next, when the process proceeds to step SP407, the information of the header portion is read from the previously specified sequencer data, and the timbre, the tempo, and the first duration (the time until the first event is present) at the start are determined. Is set.

When the process proceeds to step SP408, the start / stop signal SS (8) of the eighth channel is set to "1", whereby the generation of the tempo pulse using the eighth channel is started and the rhythm performance is started. Is started. In this way, every time the event of the automatic performance start / stop switch 14a occurs, a process such as inverting the flag RUN is performed, and a process of starting or stopping the rhythm performance is performed.

(Ii) Processing for event of tempo setting switch 14b When an event of tempo setting switch 14b is detected,
The subroutine shown in FIG. 7C is called. When the process is started in the figure, in step SP601, the set value of the tempo, that is, the number of quarter notes per "1 minute" is substituted for the variable TMP. Next, the process proceeds to step SP
In step 602, the F number FNO of the eighth channel is calculated based on the following equation (1).

FNO = (TMP × 1024 × 96) / (60 × 48 × 10 ³ ) Equation (1)

In the equation (1), the reason why the variable TMP is divided by “60” is to convert the number of quarter notes per “1 minute” indicated by the variable TMP into “1 second”. . “48 × 10 ³ ” is a sampling frequency. Thus, these divisions yield the number of quarter notes per "1" sampling clock. Next, "10
"24" means "1 kilobyte" and is the memory capacity of the waveform RAM 22 from the loop start address LS to the loop end address LE. “96” indicates the resolution. The numerical values such as “1 kilobyte” and “96” are not fixed, and may be changed as necessary.

Next, when the process proceeds to step SP603, the calculated F number FNO is supplied to the register 42 of the sound source 26 as the F number corresponding to the eighth channel. Thus, when there is an event of the tempo setting switch 14b, the F-number FNO of the eighth channel is updated based on the set tempo.

(Iii) Processing for Event of Waveform File Sounding Preparation Instruction Switch 14c When an event of the waveform file sounding preparation instruction switch 14c is detected, a subroutine shown in FIG. 9 is called. Note that, in order to actually generate the tone data stored in the waveform file, an event of a waveform file tone generation start instruction switch 14d, which will be described later, is necessary. Here, preparations for the tone generation are performed.

When the process is started in the figure, in step SP901, a waveform file to be reproduced is specified from among the waveform files stored in the hard disk 18. This waveform file is composed of waveform data composed of a plurality of blocks. Next, when the process proceeds to step SP902, an empty area among the attack areas AW1 to AWn in the waveform RAM 22 is secured. If there is no open area, only one waveform that can be erased is designated from AW1 to AWn by operating the panel SW or by automatic detection, and the attack area is reserved for writing.

Next, when the processing proceeds to step SP903, the 0th block of the specified waveform file is read from the hard disk 18 and written in the previously secured attack area. Needless to say,
The 0th block of each waveform file stores the waveform data of the attack section. Therefore, the waveform data of the attack portion is written into the waveform RAM 22, and thereafter, if there is an instruction to start generating a waveform file, the waveform can be immediately reproduced.

The hard disk 18 stores data called "cue data" corresponding to the waveform file. The cue data indicates special effects such as muting during reproduction of the waveform data. In step SP903 described above, the cue data is read together with the 0th block of the waveform file, and the cue data is stored in the RAM 8 via the SCSI input / output control circuit 16.

Next, the contents of the queue data will be described with reference to FIG. In the figure, a queue
Region Q ₁ to Q _n for storing data is provided, these regions, the attack area shown in FIG. (B) AW1
~ AWn corresponds one-to-one. Each region Q ₁ to Q _n, each plurality of queue data stored.
In the illustrated example, “m” pieces of queue data Q ₃ (1) to Q ₃ (m) are stored in the area Q ₃ .

Further, each queue data is composed of a block number QB indicating a block in which the queue occurs, a queue start time QT in the block, and a queue content QE indicating specific contents of the queue. I have. Although details will be described later, the variable CNT (i) shown in FIG. 4A matches the block number being reproduced. Thus, in FIG. (D), the queue at time t _31, i.e. if the cause of the interrupt command YS, block number QB must be set to "3". Furthermore, in this case, start time QT, it is necessary to set the time ΔT from time t ₃ ~ time t _31.

The queue contents QE are composed of a queue type and parameters according to the queue type. Here, the type of queue is specified by several alphabetic characters, and the contents of the parameters are shown in parentheses attached to the alphabetic characters, as follows. A plurality of parameters to be specified independently are separated by a symbol “,” and parameters that are selected alternatively are separated by a symbol “/”.

VOL (level, rate) This queue is used to change the volume level. The volume level after the change and the rate of change (rate)
Are parameters. The “rate” here is the speed at which the current volume level changes to the specified volume level. The same applies to the other queues listed below. MUTE (on / off) This cue changes the on / off state of muting, and the changed state (on or off) is a parameter.

FC (frequency, rate) This queue changes the cutoff frequency of the timbre control filter 38, and the frequency after change and the rate of change (rate) are parameters. FQ (Q factor, rate) This queue changes the Q factor (amplitude increase ratio) of the tone color control filter 38, and the changed Q factor and the rate of change (rate) are parameters.

PAN (position, rate) This cue changes the sound image position, and the changed sound image position and the change rate (rate) are parameters.・ PIT (Pitch change amount, rate) This cue changes the waveform playback pitch.
The changed pitch change amount and change rate (rate) become parameters. The pitch can be changed by appropriately changing the F number FNO stored in the register 42.

NON (Key Code, Velocity) This cue is to automatically generate a note-on signal to be generated by playing the keyboard 2 manually. Key speed) is a parameter. Therefore, note-on sound generated in this cue is assigned to the fifth to seventh channels of the sound source 26. NOF (Key Code) This cue is to automatically generate a note-off signal for a musical tone generated by the above-described cue content "NON", and a key code to be turned off is a parameter.

As an example, “QE = FQ (+1.1,
The queue content “3)” corresponds to the Q of the timbre control filter 38.
This means that the factor is set to “+1.1” and the rate of change (rate) is set to “3”.

(Iv) Waveform file sound generation start instruction switch 1
Processing for Event 4d When an event of the waveform file sounding start instruction switch 14d is detected in step SP3 of the main routine, the processing proceeds to a subroutine shown in FIG. 8A. When the process is started in the figure, in step SP701, the performer specifies a waveform file to be reproduced. The waveform file specified here is a waveform file selected from the n waveform files that have been subjected to the sound preparation process (FIG. 9) first, and the attack waveform A
The description proceeds assuming that the waveform file corresponding to Wx and the cue data Qx has been specified.

Next, when the process proceeds to step SP702, among the first to fourth channels of the sound source 26, if there is no vacant channel or no vacant channel, one channel being reproduced is muted and secured. Is assigned. Then, the assigned channel number is stored in the variable i. Next, the processing proceeds to step SP703.
Then, the start address of the storage area AWx on the waveform RAM 22 of the 0th block of the waveform file instructed to be reproduced is assigned to the attack start address AS (i), and the last address is assigned to the attack end address AE (i). You.

Next, when the process proceeds to step SP704, the loop start address LS (i) and the loop end address LE (i) are determined based on the channel previously allocated in step SP702. Next, when the process proceeds to step SP705, an intermediate value between the loop end address LE (i) and the loop start address LS (i) is set in the interrupt address X (i).

Next, when the process proceeds to step SP706, the CPU 4 sets the first and second blocks of the waveform file as transfer data, and sets the loop areas buf (i, 1) and buf (i, 2) as transfer destination addresses. To DMA controller 2
DMA transfer is instructed to 0. As a result, in the DMA controller 20, the first
Then, the read operation of the second block is started. Next, when the processing of steps SP707 and 708 is executed, “0” is substituted for the variable CNT (i) and the start / stop signal SS (i) of “1” is supplied to the sound source 26.

The start / stop signal SS (i) is
, The count operation is started in the address counter 34, the address signal Sad1 gradually increases, the waveform RAM 22 is read, and the waveform data in the attack section is reproduced. The address signal Sad1 at this time
Is controlled by the F number set for the i channel of the sound source 26. When trying to reproduce at the original waveform pitch when the waveform file was recorded, the value of the F number is a value of the ratio of the sampling frequency at the time of recording to the sampling frequency (48 KHz) of the sound source 26.
If it is desired to increase or decrease the reproduction pitch of the waveform file, the F number may be increased or decreased using the value of this ratio as a standard value. By the way, in step SP706, the DMA transfer is instructed to the DMA controller 20, but a time lag is inevitably caused to read the hard disk 18. Accordingly, as shown in FIG. 4 (b), when the data transfer is performed by the address signal Sad2 will later slightly than the time t _0.

[0069] In the sound source 26, the sound of the attack portion is completed at time t _1, is outputted control signal C1 is "1" from the comparator 36, the address signal Sad1 is set in the loop start address LS. Thereafter, the address signal Sa
d1 address signal Sad1 until it equals the loop end address LE is incremented, up to the time t _3, the waveform data stored in the loop region buf (i, 1) and buf (i, 2) are sequentially read.

As described above, according to the operation of the main routine and the subroutine called by the main routine, the waveform data of the attack area and each loop area buf (i, 1), b
The first waveform data of uf (i, 2) is written into the waveform RAM 22, and a sound generation process is performed based on these data. Then, for the time t ₁ subsequent processing, performed by various interrupt routines to be described below.

AES Interrupt Processing At time t ₁ in FIG. 4B, when the address signal Sad ₁ becomes equal to the attack end address AE, the attack register 44 outputs an attack end interrupt command AES. When this interrupt occurs, an interrupt routine shown in FIG. 6B is called. When the process is started in the figure, first, in step SP301, the interrupted channel number is substituted for a variable i. The variable i used in each interrupt processing routine is different from that used in the main routine.

Next, in step SP302, it is determined whether or not the variable i is in the range of "1" to "4", that is, the interrupted channel is the first channel related to the sound generation of the waveform file.
It is determined whether the channel is any of the fourth to fourth channels.
Here, if “YES” is determined, the process proceeds to step S
Proceeding to P303, the variable CNT (i) relating to the channel is set to "1", and the process returns to the main routine. Thus, in the example shown in FIG. 4 (a), at time t ₁ variable CNT (i) is set to "1".

XS / L in the first to fourth channels
When an interrupt instruction XS occurs in ES interrupt processing time t ₂ of the (i) XS interrupt processing Figure 4 at time _{t 2 (b), XS /} LES interrupt routine shown in FIG. 5 (b) is called. When the process is started in the figure, step SP
At 101, the interrupted channel number is set to the variable i
Is assigned to Next, when the processing proceeds to step SP102, the processing branches according to the value of the variable i. This time, the process proceeds to step SP103 on the assumption that an XS interrupt has occurred in the first to fourth channels.

In step SP103, FIG.
Is called. When the process is started in the figure, in step SP801, it is determined whether or not the variable CNT (i) is the number of the last block. In the example of FIG. 4, it is assumed that the last block number of the waveform file is “6”. Further, the variable CNT at this point (i) is the value at the time the interrupt command XS occurs in time t _2, the ie is "1".

Therefore, the determination is "NO" here, and the process proceeds to step SP803. In step SP803, the variable CNT (i) is incremented by "1". Therefore, here, the variable CNT (i) is set to “2”. Next, when the process proceeds to step SP804, the variable C
It is determined whether NT (i) is the number of the last block. Here also, the determination is “NO”, and the process proceeds to step SP.
Proceed to 805.

In step SP805, the variable CNT
When (i) is an odd number, “2” is substituted for the variable j,
If the variable CNT (i) is an even number, “1” is substituted for the variable j. At this point, since the variable CNT (i) is “2”, “1” is substituted for the variable j. Next, when the process proceeds to step SP806, a DMA transfer is instructed to the DMA controller 20.

Here, the data to be transferred is the {CNT (i + 1)} block in the waveform file being reproduced on the i-th channel. That is, it is the third block at this point. The transfer destination address is the loop area buf (i, j)
It is. That is, at this point, the loop area buf (i, 1)
It is. Accordingly, as shown in FIG. 4 (b), the loop region buf by the address signal Sad2 after a while after time t ₂ (i, 1) is accessed, its contents are updated in the waveform data of the third block .

Next, when the processing proceeds to step SP807, it is determined whether or not a queue exists in the CNT (i) block. In the example of FIG. 4, the queue because there only the time t ₃₁ in the third block, not in the second block exists. Therefore, here, the determination is “NO”, and the process proceeds to step SP809. Step SP8
At 09, the YS interrupt is released. In particular,
A value which the address signal Sad1 can never take is preferably set in the interrupt timing address Y. When the above processing is completed, the XS / LES interrupt routine (FIG. 5B)
The processing returns to the main routine via.

(Ii) LES interrupt processing at time t ₃ Next, when the loop end interrupt command LES is generated at time t ₃ in FIG. 4B, the same processing as in the case of the above-described interrupt command XS is performed. Done. That is, the XS / LES interrupt routine (FIG. 5B) is called, and step SP103 is executed.
, A recorder processing subroutine (FIG. 8B) is called. In this subroutine, step SP
The variable CNT (i) is set to “3” via 803, and “2” is substituted for the variable j in step SP805.

Next, when the process proceeds to step SP806, the fourth block of the waveform file is set in the loop area buf (i,
A transfer command is output to the DMA controller 20 to transfer the data to 2). Therefore, as shown in FIG.
After a while, the address signal after a lapse of time t ₃ Sad2
Accesses the loop area buf (i, 2) and updates the contents to the waveform data of the fourth block.

Next, when the processing proceeds to step SP807, the queue data (see FIG. 3C) stored in the RAM 8 is searched, and it is determined whether or not a queue exists in the CNT (i) block, that is, the third block. Is determined. In the example shown in FIG. 4, it is assumed that the queue should be executed at a time t ₃₁ in the same block exists. Therefore, here, it is determined to be “YES”, and the process proceeds to step SP80.
Proceed to 8.

[0082] In the step SP808, the waiting time until the first queue in the block is generated, i.e., in response to "Qx (1) = ΔT = t 31 -t 3 ", split register 46 Timing address Y is set. Then, when the above processing ends, the processing returns to the main routine.

Hereinafter, similarly, at time t ₄ , an interrupt command is issued.
XS is, the loop end interrupt instruction LES at time t ₅
Respectively occur, and the corresponding processing is performed. Although an interrupt command YS is generated at time t ₃₁ from the step SP808 is executed in the interrupt time t _3,
Details of the process in that case will be described later.

[0084] (iii) XS interrupt processing at time t ₆ Next, step SP803 after the interrupt instruction XS occurs at time t ₆ is executed, the variable CNT (i) is set to "6" . Since this number is the number of the last block in the example of FIG. 4, “YES” is determined in step SP804, and the processing in steps SP805 and 806 is skipped.

[0085] This is, for waveform data of the last block (6 blocks) have already been stored in the loop region buf (i, 2) in the DMA transfer instruction at time t _5, new in the current interruption process This is because there is no block to be read out. Thus, in FIG. 4 (b), the time t ₆ after the address signal Sad2 is not output.

[0086] (iv) LES interrupt processing at time t ₇ Next, after the loop end interrupt command LES occurs at time t _7, the processing proceeds to step SP801, the variable C
Since NT (i) is “6”, which is the number of the last block, it is determined “YES” here, and the processing proceeds to step SP802.
Proceed to. In step SP802, the start / stop signal SS (i) is set to "0". This allows
The waveform reproduction of the channel ends, and the channel is opened in preparation for another waveform reproduction.

[0087] When the interrupt instruction YS in YS interrupt processing time t ₃₁ in the first to fourth channels are generated, FIG. 6
The YS interrupt processing routine shown in FIG. When the process is started in the figure, in step SP201, the interrupted channel number is substituted for a variable i. Next, when the process proceeds to step SP202, the RAM
8, a queue corresponding to the current interrupt is searched. Since the interrupt timing address Y that caused the interrupt command YS was originally set in the register 46 based on the queue data in the RAM 8, the current interrupt is stored in the queue data area of the channel. There is always queue data corresponding to.

When a queue is searched, its queue contents Q
Based on E, volume control or filter control is performed. That is, according to the queue content QE, FC
The contents of the D register 48, the VCD register 52, etc. are set. Next, when the process proceeds to step SP203, the variable
In the block of the number indicated by CNT (i), it is determined whether or not there is another queue. In the example of FIG. 4, since there is only one queue in the third block, "NO" is determined, and the YS interrupt processing ends.

In the present embodiment, it is of course possible to set a plurality of queues per "1" block. If a plurality of queues are set in the same block, “YES” is determined when step SP203 is first executed, and the process proceeds to step SP204. In step SP204, based on the queue to be executed first among the remaining queues, that is, the queue with the smallest start time QT, the interrupt timing address Y
Is reset. Therefore, when the address signal Sad1 becomes equal to or higher than the reset interrupt timing address Y, an interrupt command YS is generated again, and various tone controls are executed based on the QE of the corresponding cue data.

LES Interrupt on Channel 8 (Tempo Pulse Processing) When an LES interrupt occurs on channel 8, XS
The / LES interrupt routine (FIG. 5 (b)) is called, and the processing proceeds to step SP1 via steps SP101 and SP102.
Go to 05. In step SP105, FIG.
Is called.
When the process is started in the figure, in step SP501, the variable APCNT is incremented by "1". Note that the variable APCNT is set to a predetermined time initial value when step SP406 is executed first.

Next, when the process proceeds to step SP502, it is determined whether or not there is any event of the song being reproduced in FIG. 3A at the time indicated by the variable APCNT. If "YES" is determined here, the process proceeds to step SP503, and a reproduction process of the event is performed. Next, when the process proceeds to step SP504, the time when the next event occurs is set based on the duration until the next event. If no event exists at the time in step SP502, the process proceeds to step S502.
The processing of P503 and 504 is skipped.

Next, when the process proceeds to step SP505, the variable APCNT is referred to, and it is determined whether or not the end time of the automatic performance has been reached. If the determination is "YES", the process proceeds to step SP506, and the start / stop signal SS (8) of "0" is supplied to the register 42. Thereby, the automatic performance (rhythm performance) stops.

Next, when the processing proceeds to step SP507, the CPU 4 executes processing on the fifth to seventh channels, that is, all the channels assigned for playing the keyboard 2.
Supplies a note-off signal to the sound source 26. As a result, the performance on the keyboard is forcibly terminated, and the process returns to the main routine.

As described above, in this embodiment, FIG.
The subroutine shown in (a) is provided, so that the on / off state of the automatic performance can be set by the event of the automatic performance start / stop switch 14a.
Automatic performance can be stopped automatically when a predetermined end time is reached.

In some known electronic musical instruments, an interrupt is generated at predetermined time intervals using a timer to enable automatic performance of a tempo pulse or the like. Since tempo pulse processing is performed using the loop end interrupt command LES for reproduction, there is no need to provide a timer for generating tempo pulses.

C. Modifications The present invention is not limited to the above-described embodiments, and various modifications are possible, for example, as follows. Modification In the configuration of the embodiment, the tone signal output from the interpolation circuit 32 is directly supplied to the tone color control filter 38, but a circuit for applying various special effects may be provided between them. In this case, if a register for instructing the contents of the special effect is further provided and the contents of the register are changed by the cue data, various special effects can be automatically applied to the reproduced waveform. For example, if an AM modulation circuit and an FM modulation circuit are provided at the subsequent stage of the interpolation circuit 32, for example, the following can be added to the queue contents QE described in the embodiment.

AMD (depth, rate) This cue is used for amplitude modulation given to a tone signal.
The depth is changed, and the modulation depth (depth) and the rate of change (rate) after the change are parameters. Needless to say, when the depth is set to "0", no amplitude modulation is applied. -AMF (frequency, rate) This cue is used for amplitude modulation given to a tone signal.
The modulation frequency is changed, and the modulation frequency after the change and the rate of change (rate) are parameters.

PMD (depth, rate) This cue changes the depth of the frequency modulation applied to the tone signal, and changes the modulation depth (depth) and the rate of change ( Is the parameter. -PMF (frequency, rate) This queue changes the modulation frequency of the frequency modulation applied to the tone signal, and the changed modulation frequency and the change rate (rate) are parameters.

Similarly, in the above embodiment, if a reverb circuit is inserted at an appropriate place, for example, the following can be added to the queue contents QE. REV (on / off) This cue changes the on / off state of the reverb, and the changed state (on or off) is a parameter.

REVD (depth, rate) This queue changes the reverb depth, and the reverb depth (depth) and the rate of change (rate) after the change are parameters. REVT (delay time, rate) This queue changes the reverb delay time, and the delay time (delay time) after change and the rate of change (rate) are parameters. As described above, the contents of the cue data may be appropriately determined according to various effect circuits and the like provided in the electronic musical instrument.

Modified Example In the above embodiment, the routine shown in FIG. 8A was executed when the event of the waveform file sounding start instruction switch 14d was detected, but the timing of executing the routine shown in FIG. May be determined based on events generated by a sequencer or a rhythm machine. For example, in an automatic performance performed based on a tempo pulse generated in a tempo pulse generation process using the eighth channel, a waveform start is set as a predetermined event, and when the event is reproduced, the sound of the waveform file is started. In this case, it is possible to automatically start generating the waveform file at a desired position during the performance of the music.

FIG. 10 shows a specific time chart of such an operation. In the figure, the key code C is obtained at time T-1 by the tempo pulse processing subroutine (FIG. 7B).
3 is note-on and at time T + 6 it is note-off. At time T + 1, reproduction of a waveform file is instructed as an event of automatic performance.
Therefore, an ensemble of the automatic performance and the reproduced musical tones of the waveform file is performed from this point. It is necessary to prepare the first waveform starting at the waveform start event in an attack area on the waveform RAM 22 prior to the reproduction of the event. Do it automatically. Alternatively, before the waveform start event, the song is set to reproduce an event that instructs the loading of the leading waveform of the waveform into the attack area. Alternatively, the player may instruct preparation by manual operation.

As described above, during reproduction of a waveform file, various cues such as reverb depth designation or note-on of an arbitrary key can be designated. Further, in the illustrated example, only one waveform file is reproduced, but it is also possible to perform various ensembles by reproducing the corresponding waveform files at a plurality of locations of the music.

The sequencer data in the RAM 8 is the same as that of a general sequencer, and its preparation method will not be described in detail. On the other hand, the cue data can be set in real time while playing the waveform file of the waveform for which you want to set events of various cues, the waveform shape of the waveform file can be displayed on the display unit, and the bar graph display of the waveform playback time Thus, points on the waveform reproduction time axis may be designated and set one by one. In addition, the cue event that has been set may be made to be able to delicately change the position on the time axis, edit the event type, the event content, and the like, in a manner similar to the latter input method. When the position on the time axis is changed, the interrupt register 46
Is changed according to the designated position.

In this embodiment, the waveform start is excluded from the cue event in order to avoid complication. However, the waveform start may be further started by the cue event. In this case, since the number of channels for playing the waveform file is limited, the total number of waveform starts is limited when setting the cue, and the original waveform is automatically faded out in the cue that instructs the next waveform generation. It is better to take action.

[0106]

As described above , claims 1 and 8 have the following features.
According to the invention, the timing at which the event should occur is determined.
Timing data, type data indicating the type of control target
Store queue data consisting of data and parameters
Queue data storage means and sample data readout
Generated from the address generation means that serves as the reference for imaging
Based on the read address (or the accumulated value of the accumulating means),
-Included in the queue data stored in the data storage means
At the timing indicated by the timing data, the type data and
Means for generating queues and parameters;
The control target of the type indicated by the type data is
Control means to control the playback speed of music.
Even if it is changed, the effect should be given at the correct timing.
Can be. Also, the timing is stored in the queue data storage means.
Queue consisting of data, type data, and parameters
Since data is stored, multiple storage
Several types of parameters to be controlled can be mixed and stored.
And the configuration can be simplified. Also,
According to the third aspect of the present invention, the performance for storing the performance data is stored.
Performance data storage means and performance data storage means.
An automatic performance is performed based on the performance data, and
Automatic performance means for generating a waveform generation event included in the program
And the sample at the beginning of the waveform
The pull data is stored in the performance data storage means.
Based on the performance data, prior to the waveform generation event.
Preparation means for reading from a recording medium and writing to the storage means;
As the waveform generation progresses in accordance with the waveform generation event,
And sequentially reads the sample data from the recording medium and stores the sample data in the storage medium.
Supply means for supplying to the means, and occurrence of the waveform generation event
In response to the above, first, the sample of the head part is stored by the storage means.
After reading the data, the data is continuously supplied by the supply means.
Sample data supplied next and stored in the storage means
Reading means for reading the data, and the sun read by the reading means.
Means for generating a waveform signal based on the pull data.
Therefore, load a large amount of waveforms into RAM beforehand.
You can play automatically without it. Also, automatic performance
Sample data used before the waveform generation event
At the correct timing for automatic performance.
And generate actual musical tones. Therefore, according to the present invention, it is possible to realize each of the above requirements (1) that satisfactory - a (2) <br/> tone signal reproducing apparatus.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an embodiment.

FIG. 2 is a block diagram of a main part of one embodiment.

FIG. 3 is a diagram showing various data formats in one embodiment.

FIG. 4 is a time chart of one embodiment.

FIG. 5 is a flowchart of a control program according to one embodiment.

FIG. 6 is a flowchart of a control program according to one embodiment.

FIG. 7 is a flowchart of a control program according to an embodiment.

FIG. 8 is a flowchart of a control program according to one embodiment.

FIG. 9 is a flowchart of a control program according to one embodiment.

FIG. 10 is a time chart of a modified example.

[Explanation of symbols]

 4 Central Processing Unit 18 Hard Disk 20 DMA Controller 22 Waveform RAM 26 Sound Source 38 Tone Control Filter 40 Volume Control Mixer 34 Address Counter

Claims (9)

(57) [Claims] 1. A storage means for storing waveform sample data, and timing data indicating a timing at which an event should occur.
Data, type data indicating the type of control target, and parameters.
Data storage device that stores queue data consisting of data
A stage, an address generating means for generating a read address for sequentially reading the sample data , and supplying the read address to the storage means; and a waveform for generating a waveform signal based on the sample data read according to the read address. Generating means , based on the read address generated by the address generating means.
Cue data stored in the cue data storage means
Type at the timing indicated by the timing data included in
Queue generation means for generating data and parameters, and the type of control target indicated by the generated type data
Control means for controlling according to data.
That tone signal reproducing apparatus. 2. The tone signal reproducing apparatus according to claim 1, wherein said control means includes: tone generation control, modulation control, tone color control,
A tone signal reproducing device that performs at least one of localization control and reverberation control. 3. A performance data storage device for storing performance data.
A stage, automatic performance means for performing an automatic performance based on the performance data stored in the performance data storage means , and generating a waveform generation event included in the performance data; and a recording medium for storing waveform sample data. a storage means for the sample data is read and written, the sample data of the head portion of the waveform that Ru is generated in the waveform generation event, stored in the performance data storage unit performance
Preparation means for reading from the recording medium prior to the waveform generation event based on the data and writing to the storage means; and sequentially reading sample data from the recording medium in accordance with the progress of generation of a waveform corresponding to the waveform generation event. Supply means for supplying to the storage means, and in response to occurrence of the waveform generation event,
First, after reading out the sample data at the head, reading means for reading out the sample data successively supplied by the supply means and stored in the storage means, and the sample data read out by the reading means Means for generating a waveform signal based on the tone signal. 4. The tone signal reproducing device according to claim 3, wherein a data specifying a waveform to be reproduced in the performance data is stored in a header of the performance data. 5. The tone signal reproducing apparatus according to claim 3, wherein the preparation means prepares the sample data of the head portion when the automatic performance means starts automatic performance. 6. The tone signal reproducing apparatus according to claim 3, wherein said performance data further includes a waveform preparation event, and said preparation means is provided when said automatic performance means reproduces said waveform preparation event. Is a tone signal reproducing device for preparing the sample data of the head portion. 7. The tone signal reproducing device according to claim 3, wherein the automatic performance means further generates a tone generating event, and the tone signal reproducing device further generates a tone in response to the tone generating event. A tone signal reproducing device including a generating means. 8. A storage means for storing waveform sample data, and timing data indicating a timing at which an event should occur.
Data, type data indicating the type of control target, and parameters.
Data storage device that stores queue data consisting of data
Stage, read rate generating means, accumulating means for accumulating a read rate generated by the read rate generating means, and reading sample data at a position corresponding to the accumulated value of the accumulating means from the storage means. Reading means; waveform generating means for generating a waveform signal based on the sample data read by the reading means; and arbitrary cue data in response to an operation of an operation element.
And queue data setting means for writing in the data storage means, based on the accumulated value of the accumulating means, the queue data storage means
Data included in the queue data stored in the
Type data and parameters are issued at the timing indicated by
The generated queue generation means and the type of control target indicated by the type data generated are parameters.
Control means for controlling according to data.
That tone signal reproducing apparatus. 9. The tone signal reproducing apparatus according to claim 8, wherein said control means comprises : tone generation control, modulation control, tone color control,
Localization control, reverberation control, and less of tempo pulse generation
A tone signal reproducing device that performs one control .