JP3358707B2 - Music generator - 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 generator having an effector for controlling the waveform of tone data.

[0002]

2. Description of the Related Art Conventionally, a musical tone is generated with a tone based on tone data selected by a user, and various effects are added by switching an algorithm of an effect (effect imparting processing such as equalizing or reverb applied to the tone). There has been a musical sound generation device provided with an effector capable of performing the following. In such a tone generator, the tone color of the tone data to be generated is controlled according to the selected tone color data. In addition, control is performed so as to select an algorithm of an effect process to be applied to generated tone data in accordance with the selected tone color data.

[0003]

There are two types of effect processing performed by the effector, one for stereo and the other for monaural. In the above-described tone data in the conventional tone generator, tone data is generated in monaural. The designation of whether to generate the sound in stereo or stereo and the designation of effect processing in monaural or stereo have been performed independently of each other. Therefore, depending on the setting of the tone color data, there is a problem in that monaural effect processing is performed on musical sound data generated in stereo, and stereo effect processing is performed on musical sound data generated in monaural.

[0004] It is an object of the present invention to provide a musical sound generating apparatus in which effect processing is automatically performed in an optimal manner in accordance with an output form (monaural or stereo) of generated musical sound data. And

[0005]

In order to achieve this object, a musical tone generating apparatus for a single musical tone part having a plurality of sounding channels is provided with a tone color of musical tone data to be generated. Tone number for selection
Tone number input means for inputting a tone number and corresponding to the tone number
A storage means for storing tone color data for each timbre de <br/> over data, respectively, the waveform data used for tone generation state
Tone-like or those which generate a musical tone data mono, and those containing designation information specifying either and generates a musical tone data of a stereo, which is the input
Waveform data selecting means for selecting the waveform data based on the tone color data corresponding to the number, sounding instruction information inputting means for inputting a sounding instruction for instructing generation of musical tone data based on the selected waveform data; Sound data generating means having a sounding channel, wherein the designation information indicates whether the selected waveform data is to generate monaural sound data or stereo sound data in accordance with a sounding instruction; And the generation of the selected monaural tone data using a single sound channel or the use of at least two sound channels corresponding to the number of output systems of the sound data in accordance with the judgment result. Performs one of the generation of the selected stereo musical sound data, and generates the musical sound data for the plurality of tone generation channels. And accumulating means for accumulating the tone data of a plurality of tone generation channels generated by the tone data generating means, wherein the selected waveform data generates monaural tone data. In some cases, the result of accumulating the tone data of all tone channels is output as a single tone signal. If the waveform data is for generating stereo tone data, the tone of the tone channel corresponding to each output system is output. A plurality of data signals each of which is obtained by accumulating data, and effect processing means for effecting the cumulatively calculated musical sound signals, wherein the waveform data is selected by the waveform data selecting means. When selected,
Whether the selected waveform data is to generate monaural tone data or stereo tone data is determined based on the designation information, and according to the determination result, a monaural tone data is generated. Automatically switching to effect processing for the single tone signal by an algorithm, or effect processing for each of the plural tone signals by a stereo algorithm.

[0006] For a single tone part, a tone of one part, that is, a tone of one tone is generated. For example, a sound source to which only one MIDI channel (one tone color is assigned) is for a single tone part. The stereo may be a multi-channel stereo having any number (plurality) of output channels.
When monaural is specified, the same monaural tone data is normally output from a plurality of output channels, but monaural tone data may be output from only one predetermined output channel. The stereo algorithm is, for example, an algorithm that processes the LR in a divided manner in the case of LR stereo. The monaural algorithm is an algorithm for processing LRs collectively.

[0007]

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

FIG. 1 is a block diagram of an electronic musical instrument to which a tone generator according to the present invention is applied. The electronic musical instrument includes a central processing unit (CPU) 101, a read only memory (ROM) 102, and a random access memory (RA).
M) 103, MIDI (Musical Instrument Digital Int)
interface 104, display (PI) 105, panel switch (PSW) 106, tone generator (TG) 107, waveform memory (WM) 108, digital / analog (D / A) converter 109, sound system (SSL) , SSR) 110L, 110R, and a bus line 111.

The CPU 101 controls the operation of the electronic musical instrument as a whole. The operation of the CPU 101 will be described later with reference to the flowcharts of FIGS. ROM102
Stores a control program executed by the CPU 101 and various constant data. RAM 103 is a CPU
Used for the work area 101 and the like. MIDI 104
This interface is used to connect to an external MIDI device, and an MIDI message such as program change or note-on is sent from the external MIDI device. The display 105 is a display device provided on a panel of the electronic musical instrument, and displays various information. The panel switches 106 are a plurality of switches provided on a panel of the electronic musical instrument. Panel switch 106 includes a part selection switch for selecting a part to be pronounced by the electronic musical instrument. Since one part corresponds to one MIDI channel, selecting a part means specifying one MIDI channel assigned to the electronic musical instrument.

A sound source section 107 reads out waveform data from a waveform memory 108 based on an instruction from the CPU 101,
Processing such as adding effects is performed and output as musical sound data. The D / A converter 109 includes the sound source unit 10
7 is converted from digital to analog (D / A) to output an analog musical sound signal. D /
The analog tone signal output from the A converter 109 is output to a sound system (SSL, SSR) 110L, 110R.
It is emitted by.

The tone generator 107 has 32 sound channels. The sound source unit 107 is a sound source of a waveform memory reading method, and the form of the waveform data used for generating a musical tone is stereo (left and right (LR) two-channel output) or monaural. From the output side of the musical sound,
Since there is an output on the left (L side) and an output on the right (R side),
These are referred to as an L-side output channel and an R-side output channel. If the tone selected by the user is a tone that uses waveform data in a monaural output mode, L
The same monaural tone data is output from the side output channel and the R side output channel. In this case, 32 sounds can be generated simultaneously. If the tone selected by the user is a tone using waveform data of a stereo output mode,
Stereo tone data having different waveform shapes are output from the L-side output channel and the R-side output channel. That is, stereo left musical tone data is output from the L output channel, and stereo right musical tone data is output from the R output channel. In this case, 16
Simultaneous sound generation is possible.

In FIG. 1, reference numeral 121L denotes a signal line for outputting tone data of the L-side output channel of the tone generator 107, and reference numeral 121R denotes a signal line for outputting tone data of the R-side output channel of the tone generator 107. D / A converter 109
Converts the L-side and R-side tone data output from the tone generator 107 independently from one another into D / A signals.
Output from 2L and 122R, respectively. The L-side analog tone signal from the output signal line 122L is emitted by the sound system (SSL) 110L, and the output signal line 122R.
From the R-side analog tone signal from the sound system (SS
R) Sound is emitted at 110R.

FIG. 2 is a detailed block diagram of the tone generator 107 shown in FIG. The sound source unit 107 includes a sound source register 201,
Readout and interpolation circuit 202, digital filter 2
03, amplitude envelope generating circuit 204, multiplier 20
5. Accumulator (accumulator ACC1, ACC2) 20
6, 207, an equalizer circuit 208, and a reverb (reverberation) circuit 209.

The sound source register 201 is a sound source register for storing designation information (such as a command and parameter information for the sound source unit 107) sent from the CPU 101. CP
The U101 sets the predetermined designation information in the tone generator register 201, thereby instructing the start of sound generation and changing the tone.

Upon receiving a tone generation start instruction from the CPU 101, the tone generator 107 starts a tone data generation operation for a plurality of time-division multiple channels (32 channels). First, the read-out and interpolation circuit 202 reads out the waveform data in the waveform memory 108 at a speed corresponding to the pitch specified by the note number, and performs an interpolation process. The interpolated musical tone waveform data is filtered by a digital filter 203, multiplied by an envelope waveform output from an amplitude envelope generating circuit 204 by a multiplier 205, and subjected to amplitude control. The result of this multiplication (musical sound waveform data to which an envelope has been added) is input to accumulators 206 and 207. The accumulator 206 corresponds to the L-side output channel, and the accumulator 207 corresponds to the R-side output channel. Accumulators 206 and 207
Then, the tone waveform data from the multiplier 205 is channel-accumulated to generate tone data of the L-side output channel and tone data of the R-side output channel.

The processing of the above-mentioned blocks 202 to 207 operates on a 32 channel time division basis. When the waveform data used is stereo, the CPU 101
When assigning any one of the two sound channels (two channels) and instructing the start of sound generation, the panning parameters are appropriately designated (set in the tone generator register), and the tone data of the sound channel is set to the L side. Accumulator 2
06 or to the accumulator 207 on the R side (weighting may be applied to both the left and right sides and input to both sides). As a result, the tone data generated in one sounding channel mainly flows to the L side, and the tone data generated in another sounding channel mainly flows to the R side, thereby generating left and right stereo tone data.
When the mode of the waveform data to be used is monaural, CP
When one of the 32 sound channels (one channel) is assigned from U101 and the start of sound generation is instructed, the tone data of the sound channel is stored in the accumulator 2 on the L side.
The panning parameters are specified so as to be input to the sub-command 06. As a result, the tone data generated in all the tone generation channels is caused to flow to the L side to generate monaural tone data.

The accumulated values on the L and R sides are input to an equalizer circuit 208. The equalizer circuit 208 performs an equalizing process on the musical sound data, that is, a process for controlling a signal level for each predetermined frequency band. The tone data after the equalizing process is input to the reverb circuit 209. The reverb circuit 209 performs a reverb adding process on the musical sound data and outputs a processing result.

As described above, the tone generator 107 has 32 sound channels, but there are two output channels viewed from the output side: an L-side output channel and an R-side output channel. When the form of the waveform data used for generating a tone is stereo, the processing of the equalizer circuit 208 and the reverb circuit 209 is performed independently on the L side and the R side. 211L, 212L, 213L, 214L is L
2 shows the flow of the musical tone data on the side 211R, 212R, and 2R.
13R and 214R show the flow of the tone data on the R side. On the other hand, if the form of the waveform data used for generating the musical tone is monaural, only the L-side output channel is used without using the R-side output channel and 211L, 212L, 213
Monaural tone data is generated according to the flow indicated by L.
However, tone data (monaural) as a result of the reverb addition performed by the reverb circuit 209 is output from the signal lines 214L and 21L.
The same is output from both 4Rs (apparently, the same tone data is output from both the L and R output channels).

The tone generator 107 is specifically constituted by a digital signal processor (DSP). Therefore, the effector portion (the equalizer circuit 20)
8 and a processing portion of the reverb circuit 209) are defined by a microprogram and parameters given to the DSP. In this electronic musical instrument, the algorithm of the effector is automatically switched between monaural and stereo according to whether the form of the waveform data used for tone generation in the tone generator 107 is monaural or stereo. The algorithm for stereo is an algorithm for processing the LR in a divided mode, and the algorithm for monaural is an algorithm for processing the LRs collectively. Also, the tone data contains information for specifying the waveform data to be used, and the form of the used waveform data is either monaural or stereo. Therefore, the output form of the tone data is monaural or stereo. Will be included. That is,
In this electronic musical instrument, when the user specifies a tone,
The algorithm of the effector is automatically switched to monaural or stereo according to the designation of monaural / stereo of the tone color data corresponding to the tone color. Hereinafter, the effector portion will be described in detail focusing on the equalizer circuit 208.

FIG. 3A is a block diagram showing an algorithm of the equalizer circuit 208 when the form of the waveform data used for generating the tone in the tone specified by the user is monaural. FIG. 3B is a block diagram showing an algorithm of the equalizer circuit 208 when the form of the waveform data used for generating the tone in the tone specified by the user is stereo.

In FIG. 3A, in the processing of the monaural equalizer, the accumulation result (monaural sound data) of the L-side accumulator 206 is obtained by a low shelving filter (LSF) 301 and a peak dip.・ Filter (PD
F) 302, and a high shelving filter (H
SF) 303 to perform equalizing processing and output. LSF is low frequency band, PDF is medium frequency band, H
SF is a filter for increasing or attenuating (boost / cut) the amplitude level in a high frequency band. As a result, a monaural three-band equalizer is realized.

In FIG. 3B, in the processing of the stereo equalizer, the accumulation result (L side musical tone data) of the L side accumulator 206 is equalized through the LSF 311 and the HSF 312, and the R side accumulation is performed. The accumulation result (R side musical sound data) of the arithmetic unit 207 is LSF321 and HSF3.
The equalizing process is performed via the output signal 22 and the tone data is independently output as the tone data of the L-side output channel and the R-side output channel.

When monaural is designated, FIG.
Since only one series of processing is required as in (a), a large number of DSP microprogram steps can be secured, so that the number of bands to be equalized can be increased. In FIG. 3A, the equalizer has three bands of a high band, a middle band, and a low band. When stereo is designated, it is necessary to perform the processing of two lines on the L side and the R side as shown in FIG. 3B, so that the number of bands in one line cannot be so large. In FIG. 3B, two bands of high and low bands are provided on the L and R sides, respectively.

Next, the operation of the CPU 101 in FIG. 1 will be described in detail.

FIG. 4A shows a flowchart of a main routine executed by the CPU 101 when the power of the electronic musical instrument is turned on. First, in step 401, various initial settings are performed. Next, a MIDI process is performed in step 402, a panel process is performed in step 403, and the process returns to step 402. The MIDI processing in step 402 is as follows.
This is a step of detecting a MIDI message input from the MIDI 104 and performing processing in accordance with the message, which will be described later in detail with reference to FIG. The panel process of step 403 is a step of detecting when the user operates the panel switch 106 and performing a process according to the operation.

FIG. 4 (b) is a flowchart showing step 40 in FIG. 4 (a).
3 is a flowchart of part selection switch-on event processing executed when the part selection switch is turned on in panel processing of No. 3; The part selection switch is a switch for selecting which part the electronic musical instrument is responsible for, in other words, which MIDI channel message to receive and execute the corresponding process. In the part selection switch-on event processing, step 411 is executed.
Then, the input numerical value (the numerical value indicating the MIDI channel) selected by the user is set in the register RMC, and the routine returns. The register RMC is in charge of M
This register stores the IDI channel number.
A numerical value between 16 and 16 is set. After the part is selected, the electronic musical instrument receives only the MIDI message of the designated MIDI channel and executes the corresponding processing.

FIG. 5A is a flowchart showing the operation of step 40 in FIG.
2 shows a detailed flowchart of the MIDI process 2. MI
In the DI processing, first, at step 501, it is determined whether or not there is MIDI reception data. When there is no received data,
Return as it is. If there is received data, it is determined in step 502 whether the received data is a MIDI channel message. If the message is a MIDI channel message, the value of the MIDI channel in the received data is set in the register m in step 503, and the value of the register m is stored in the register R in step 504.
MC (in step 411 of FIG. 4B, MI to be received)
DI channel number is set). If they match, it means that the received message should be processed by this electronic musical instrument, so that processing corresponding to the received data is executed in step 505, and the process returns. If the value of the register m is different from the value of the register RMC in step 504, it means that the message is not a channel message to be processed by the electronic musical instrument, and the process returns. If the data received in step 502 is not a MIDI channel message, it is a system message, so step 50
At 6, the other message processing is performed, and the process returns.

FIG. 5B is a diagram showing step 50 in FIG.
5 is a flowchart showing processing when the received data is a program change (that is, selection of a timbre) among the processing corresponding to the received data of No. 5; When a program change is received, first in step 511,
The input numerical value (tone color number indicating the tone color to be selected) is set in the register TC. Next, in step 512, the equalizer 208 is monaural (see FIG. 3 (C)) based on the selected waveform of the tone data of the tone number indicated by the register TC (that is, based on whether the form of the selected waveform data is monaural or stereo). a)) or stereo (FIG. 3B). The tone data corresponding to the tone number is stored in the ROM 102.
Shall be stored in This timbre data is data including information indicating whether the output mode is monaural or stereo. If the selected timbre data specifies monaural, the step 512 becomes the equalizer 208 of the monaural algorithm. And send the micro program and parameters for monaural to DSP,
When the selected timbre data designates stereo, processing for sending a stereo microprogram and parameters to the DSP is performed so as to be an equalizer 208 of an algorithm for stereo. After step 512, in step 513, other settings are made based on the selected timbre data, and the process returns.

FIG. 6 is a flowchart of a process executed when the received data is a note-on (sound generation instruction), among the processes corresponding to the received data in step 505 of FIG. 5A. First, in step 601, the received note-on note number is set in the register NN.
In step 602, the selected waveform of the tone color data of the tone color number indicated by the register TC is checked. And step 60
At 3, it is determined whether the mode of the selected waveform is stereo or monaural. If it is monaural, in step 604,
One channel is assigned from the 32 sounding channels (in some cases, a tone that is already being sounded is truncated). Next, in step 605, a tone generator register for one channel is set based on the currently selected tone color data. Since the form of the waveform data used for generating the musical sound is monaural, the panning parameters are set so that the musical sound data is transmitted to the L side. Then, in step 606, note-on is sent to one set sounding channel, and the process returns. Thus, as described with reference to FIG. 2, equalization and reverb addition are performed using only the L-side output channel to generate monaural tone data, and the tone data (monaural) resulting from the reverb addition is converted to a signal line 214L. , 21
Output from both 4Rs.

If the selected waveform is stereo at step 603, at step 607, two channels necessary for stereo sound generation are allocated from the 32 sound channels (in some cases, the tones already being sounded are truncated). . Then, in step 608, the tone generator register of the L side output channel is set based on the currently selected tone color data, and in step 609, the tone generator register of the R side output channel is set based on the currently selected tone color data. The setting of the tone generator register of the L-side output channel is performed by setting a parameter in the tone generator register so that data for generating the L-side musical tone is generated on one of the two sound channels of the two channels secured in step 607. In this case, the panning parameters are set so that the musical tone is mainly output to the L-side output channel (including output to both the LRs with appropriate weighting). The setting of the tone generator register of the R side output channel means that the data for generating the tone on the R side is stored in step 6.
07 is to set a parameter in the tone generator register so as to sound on the other sounding channel of the two channels secured in step 07. In this case, the parameter of panning is to output the musical tone mainly to the R-side output channel (weighted appropriately). (Including output to both LRs). Next, in step 610, the 2
Sends note-on to one sounding channel and returns. As a result, as described with reference to FIG. 2, stereo tone data that is effect-processed independently for the L-side output channel and the R-side output channel is generated and output.

In the embodiment of the present invention, the effect processing is automatically switched between monaural and stereo for the selected tone color, depending on whether the mode of the waveform data is monaural or stereo. You don't have to check each time whether the tone is monaural or stereo. Tone switching is performed by a MIDI program change. In this case, the present invention is applied since the received tone generator may use a stereo waveform or a monaural waveform even with the same tone number. Otherwise, the MIDI sending side must check the tone color setting (whether the waveform data of the tone number is stereo or monaural) of the receiving tone generator before sending out the MIDI effect control data. According to the present invention, such a complicated procedure is not required.

In the above embodiment of the present invention, the LR
, But a multi-channel stereo having a larger number of channels may be used. When monaural is designated, all the generated musical tones are output to the L side inside the tone generator 107 for effect processing, and the output portion of the effector (the reverb circuit 209) is output.
Output portion), the same monaural tone data is output to both the L-side and R-side outputs.
07, monaural tone data may be output only from one output channel, and the same tone may be emitted left and right by the sound system. Further, in the embodiments of the present invention described above, an equalizer is described as an example, but the present invention can be similarly applied to other effectors.

In the above embodiment of the present invention, an example in which the present invention is applied to a waveform memory sound source has been described. However, the present invention can be applied to a sound source other than the waveform memory sound source. The sound source may generate waveforms of different shapes for L and R in stereo, and generate a waveform of one shape in monaural.

Next, a modified example of the above embodiment of the present invention will be described. This modification adds processing when there are no more than two empty channels when assigning sounding channels when the mode of the selected waveform is stereo. Specifically, the note-on reception processing of FIG. 6 in the embodiment of the present invention is replaced with the processing of FIG.

In FIG. 7, steps 601-610
Is the same process as Steps 601 to 610 in FIG. In the process of FIG. 7, when the mode of the selected waveform is stereo in step 603, step 701 is executed.
To determine whether there are two or more empty channels. If there are two or more channels, step 60
Proceeding to 7, the same processing as in the above-described embodiment of the invention is performed. If there are not two or more vacant channels, it is determined in step 702 whether or not the vacant channel is one. If the number of available channels is one, the process directly proceeds to step 704. If the number of available channels is not one, that is, if there are no available channels, a truncation process is performed in step 703 to secure one channel, and the process proceeds to step 704.
In the truncation process, one of a plurality of sound channels is muted according to a predetermined rule (for example, the oldest key released, the one having the smallest amplitude, and the like). This is the process of allocating for this new pronunciation.

Next, in step 704, a sound source register for one channel reserved is set based on the currently selected timbre data. In step 705, the reverb circuit 209 is instructed to give a pseudo stereo effect. The pseudo-stereo effect means that a series of musical sound signals is
This is a process for realizing pseudo stereo by branching to R and making the phase, frequency characteristic, amplitude, etc. different for L / R. Then, step 706
Then, note-on is sent to one set sounding channel, and the process returns. As a result, the tone generated with the pseudo stereo effect by generating the tone in one channel is converted to L / R
And output.

In this modification, the reverb circuit 20
In step 9, it is necessary to add a normal stereo effect and a pseudo stereo effect in parallel. On the other hand, in FIG. 2, the accumulators 206 and 207 are provided corresponding to the L / R, respectively. In the monaural case, the accumulators 206 and 207 flow to the L side. The output (the output to which the pseudo-stereo effect is to be added in the subsequent stage) is used as the accumulator 20.
6 and input to the reverb circuit 209, so that the reverb circuit 209 cannot add a normal stereo effect and a pseudo-stereo effect in parallel. Therefore, in this modification, a monaural accumulator is provided separately from the accumulators 206 and 207, and the output of the channel set in step 704 is accumulated by the monaural accumulator. Then, the output of the accumulator is input to the reverb circuit 209, and the output of the stereo and the output of the monaural are separated and input to the reverb circuit 209. In the reverb circuit 209, a normal stereo effect and a pseudo-stereo effect are respectively added to a stereo output and a monaural output which are separately input. As described above, the reverb circuit 209 adds the normal stereo effect and the pseudo stereo effect in parallel.

In the above modification, when the number of available channels is 0, one sounding channel is secured by the truncation processing to add the pseudo stereo effect. In this case, two channels are secured by the truncation processing. Hereinafter, the processing of steps 608 to 610 may be performed. If only one sounding channel is truncated, there is a problem that one of the L / R musical tones for one sounding instruction is muted and the other is continuously sounded. However, the two channels are truncated as described above. If this is ensured, this problem can be prevented.

[0039]

As described above, according to the present invention, the algorithm of the effector is automatically switched between monaural and stereo according to the output mode (monaural or stereo) of the generated musical sound data. In this way, the effect processing is automatically performed in an optimum manner according to the output mode (monaural or stereo) of the generated musical sound data. That is, it is possible to automatically avoid setting inappropriate effects.

[Brief description of the drawings]

FIG. 1 is a block diagram of an electronic musical instrument to which a tone generator according to the present invention is applied.

FIG. 2 is a detailed block diagram of a sound source unit.

FIG. 3 is a block diagram showing an algorithm of a monaural and stereo equalizer circuit;

FIG. 4 is a flowchart of a main routine and a part selection switch-on event process executed by a CPU;

FIG. 5 is a flowchart of MIDI processing and program change reception processing.

FIG. 6 is a flowchart of a note-on reception process.

FIG. 7 is a flowchart of a note-on reception process according to a modified example.

[Explanation of symbols]

101 central processing unit (CPU), 102 read-only memory (ROM), 103 random access memory (RAM), 104 MIDI, 105 display (PI), 106 panel switch (PSW), 107
... sound source section (TG), 108 ... waveform memory (WM), 10
9 Digital / analog (D / A) converter, 110
L, 110R ... Sound system (SSL, SSR)
111: bus line, 201: tone generator register, 202:
Readout and interpolation circuit, 203: digital filter, 204: amplitude envelope generation circuit, 205: multiplier, 206, 207: accumulator (accumulator ACC)
1, ACC2), 208: equalizer circuit, 209: reverb (reverberation) circuit.

Claims (1)

(57) [Claims] In a tone generator for a single tone part having a plurality of tone generation channels, a tone number for selecting a tone of tone data to be generated is provided.
A tone number input means for inputting, and a storage means for storing tone color data corresponding to the tone number , wherein each tone data is used for tone generation.
Corresponding to the input tone color number, including specification information for specifying whether the form of the waveform data to be generated is to generate monaural tone data or to generate stereo tone data. Based on the tone data
A waveform data selection means for selecting the waveform data Te, and sounding instruction information input means for inputting sound instruction to generate the musical sound data based on the waveform data selected, the tone data generating means having a plurality of tone generation channels In response to the sounding instruction, while determining whether the selected waveform data is to generate monaural tone data or stereo tone data based on the designation information, According to the determination result, generation of the selected monaural tone data using a single sound channel or generation of the selected stereo tone data using at least two sound channels corresponding to the number of output systems of the tone data. One capable of simultaneously generating tone data for the plurality of tone generation channels; Accumulating means for accumulating the tone data of a plurality of sound channels generated by the data generating means; if the selected waveform data is to generate monaural tone data, tone data of all sound channels; Is output as a single tone signal, and if the waveform data is to generate stereo tone data, a plurality of tone tone data of the sound channels corresponding to each output system are accumulated. And an effect processing unit for performing an effect process on the accumulated tone signal, wherein the waveform data selection unit selects the waveform data when the waveform data is selected by the waveform data selection unit. wherein if waveform data and generates a stereo musical sound data or is intended to generate a musical tone data mono With determined based on the constant information, in accordance with the determination result, the effect processing for the single tone signals in an algorithm for mono, or to perform effect processing algorithm for stereo for each tone signals of said plurality of channels A musical sound generating device comprising: a device for automatically switching.