JPH11133970A - Sound source device of electronic musical instrument - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an oscillator structure of sound source device which can read the next musical sound waveform simultaneously, during the time to continue to read out a musical sound waveform from a memory means. SOLUTION: Wave read unit WRU 301 reads a waveform data from a waveform data ROM 399, depending on a wave clock WCK. The waveform data read from the TOM 399 is output to the outer side from a selector SEL 302 as a waveform signal SG0, and when it is informed that the waveform signal reaches to the head of the repeating part in the waveform data, the wave read unit 301 delivers the waveform data to a RAM 303 to memorize there, and when it is informed that is comes to the finish of that part, the wave read unit stops the transfer of the waveform data. And when there is a repeating part in the waveform data read from the ROM 399, after that, the selector 302 converts to the reading from the ROM 303, instead of the reading from the ROM 399, and it starts the next waveform reading process while outputting the waveform signal SG0.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a sound source device for an electronic musical instrument.

[0002]

2. Description of the Related Art To generate a musical tone in a tone generator of an electronic musical instrument, a basic musical tone waveform is generated by an oscillator such as a digitally controlled oscillator DCO or a voltage controlled oscillator VCO, and a digitally controlled filter DCF or a voltage controlled filter is generated. The tone is controlled by a filter such as a VCF, and the volume is controlled by an amplifier such as a digitally controlled amplifier DCA or a voltage control amplifier VCA to be emitted as a musical tone. The generation of the musical tone waveform is performed by reading out the musical tone waveform stored in advance in a storage means such as a ROM or a RAM by the waveform reading means.

[0003]

If the musical tone waveform to be read has a repetitive portion, as described above, if this is continued to be read from a single storage means, access to the storage means becomes impossible during this time. Cannot read another tone waveform. Therefore, even if the processing speed is to be improved, it becomes a bottleneck, and the reading speed of the musical tone waveform cannot be increased, which is one of the factors that hinder the improvement of the processing speed of the entire circuit.

The present invention has been made in view of the above-mentioned problems of the prior art, and does not provide a configuration in which one musical tone waveform can be simultaneously read out from the storage means while the next musical tone waveform can be read out at the same time. It is assumed that.

[0005]

According to the present invention, a tone generator for an electronic musical instrument according to the present invention comprises a first waveform storing means for storing a musical tone waveform including a repetitive portion, and simultaneously storing the repetitive portion when the repetitive portion is read. The waveform can be read out from the second waveform storage means and the first waveform storage means which can be read separately, and when reading the waveform of the musical tone waveform stored in the waveform storage means, A waveform reading unit that can identify that the waveform has reached the beginning of the repetitive portion of the waveform and that it has reached the end of the repetitive portion, and can also read the waveform from the second waveform storage device. When reading the waveform of the musical tone waveform stored in the first waveform storage means, when it is determined that the waveform has reached the beginning of the repeated portion of the waveform,
When the waveform readout means stores the read musical tone waveform in the second waveform storage means, and when the waveform readout means identifies the end of the repetition portion, the waveform readout means performs the readout. As a basic feature, switching from the first storage means to the second storage means is provided.

[0006]

In the above arrangement, when reading out the tone waveform from the first waveform storage means, the waveform reading means stores the repetition portion in the second waveform storage means, and reads the repetition portion from the second time on. If it is attempted to read out the data, the read-out destination is switched to the second waveform storage means, so that the next musical tone waveform can be read out from the first waveform storage means, and the pre-reading can be performed. Thus, the reading speed of the musical tone waveform can be improved, and the processing speed of the entire circuit can be increased.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of a tone generator for an electronic musical instrument according to the present invention will be described below. FIG. 1 is a block diagram showing a circuit configuration of an electronic musical instrument according to an embodiment of the present invention, and FIG. 2 is a block diagram showing a circuit configuration of a sound processing unit (SPU) as a sound source device used therein.

In FIG. 1, MCT, MAD, and MDT are buses of a microcomputer used in an electronic musical instrument, of which MCT is a control bus, MAD is an address bus,
MDT indicates each data bus, and as a configuration for exchanging control signals, address signals and data signals with these, a microprocessing unit MPU2 for performing overall control and arithmetic processing, a sound processing unit SPU1, A storage device 3 including a ROM 3a and a RAM 3b for reading and writing data and programs;
An I / O unit 4 for exchanging data and the like with the outside, a panel interface 5, and a keyboard 6 for transmitting key information corresponding to the key touch by the player are connected. The digital signal transmitted from the SPU 1 is converted into an analog tone signal by a digital-to-analog converter DAC7, and is converted into an amplifier AM.
The signal is amplified by P8 and transmitted as an audio signal from the speaker SPK9.

The structure of the SPU 1 shown in FIG. 2 comprises a low-frequency oscillator LFO100 for giving a sawtooth wave to each circuit,
Digitally controlled oscillator DCO101 which becomes an oscillation circuit as a configuration of a waveform generating means for generating a musical tone waveform
And a digital gain controller D as a configuration of a variable gain circuit capable of changing the gain of the tone waveform.
GC103 and digitally controlled filter DCF104 for removing unnecessary harmonic components from the above tone waveform
And a digitally controlled amplifier DCA 105 for providing an amplitude envelope by the amplitude envelope generator EG114 while amplifying the tone signal.
And In this electronic musical instrument, these circuits (LF
O100, DCO101, DGC103, DCF10
4. DCA 105 and other FGs 108 to 110 to be described later
And EGs 112 to 114 are provided for 16 channels (of course, the number of channels is arbitrary), but they are omitted for simplicity. Also, a digital mixer DMX1 for mixing a tone signal transmitted from each DCA 105 of 16 channels to obtain a polyphonic effect.
06 and a digital effector DEF107 for adding various effects to each tone signal.
Further, the DCO 101 and the DCF 104 have a flicker generator FG1 that gives a fluctuation to the frequency of the generated waveform or a fluctuation to the cutoff frequency of the digital filter based on the envelope given by the envelope generators EG112 and EG113.
08 and 110, and the DGC 103 includes:
A flicker generator FG109 for providing frequency and amplitude fluctuations to the input waveform is provided.
5 is provided with an envelope generator EG114 for adding an envelope to the output waveform.

Since the above-mentioned SPU 1 is configured to obtain a polyphonic effect for 16 channels as described above, each of its constituent circuits is provided with an internal RAM for storing respective input / output signals, and one operation is performed. The signal given from the preceding stage is output to the internal R every cycle (one cycle of MCK described later)
Time-division processing is performed in which data is read from the AM, processed, and written to the internal RAM of the next stage. FIG.
4 is a timing chart showing output timing of a clock circuit (not shown) of PU1. MCK is obtained by dividing the frequency of an externally applied clock several times as necessary, and serves as a system master clock. CK0-3
Is the time-division clock of the oscillator, with the higher number attached to the higher and the lower number attached to the lower (in this example)
It becomes a 4-bit signal. Therefore, there are 2 oscillators = 16 oscillators. These CK0-3
Is supplied to all circuits as a signal indicating the oscillator number.

FIG. 4 shows a circuit configuration of the LFO 100. In this configuration, the oscillation frequency L is stored in the RAM 201.
When the SP is given, the adder ADD202 causes the RAM 20
Previous frequency held at 0 (0 if no previous value)
The oscillation frequency from the RAM 201 is added to the RAM 2 to be taken out as a low frequency signal LAD.
00 and is held as the next frequency to be added.
The added frequency is output by the latch clock LCK at the time of the next addition processing of the ADD 202.

FIG. 5 shows a circuit configuration of the DCO 101 having a first waveform storage unit, a second waveform storage unit, and a waveform reading unit as a configuration of the present invention. In this configuration, a waveform data ROM 399 (described later) is used as the first waveform storage means, and an R data (described later) is used as the second waveform storage means.
The AM 303 further uses a wave read unit WRU 301 as a waveform reading means. The WRU30
1 is the waveform data R based on the wave clock WCK.
Read the waveform data from OM399. The waveform data read from the ROM 399 is stored in the selector SEL30.
2 is sent to the WRU 301, where it is subjected to a vibrato addition process as described later, if necessary, and is output to the outside as a waveform signal SG0. At this time, when the MPU 2 is notified that the beginning of the repetitive part (loop top) in the waveform data has been reached, the WRU 301 sends the waveform data to the RAM 303 and stores it, and the end of the part (loop end) is reached. Is notified from MPU2, W
The RU 301 stops transferring the waveform data. Then, when the waveform data read from the ROM 399 has a repeated portion thereafter, the selector 302
Switching to reading from the RAM 303 instead of reading from 99, and further outputting the waveform signal SG0, starts the next waveform reading process. Also, FIG.
The circuit details of M303 are shown (the operation is the same as that of a normal RAM, so that the RD / WR signal and the like are omitted). A signal line indicating the number of the sounding oscillator is connected to the upper part (or part) of the address. That is, the RA
CK3 of the above-described oscillator is set to Am + n-1 (= Am + 3) of M303, and CK2 is set to Am + n-2 (= Am + 2).
Is connected to Am in the same manner as above (in this example, n = the number of CK = 4). The waveform write / read address is connected to the lower (remaining) address of the RAM 303. The waveform address given to the RAM 303 is, as shown in Table 1 below, an external waveform ROM 399.
Is the value (zy) obtained by subtracting the head address y of the loop from the address z given to. When applying vibrato to the readout waveform, waveform shaping is performed by the vibrato generation unit VGU. That is, the low frequency signal LAD and the vibrato depth VDP are input to the VGU 300, and the waveform shaping (deformation) read from the waveform conversion table by the vibrato wave select signal VWS.
According to the function [f (LAD + VDP) * VDP], the V
The GU 300 extracts a waveform conversion signal for frequency modulation of the waveform data. On the other hand, when the vibrato phase shift VPH is input to the VGU 300, the VW
After S is added to (LAD + VPH), the waveform data is offset to a predetermined phase by ignoring the upper bits, and VGU 300 outputs to WRU 301. On the other hand, when an attempt is made to add a non-periodic fluctuation to the readout waveform, a frequency fluctuation signal FS0 from the FG 108, which will be described later, is applied to the wave read unit WRU 301 to perform the processing.

[0013]

[Table 1]

The floating point conversion circuit FXL102 is
As shown in FIG. 7, the fixed-point format waveform signal SG0 output from the DCO 101 is input, and the MSF 400
To find the MSB (for example, the presence of the MSB at the fifth uppermost bit), shift the value of SG0 by the barrel shifter BSH401 by that amount (for example, shift leftward by 5 bits), and add an exponent (for example, the uppermost bit). It is output to the DGC 103 as a floating-point SG1 (with information on how many bits shifted) including a 4 bits) and a mantissa (for example, lower 12 bits).

FIG. 8 shows a circuit configuration of the DGC 103. In the multiplier 500 shown in FIG.
For the input waveform signal SG1 from the F.102, if necessary, an F which gives a frequency and amplitude fluctuation to an input waveform output from a flicker generator FG109 described later.
S1 is multiplied. In another multiplier 501, the multiplication result is multiplied by a gain GIN for controlling a waveform level. The multiplied value is divided into an integer part and a decimal part by the INTAGE fraction separator I / F 502. Then, non-linear data is read from the function table FNC 503 based on the integer part, and the interpolator IPC 504 interpolates the data by using the decimal part separated by the I / F 502 to obtain a waveform. Waveform signal SG2 with gain changed (and fluctuation added if necessary)
Is output.

FIG. 9 shows a circuit configuration of the DCF 104. The input waveform signal SG from the above-mentioned DGC 103
For II, the IIR filter IIR602 cuts overtone components other than the predetermined frequency band at the cutoff frequency OFS given by the offset value. Growl processing (processing of giving a fluctuation to the cutoff frequency) can also be performed on the cutoff frequency OFS by the growl generation unit GGU600.
That is, the low frequency signal LAD and the growl depth GDP are input to the GGU 600, and the GGU 600 is operated according to the waveform shaping (deformation) function [f (LAD + GDP) * GDP] read from the waveform conversion table by the groove wave select signal GWS. A waveform conversion signal for modulating the cutoff frequency is extracted. On the other hand, this GGU6
00, when the gross phase shift GPH is input, the GWS adds (LAD + GPH) and ignores the upper bits, thereby offsetting the waveform data for frequency modulation to a predetermined phase,
Is output to the ADD 601. In the ADD 601, the waveform conversion signal is added to the cutoff frequency OFS and sent to the co-efficient generation unit CGU 603 for controlling the coefficient of the IIR 602. The CG
In U603, the added value and the cutoff frequency fluctuation signal FS2 from the FG110 described later are input, and the IIR
A control signal is output to 602 to change necessary coefficients in the IIR 602. In the IIR 602, based on the cutoff frequency OFS given the fluctuation, the SG2
To cut off harmonic components outside the predetermined frequency band,
It is output to the outside as a waveform signal SG3.

FIG. 10 shows a circuit configuration of the DCA 105. The input waveform signal S from the DCF 104 described above
For G3, an envelope generator EG11 described later is used.
4 multiplied by the amplitude envelope ES2 from the waveform signal S
G4 is output. At this time, tremolo (periodical change in amplitude) can be applied to SG4 by tremolo generation unit TGU700. That is, the low frequency signal LAD and the tremolo depth TDP are equal to TGU7.
00 and the tremolo wave select signal TWS
According to the waveform shaping (deformation) function [f (LAD + TDP) * TDP] read from the waveform conversion table by
The TGU 700 extracts a waveform conversion signal for amplitude modulation of the waveform. On the other hand, when the tremolo phase shift TPH is input to the TGU 700, the TWS adds (LAD + TPH) and then ignores the upper bits to offset the waveform data for amplitude modulation to a predetermined phase. , TGU 700 output to MUL 702. The MUL 702 multiplies the multiplied value from the MUL 701 by the waveform conversion signal from the TGU 700,
It is output to the outside as a waveform signal SG4.

FIG. 11 shows a circuit configuration of the DMX 106. The input waveform signal SG4 from each of the DCAs 105 for 16 channels is supplied to multipliers MUL803 and MUL803.
PA that is input to the receiver 04 and sets the stereo location
N and (1-PAN) calculated by the subtractor 800 are input to multipliers MUL 803 and 804, respectively, and are divided as left and right stereo waveforms as a result of these multiplications.
Each of these stereo waveforms is added to adders ADD 805 and 8
06, but also to the other multipliers 801 and 802, and the left and right effect level signals ELL and ELR
Are respectively multiplied by the external digital effector DEF107 as left and right effector transmission signals ESL and ESR.
Sent to The signals subjected to the effect processing are sent to the adders ADD 805 and 806 as left and right effector reception signals ERL and ERR, and are added to the stereo waveform without effect. The output is further sent to adders ADD 807 and 808 and added to the waveform signal of the previous channel, then latched by latch circuits LTC 809 and 810, from which it is fed back to adders ADD 807 and 808 and
The above addition process is repeated for six channels (0 to 15),
The signals are output to the outside as stereo signals SGL and SGR for 16 channels. Thereafter, as described above, the floating-point format is converted to the fixed-point format by the fixed-point conversion circuit FLX111, from which the DAC is converted into a serial signal.
7 is output.

FIG. 12 shows a circuit configuration of the flicker generators 108, 109 and 110 that generate fluctuations. In this configuration, the value stored in the RAM 900 is set as the upper bits by the write clock WCK.
If there is a previous value stored in the RAM 901, that value is set as a middle bit and the random number generation circuit NGU 90
The random number generated in 2 is given as a lower bit to the INTAGE fraction separator I / F 903, where it is divided into an integer part and a decimal part. Then, based on the integer part, the non-linear data is read from the function table FNC904, and the interpolator IPC905 interpolates the data using the decimal part separated by the I / F502, and performs interpolation on the data. Is returned to the RAM 901. The value is stored in the RAM 90 at the next write clock WCK.
0, and thereafter the above processing is repeated. The IPC 9
The data of the fluctuation output from the data 05 is a parameter flicker depth FDP of the fluctuation depth and a multiplier MUL9.
06, and the result of the multiplication is output as it is as the frequency fluctuation signal FS1 in the FG109. On the other hand FG
At 108 and 110, the result of the multiplication and the envelope signals ES0 and ES2 output from the envelope generators EG112 and EG113 are added by an adder 907 and output as frequency fluctuation signals FS0 and FS2.

In this configuration, a DAC 7 having a normal configuration is used to convert a digital musical tone signal output from the SPU 1 into an analog signal.
The fixed point conversion circuit FLX111 for converting the floating point format to the fixed point format is provided on the output side of the DMX 106, and is output to the outside from the output side of the DMX 106. In this part, as shown in FIG. 13, the left and right waveform signals SGR and SGL are input to the latches LTC1000 and 1001, and the power mantisser separators PMS1002 and 1003 use the exponent part (for example, the upper four bits) and the mantissa part (for example, Lower 1
2 bits), and the mantissa is shifted (for example, 5 bits to the right) according to the exponent by barrel shifters BSH1004 and 1005 to return to the fixed-point format.
These parallel signals are converted into serial signals and extracted as external output.

In the above-described configuration, in the configuration of the DCO 101, when the WRU 301 reads a tone waveform from the waveform data ROM 399, the WRU 301
When the repetition portion is to be read out for the second time or later, the readout destination is switched to the RAM 303, so that the next musical tone waveform can be read out to the ROM 399. As a result, the reading speed of the musical sound waveform can be improved. In addition, since the floating point conversion circuit FXL102 is provided between the DCO 101 as the waveform generating means and the DGC 103 as the variable gain circuit, the fixed point waveform signal SG0 generated by the DCO 101 is converted into the floating point format and output to the DGC 103. Therefore, the data bit length of the arithmetic processing does not become large, and therefore, the circuit scale of the DGC 103 can be reduced in the circuit configuration of the tone generator of the electronic musical instrument, and the calculation accuracy can be reduced. You will be able to keep it high.
In this configuration, a digital tone signal output from the SPU 1 is converted into an analog signal by using a DAC of a normal configuration.
Therefore, a fixed-point conversion circuit FLX111 for converting a floating-point format to a fixed-point format is provided on the output side of the DMX 106. However, a DAC that can process a floating-point format waveform signal without using a DAC is used. In this case, there is no need to provide the FLX 111. In addition, DCO
If the floating point format waveform data can be stored in the waveform ROM 399 provided in the
Even the configuration of 102 becomes unnecessary.

[0022]

According to the configuration of the present invention described in detail above, the waveform reading means stores the repetition portion in the second waveform storage means when reading out the tone waveform from the first waveform storage means. When the repetitive portion is read for the second time or later, the read destination is switched to the second waveform storage means, so that the next waveform can be read from the first waveform storage means. As a result, it becomes possible to perform pre-reading and the like, so that the reading speed of the musical sound waveform can be improved, and further, the processing speed of the entire circuit can be increased.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a circuit configuration of an electronic musical instrument according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a circuit configuration of a sound processing unit as a sound source device used in the circuit configuration of the electronic musical instrument according to the embodiment.

FIG. 3 is a timing chart showing an output timing of a clock circuit of an SPU.

FIG. 4 is a block diagram showing a circuit configuration of a low-frequency oscillator LFO.

FIG. 5 is a block diagram showing a circuit configuration of a digitally controlled oscillator DCO.

FIG. 6 is a circuit diagram showing details of a circuit of a tone generator cache RAM;

FIG. 7 is a block diagram illustrating a circuit configuration of a floating-point conversion circuit FXL.

FIG. 8 is a block diagram illustrating a circuit configuration of a digital gain controller DGC.

FIG. 9 is a block diagram illustrating a circuit configuration of a digitally controlled filter DCF.

FIG. 10: Digitally controlled amplifier D
FIG. 3 is a block diagram illustrating a circuit configuration of CA.

FIG. 11 is a block diagram illustrating a circuit configuration of a digital mixer DMX.

FIG. 12 is a block diagram showing a circuit configuration of a flicker generator FG.

FIG. 13 is a block diagram illustrating a circuit configuration of a fixed-point conversion circuit FLX.

[Explanation of symbols]

 Reference Signs List 1 SPU 2 MPU 3 Storage device 3a ROM 3b RAM 4 I / O unit 5 Panel interface 6 Keyboard 7 DAC 8 AMP 9 SPK 100 LFO 101 DCO 102 FXL 103 DGC 104 DCF 105 DCA 106 DMX 107 DXF

Claims (1)

[Claims] 1. A first waveform storage means for storing a tone waveform including a repetitive part, a second waveform storage means for simultaneously storing the repetitive part when the repetitive part is read, and separately reading the same. A waveform can be read from the first waveform storage means, and when the waveform of the musical tone waveform stored in the waveform storage means is read, it is determined that the waveform has reached the beginning of the repetitive portion of the musical tone waveform and the end of the repetitive portion. And a waveform readout means for reading out the waveform from the second waveform storage means, and for reading the waveform of the musical tone waveform stored in the first waveform storage means. At this time, if it is identified that the waveform has reached the beginning of the repeated portion of the waveform, the waveform reading means stores the read musical tone waveform in the second waveform storage means. In addition, when it is determined that the waveform has reached the end of the repeated portion, the waveform reading means switches from the first storage means to the second storage means as the storage means for performing the reading. Sound source device for electronic musical instruments.