JPH0345995A - Electronic musical sound generating device - Google Patents

Abstract

PURPOSE:To obtain the electronic musical sound generating device which has small deterioration in sound quality at the lowest price by providing a time difference address generating means which generates respective addresses for reading time difference data having a specific time difference. CONSTITUTION:An address correcting means provided on an address transmis sion line is so constituted as to generate addresses having the time difference DELTA=2<1> in order when an output address of an address counter 33 consists of bits Qo - Qj. In this case, when the time difference is regarded as a DELTA=2<1> sam ple time, the contents of a PCM memory 39 can be set so as to consist of (k) DELTA sample areas. Here, data are read out of the PCM memory 39 while having time differences -DELTA, -2DELTA and -3DELTA and then adjusted in amplitude before their composition, thereby generating data for an effect sound. Consequent ly, the electronic musical sound generating device which has a small deteriora tion in sound quality is obtained at the lowest price.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to an electronic musical tone generating device that gives spatial expansion to an output musical tone signal in an electronic musical instrument using a digital waveform memory.

(Prior Art) Conventionally, digitally processed electronic musical instruments sample musical sounds obtained from musical instruments such as wind instruments and string instruments (hereinafter simply referred to as musical instruments to distinguish them from electronic musical instruments that generate musical sounds electrically). Some devices generate musical tones by storing waveform data in a waveform memory and recalling the stored waveform data.

FIG. 7 is a block diagram showing a conventional electronic musical tone generator, which is disclosed in Japanese Patent Publication No. 52-46088. As the waveform memory, a waveform memory 21 is used which samples and stores one cycle of musical tones. That is, N pieces of waveform data obtained by sampling the -period of a musical tone with a fundamental frequency of 10 are stored in the period memory 21. The frequency information generator 22 provides information indicating the fundamental frequency fO to the oscillator 23, and the oscillator 23 provides the N-ary address counter 24 with a signal having a frequency N times the fundamental frequency fO.

The N-ary address counter 24 counts the signals from the oscillator 23 and specifies the address of the waveform memory 21 based on the count value. In this way, data is sequentially read from the waveform memory 21, and continuous waveform data can be obtained by repeatedly reading data for a cycle. This waveform data is enveloped and D/A (digital analog)
The converter 25 applies an envelope and converts it into an analog signal, and a musical tone signal is outputted via the amplifier circuit 26. These N-ary address counter 24 and envelope adding and D/A conversion circuit 25 are connected to the sound generation control circuit 2.
7, and the sound generation control circuit 27 causes the N-ary address counter 24 to start counting by key operation of an electronic musical instrument (not shown). In addition, the oscillator 23
By changing the frequency of the signal output from
Read frequency from waveform memory 21: A! It is possible to change the frequency (pitch) of the musical tone signal.

Furthermore, according to the above device, a waveform memory is prepared to store waveform data obtained by sampling musical tones with different proportions of harmonic components (different harmonic structures), and is used to store waveform data obtained by sampling musical tones with different proportions of harmonic components (different harmonic structures). By selecting a periodic memory based on , it is possible to generate complex musical tones whose spectral structure changes with time and intensity.

Furthermore, the signal obtained from the amplifier circuit 26 is input to the sound effect generating bag@30, and the signal is inputted into the sound effect generating bag@30 to generate an echo (Yamahiko) effect or reverb (
It is output with a characteristic called a reverberation effect.

FIG. 8 shows an example in which a large-capacity PCM memory that stores waveform data within a predetermined period of time is employed as the waveform memory. Components that are the same as those in the device shown in FIG. 7 are designated by the same reference numerals. This device samples all musical tones obtained from musical instruments within a predetermined period of time and stores them in a PCM memory 29. Musical tones are generated by sequentially reading the waveform data stored in the PCM memory 29. The pitch of a musical tone can be changed by changing the reading cycle of waveform data from the PCM memory. This method requires extremely large capacity PCM memory. Therefore,
Sampling is not performed for the last few rounds of the musical tone to be sampled, so that the required memory capacity can be reduced.

That is, during the last few cycles of a musical tone, the envelope of the musical tone is often in a steady state or attenuating (see L1 to L2 of the waveform data in FIG. 9), and the change in the spectral structure is relatively small. Therefore, the last few cycles of musical tones are generated by repeatedly reading out the waveform data (Ll to L2) stored at the last address of the PCM memory 29. This makes it possible to reproduce the musical tones of musical instruments that have long sounding times with a relatively small memory capacity.

The circuit that performs this control is the looping control circuit 28. This control circuit 28 controls the address counter 2 when the output of the counter 24 outputs an address corresponding to L2.
4 is loaded into Ll so that the data (Ll to L2) is read out repeatedly. Also in this device,
A sound effect generator 30 is provided to enrich musical sounds.

By the way, the sound effect generating device 30 is a device that obtains spatial spread of sound.

FIG. 10 shows the principle of the sound effect generator.

The musical tone signal input to terminal 1 is sent to adder 2 and mixing circuit 5.
supplied to The output of the adder 2 is delayed by a delay circuit 3 and fed back to the adder 2 via a volume 4. The feedback rate α is α minus 1. The operation of this circuit will be explained with reference to FIG.

FIG. 11(A) shows the original sound, FIG. 11(B) shows the signal delayed once, FIG. 11(C) shows the signal delayed twice, and FIG. 11(D) shows the signal delayed three times. These signals have feedback rates α of α and α2 α3, respectively.
It's been doubled. All waveforms at the same time are mixed and output.

When the delay amount Δ is large enough that the delayed sound can be clearly recognized, it is called an echo (Yamahiko) effect, and when Δ becomes so small that it cannot be recognized, it is possible to create a characteristic called a reverberation effect. It is.

Generally, the mixing circuit 5 and volume 4 shown in FIG. 10 are constituted by a variable resistor Rx as shown in FIG. 12. As the delay circuit 3, there is a method of analog delay using a BBD (charge transfer device), etc., and a method of performing analog-to-digital (A/D) conversion and digital-to-analog (D/A) conversion and delay using a digital memory element. There is a way.

As sound effect generators, digital systems are more advantageous in terms of frequency band, delay time, and S/N ratio, and are currently in widespread use.

FIG. 13 shows a digital delay circuit, which is composed of an A/D converter 11, a memory control circuit 12, a memory 13, and a D/A converter 14.

(Problems to be Solved by the Invention) The conventional digital sound effect generation device W130 described above requires a fairly large memory 13 depending on the amount of delay. Furthermore, unless the number of quantization bits is increased to a certain extent, there is a problem that quantization noise will occur due to A/D and D/A conversion, so it is not suitable for reducing the number of quantization bits and reducing the circuit size. be.

Therefore, the present invention provides a digital electronic musical sound generation device using a waveform memory, which eliminates the need to use memory, A/D, and D/A converters as a sound effect generation circuit, and achieves sound quality at a minimum cost. An object of the present invention is to provide an electronic musical tone generating device that does not deteriorate.

[Structure of the Invention] (Means for Solving the Problems) The present invention provides a digital electronic musical sound generation device that electronically synthesizes a sound signal by reading data from a waveform memory, which comprises: properly reading data from the waveform memory; In addition to the direct data, a time difference address generating means generates each address for reading at least one time difference data having a predetermined time difference from the direct data, and an address from the time difference address generating means is used to read out the direct data and the time difference data. time-division multiplexing control means for time-divisionally processing the time-difference data and outputting it from the waveform memory, and adding all or time-divisionally adding and outputting the time-division data; and compressing the time-difference data outputted in the time-division manner at a predetermined ratio. and a time difference data output control means for maintaining the direct data and each time difference data at a predetermined time difference at the time of key-on and key-off of a musical tone, without using any delay element or memory. It is constructed to generate a musical tone signal with spatial expansion.

(Function) With the above means, the time difference data can be read out simultaneously in the waveform data generation section, so a special A/D converter, memory, D/A, etc. are required to obtain the time difference data for obtaining sound effects. No converter required.

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

FIG. 1 shows an embodiment of the present invention. In Figure 1, 3
1 is a frequency information generating device, and an oscillating rA 32 obtains an oscillating output according to frequency information outputted from the frequency information generating device.

The clocks with frequencies fO and 4fO obtained from this oscillator 32 are respectively supplied to an address counter 33 and a time division multiplexing counter 52 (;). The address counter 33 basically inputs the count value to the address input of the PCM memory 39. In this embodiment, address modification means is provided on the address transmission path.

Assuming that the output address of the address counter 33 is composed of bits from QO to Qj, this address correction means is configured to sequentially generate addresses having a time difference Δ, assuming that the time difference is Δ−2′. There is.

Now, as shown in FIG. 5, if the above-mentioned time difference is Δ-2' sample time, the contents of the PCM memory 39 can be set to consist of an area of Δ samples. Therefore, data for sound effects can be created by reading data from the PCM memory 39 with time differences of -Δ, -2Δ, and -3Δ, and adjusting and synthesizing the amplitude of the read data.

In the above address correction means, the address counter 3
3 lower addresses Q i -Q j are fed directly to multiplexer 38 and to subscriber 35.3.
Also provided on 6.37. In addition, the address counter 33
The lower bits Q O-Q (1-1) of are directly supplied to the PCM memory 39 . The subscriber 35 outputs an address obtained by subtracting Δ from the address of the upper and lower bits Q i -Q j. Further, the subscriber 36 outputs an address obtained by subtracting 2Δ from the address of the upper bit QL-Qj, and the subscriber 37 outputs an address obtained by subtracting 2Δ from the address of the upper bit QL-Qj.
Outputs the address obtained by subtracting 3Δ from the address of j.

The multiplexer 38 has lower bits QO to Q(1-1
) is supplied, and the upper pin) Qi-
Any address of Qj is selected and supplied.

A control signal to this multiplexer 38 is obtained from a time division multiplexing counter 52.

FIG. 2 shows a timing chart of the selected addresses of the multiplexer 38. (A) in the figure shows the clock CLK, and CB) and (C) in the figure show output bits of QO and Ql of the address counter 33, respectively.

In the same figure, (D) and (E) are time division multiplexing counters 52.
The control signals (MPO, MPI) from the

In the multiplexer 38, (MPO.

MPI ) - (0, O), direct address, (1°0
), a 2Δ delay address is selected, (0, 1), a 3Δ delay address is selected ((F) in the same figure). In other words, the address correction means is the address counter 33, and when one type of address is generated,
The actually generated direct address, the Δ delayed address, and 2
Four of the Δ delay addresses and 3 delay addresses are obtained at the same time. These addresses are then selected by the multiplexer 38 in a time-division manner.

For this reason, as shown in FIG. 2(G), the PCM memory 39 receives address time difference (-Δ) (-
2Δ) (−3Δ) (hereinafter referred to as 1Δ data, −2Δ data, and −3Δ data) are output in a time-division manner.

Therefore, by adjusting the amplitude of data with a time difference, for example, -Δ data can be reduced to 1/2, -2Δ data to 1/4, -3Δ data
By outputting the data at 1/8 and creating and outputting data in the same manner for the next clock period, it is possible to obtain sound effects such as echoes.

For this purpose, the output of the PCM memory 39 is supplied through amplitude adjustment means to a multiplexer 48, and via this multiplexer 48 to an enveloping and D/A conversion circuit 49. The multiplexer 48 also receives the control signal (MP(1, M
PI ), direct data, -Δ data, -
2Δ data and -3Δ data are selected and output in a time-sharing manner.

Direct data from the PCM memory 39 is sent to the gate circuit 44.
is supplied to multiplexer 48 via. Also, −Δ
data, -2Δ data, and -3Δ data are each l/2
The signal is supplied to a multiplexer 48 via a multiplier 41, a 1/4 multiplier 42, a 178 multiplier 43, and further via gate circuits 45, 46, and 47. Therefore, −Δ data, −2
Δ data and −3Δ data are each sent to an l/2 multiplier 41,
The amplitude is adjusted in the 1/4 multiplier 42.178 multiplier 43.

The enveloping and D/A conversion circuit 49 and the amplifier 50 to which the output of this circuit is supplied are the same as previously described. Further, the relationship between the sound generation control circuit 51, the envelope applying and D/A converting circuit 49, and the address counter 33 is also as explained with reference to FIG.

Even if the data is output in a time-division manner as shown in FIG. 2(G), the human sense of hearing perceives it as the addition of musical tones based on each piece of data.

By the way, in the above explanation, it is assumed that data with a time difference is output from the beginning, but with electronic musical instruments, echoes do not appear from the moment the key switch is turned on, but after a while. occurs. For this reason, in this embodiment, the time difference data output control circuit 53 is configured to operate after the key switch is turned on.

The time difference data output control circuit 53 detects that the key switch is turned on based on the control signal from the sound generation control circuit 51. Upon obtaining this detection, the time difference data output control circuit 53 controls the gate circuits 44, 45, 46, 47 as shown in the timing chart of FIG.

That is, FIGS. 3A and 3B show pulses that serve as reference signals for the time difference data output control circuit 53, and are the outputs Q (1-1) and Q (1-2) of the address counter 33. (C) to (F) in the figure are gate control signals output from the time difference data output control circuit 53, respectively, and D
EO controls the gate circuit 44, and DEI controls the gate circuit 4.
5, DE2 controls the gate circuit 46, DE3
controls the gate circuit 47. Time t1 in FIG. 3 is the time when the key switch is turned on. Then the control signal DE
O is output, making the gate circuit 44 conductive. Then,
Although the data is output as shown in FIG. 2(G), only the direct data is selected and supplied to the multiplexer 49. Next, when the period T has elapsed, the control signal DEI is output, and the gate circuit 45 is turned on. As a result, FIG. 3(H).

As shown in (1), the Δ-delayed data and the direct signal are supplied to four multiplexers. In this way, the gate circuits 44, 45, 46, 47 are turned on every time Δ time elapses, and the data shown in FIGS. 3(H) to 3(K) are sequentially supplied to the multiplexer 48. In this example, all data is supplied in the third cycle. From this point on, the data shown in FIG. 2(G) is output from the multiplexer 49 in a time-division manner.

In order to operate as described above, a sound effect in which an echo is heard can be obtained some time after the key switch is turned on.

Furthermore, when the key switch is turned off, the echo is configured so that it does not immediately disappear from the moment the key switch is turned off. When the key switch is turned off, first the sound caused by the direct data disappears, then the Δ delay data, 2Δ delay data, and 3Δ delay data gradually disappear in this order.

FIG. 4 shows the operation timing at that time. Time t2 is the time when the key switch is turned off, and this is detected by the time difference data output control circuit 53 via the sound generation control circuit 51.

Then, the time difference data output control circuit 53 operates as shown in FIG.
) to (F), the control signal DE
Switch O to DE3 to low level and make the corresponding gate circuit non-conductive. As a result, the echo sound gradually remains and disappears, that is, reverberant sound can be obtained.

In the above-mentioned electronic music device, the sound effect generator is integrated, and if the time difference Δ is large, the reverb effect can be obtained if the echo Δ is small. In this example, 3
Although up to Δ delayed data is created, the present invention is not limited to this, and delayed data that is further delayed may be created to change the spread effect. Moreover, the multipliers 41, 42, 43 for data compression can be created using a simple hardware pit slicer. Furthermore, if this compression value is made variable so that it can be changed arbitrarily, the degree of spatial spread of the sound can be changed. Furthermore, in this embodiment, addition is performed by time-sharing processing without using an adder for the data input of the envelope adding and D/A conversion circuit 49, contributing to hardware reduction.

The above embodiment is a circuit using PCM memory, but
In particular, it is not necessary to provide an envelope imparting circuit. However, when performing looping control, it is preferable to provide an envelope applying circuit.

FIG. 6 shows another embodiment of the invention.

This embodiment employs an envelope memory 61 instead of the periodic memory shown in FIG.
The signal is supplied to the conversion circuit 62. The lower bits of the output address of the address counter 33 are supplied to the envelope memory 51, and the higher bits are supplied to the multiplexer 38 and subscribers 35, 36, 37. The outputs of the subscribers 35, 36, and 37, Δ delay data, 2Δ delay data, and 3Δ delay data, are also supplied to the multiplexer 51.

As in the previous embodiment, the multiplexer 51 is controlled by a control signal from the time division multiplexing counter 52. Further, the envelope memory 61 receives the control signal D from the time difference data output control circuit 53 described in the previous embodiment.
Controlled by EO-DE3. Also in this embodiment, sound effects can be obtained with a simple configuration.

[Effects of the Invention] As explained above, the present invention eliminates the need to use a memory or an A/DSD/A converter as a sound effect generation circuit in a digital electronic musical tone generation device using a waveform memory. To realize electronic musical sound generation without deterioration of sound quality at a minimum cost.

[Brief explanation of drawings]

FIG. 1 is a circuit diagram showing one embodiment of the present invention, FIGS. 2 to 4 are timing charts shown to explain the operation of the circuit in FIG. 1, and FIG. 5 is a PCM shown in FIG. 1. An explanatory diagram of the memory, FIG. 6 is a circuit diagram showing another embodiment of the present invention, FIGS. 7 and 8 are circuit diagrams each showing a conventional electronic musical tone generator, and FIG. 9 is a circuit diagram of FIG. 8. FIG. 10 is an explanatory diagram showing the principle of the sound effect generation circuit, FIG. 11 is a signal waveform diagram shown to explain the operation of the circuit in FIG. 10, and FIG. This figure shows an example of the volume shown in FIG. 10, and FIG. 13 is a circuit diagram showing an example of a digital delay circuit. 31... Frequency information generator, 32... Oscillator, 3
3... Address counter, 34... Looping control circuit, 35-37... Subscriber, 38... Multiplexer, 39... PCM memory, 41-43.
... Multiplier, 44-47 ... Gate circuit, 48 ...
Multiplexer, 49...envelope provision and D/
A conversion circuit, 50... Amplifier, 51... Sound generation control circuit, 52... Time division multiplexing counter, 53... Time difference data output control circuit.

Claims (1)

[Claims] In a digital electronic musical tone generation device that electronically synthesizes a sound signal by reading data from a waveform memory, in addition to the direct data that is normally read from the waveform memory, there is also a predetermined time difference. time difference address generation means for generating each address for reading out at least one time difference data having a time difference of 0.01, and time difference address generation means for generating each address for reading out at least one time difference data having time-division multiplexing control means for adding all or time-divisionally adding them and outputting them; time-difference data compression means for compressing the time-difference data output in the time-division manner at a predetermined ratio; and key-on and key-off control means for musical tones. 1. An electronic musical sound generating device comprising: time difference data output control means for maintaining the direct data and each time difference data at a predetermined time difference.