JPH0456897A - Electronic musical instrument - Google Patents

Abstract

PURPOSE:To accurately apply a pronunciation instruction or a muting instruction by generating the pronunciation instruction when the absolute value of a peak value is more than a previously determined 1st threshold and a changing state is changed to a previously determined 2nd threshold or more. CONSTITUTION:When a peak detecting means 2 successively detects peak values and a change detecting means 4 successively detects the changing states of respective peak values, e.g. the rate of changes of respective peak values, the changing states are inputted to a pronunciation instructing means 6. On the other hand, an input signal is also inputted to the means 6, which outputs a pronunciation instruction only when the changing state, i.e. the rate of change, exceeds the previously determined 2nd threshold in the state of the peak value more than the 1st threshold. Thereby, even if a noise is superposed to the peak value or the peak value is changed by the change of a higher harmonic phase, a pronunciation instruction can be accurately applied.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an electronic musical instrument, and particularly to one that controls a sound source section in response to an input waveform signal.

[Prior Art] As an example of the above-mentioned electronic musical instrument, there is a kitar synthesizer, which controls a sound source section to generate a musical tone when a waveform signal is generated, and mutes the sound when the waveform signal is attenuated. It is necessary to control the sound source section and also control the envelope of the musical tone signal generated from the sound source section in accordance with the envelope of the waveform signal.

Conventionally, techniques for giving a sound generation instruction include sequentially detecting positive peak values and negative peak values of a waveform signal, and giving a sound generation instruction when both of them exceed a predetermined threshold. 100598), or when either the positive or negative level of the waveform signal exceeds a predetermined value, for example, if the positive side exceeds the predetermined value, the time interval between adjacent peak values on the positive side is measured. , which instructs the generation of a musical tone with a frequency corresponding to this time interval (Japanese Patent Laid-Open No. 63-24
No. 3998). Furthermore, as a technique for giving a mute instruction, there is a technique that detects the positive peak value and negative peak value of an input waveform signal, and gives a mute instruction when both of them are smaller than a predetermined value (Japanese Patent Application Laid-Open No. 63-1989-1). No. 243997), and others that give a mute instruction when the input waveform signal remains below a predetermined level for a predetermined period of time (Japanese Patent Laid-Open No. 100597/1999). In addition, the technique for detecting the envelope is to find the maximum peak value at the rising edge of the input waveform signal, sequentially find the peak values that occur after that, and each time this peak value is found, the average value with the previous peak value is calculated. is calculated, and the ratio between this average value and the maximum peak value is sequentially calculated. However, each time the above peak value is calculated, the ratio between this value and the maximum peak value is calculated (Japanese Patent Application Laid-Open No. 2-7096 ).

[Problem to be solved by the invention] However, the above-mentioned sound generation instruction and mute instruction techniques do not cause fluctuations in the peak value due to changes in the phase of harmonic components of the input waveform signal, or noise or noise superimposed on the input waveform signal. There was a problem in that it was not possible to accurately give instructions to produce sound or mute the sound due to fluctuations in the peak value caused by the noise. Further, the above-mentioned envelope detection technique requires complicated calculations, so there is a problem in that the processing time becomes long.

An object of the present invention is to provide an electronic musical instrument that solves each of the above-mentioned problems.

[Means for Solving the Problem] As shown in FIG. change detection means 4 for detecting a state of change between the peak value and the previously detected peak value;
They have these in common.

The first invention includes a sound generation instruction means 6 that generates a sound generation instruction when the absolute value of the peak value is equal to or higher than a predetermined first threshold value and the above-mentioned change state changes to a predetermined second threshold value or higher. ing.

The second invention includes a mute instruction means 8 which generates a mute instruction when each successively detected change state indicates a tendency of attenuation of the input signal.

The third invention is to output the current peak value as envelope information of the input signal when the change state indicates that the peak value detected this time is greater than the peak value detected last time. , comprising an envelope information generating means IO for outputting the previous peak value as the envelope information when the change state indicates that the currently detected peak value is smaller than the previously detected peak value. ing.

[Operation] According to the first invention, the peak detection means 2 sequentially detects the peak values as shown in FIG. 4, and the change detection means 4 sequentially detects the state of change of each peak value, for example, the rate of change of each peak value. Detect. This changing state is input to the sound generation instruction means 6. −
The universal signal is also input to the pronunciation instruction means, and the pronunciation instruction means l
O issues a pronunciation instruction when the change state, for example, the rate of change, exceeds a predetermined second threshold in a state where the peak value is above the first threshold as shown by the dotted line in FIG. Let's do it.

In the second invention, each peak value is sequentially detected by the peak detection means 2 as shown in FIG. Ru. The mute instruction means 8 issues a mute instruction when this change state indicates a decreasing tendency of the input signal.

In the third invention, peak values are sequentially detected as shown in FIG.
0. This envelope information generating means 10 generates, for example, h103(-23) and h102(-1) in FIG.
7), if the current peak value h103 is larger than the previous peak value h102, the current peak value h103 is output as the current envelope information as shown in FIG. h104 (=19) and the previous peak value h103 (=23), if the current peak value h104 is smaller than the previous peak value h103, as shown in the same figure (C), the previous peak value The peak value is set as the current peak value.

[Embodiment] In FIG. 2, reference numeral 21 denotes an input amplifier, which amplifies an input waveform signal generated by, for example, plucking the strings of a guitar synthesizer or an input waveform signal input from a microphone, and is constituted by, for example, an operational amplifier. ing. The output of this input amplifier 21 is supplied to a low-pass filter 22, where harmonic components of the input waveform signal are removed. That is, the fundamental wave of the input waveform signal is generated at the output of the low-pass filter 22. The fundamental wave of this input waveform signal is
The data is input to the zero-cross converter 23, and the converter 23 supplies zero-cross data to the external timer terminal 28a of the microcomputer 28 every time the zero-cross point of the fundamental wave is detected.

On the other hand, the fundamental wave of the input waveform signal is also supplied to the rectifier circuit 25, and its positive polarity portion is half-wave rectified by the rectifier circuit 25. In addition, the fundamental wave of the input waveform signal is
4 to the rectifier circuit 26, and its positive polarity portion (negative polarity portion of the original fundamental wave) is subjected to half-wave rectification. The outputs of these rectifying circuits 25 and 26 are added by an adder 33. Therefore, the output of this adder 33 is a full-wave rectified version of the input waveform signal. The output of this adder 33 is supplied to a peak hold circuit 27, and the output of this peak hold circuit 27 is
It is supplied to the D converter input terminal 28c. Also,
This peak hold circuit 27 has a peak hold reset input terminal 27a for resetting the held peak value, and this is connected to a peak hold reset output terminal 28b of the microcomputer 28.

The microcomputer 28 includes a central processing unit, a working memory, an A/D converter, etc., and outputs a reset signal to a peak hold reset output terminal 28b every time zero cross data is input to an external timer input terminal 28a. . Therefore, the peak hold circuit 27
is reset every time the fundamental wave of the input waveform signal crosses zero, and holds the peak value every half period of the fundamental wave of each input waveform signal. Then, these peak values are sent to the A/D converter in the microcomputer 28.
Therefore, it is sequentially digitized and digital peak information h
(i).

The microcomputer 28 processes the digital peak information h(i) as described later, and outputs a sound generation instruction (gate on) signal, a mute instruction (gate off signal) and envelope information as MIDI signals to the MIDI output terminal 28.
Output from d. This MIDI signal is supplied to a sound source section (not shown) via a MIDI output circuit 29.

A MIDI input circuit 30 receives MIDI signals from other electronic musical instruments (not shown) and supplies them to the microcomputer 28. The microcomputer 28 is
The supplied MIDI signal is supplied to the sound source section via the MIDI output circuit 29. Reference numeral 32 denotes a group of operators, which is used to set parameter information to be supplied to the sound source unit in the microcomputer 28 via the operator information input terminal 28g, including information other than gate-on, gate-off, and envelope information. The parameter information thus obtained is also supplied as a MIDI signal from the microcomputer 28 to the sound source section via the MIDI output circuit 29. A display circuit 31 receives information set by the operator group 32 from a display information output terminal 28f of the microcomputer 28 and displays it.

Next, we will explain how the microcomputer 28 processes the digital peak information h(i) to generate gate-on signals, gate-off signals, and envelope information. The various registers set in the memory will be explained with reference to FIG. Register HI is a register that stores digital peak information h(i) when it is input, and register HN is a register that stores digital peak information h(i) when it is stored in register HI. This register stores digital peak information stored in HI, that is, h(i-1). register)
IM is a register that stores the larger of the values stored in registers H1 and HM. register RI
is a register that stores a change coefficient of digital peak information calculated as described later based on the values stored in registers H1, HM, and HN, in order to be used to determine whether or not to generate a gate-on signal. . The register HL is a register that stores a threshold value H(Iim) that is compared with the digital peak information h(i), as shown by the dotted line in FIG. 4, in order to determine whether or not to generate a gate-on signal. . Register RL is a register that stores a threshold value R (Iis) that is compared with the variation coefficient stored in register RI to determine whether or not to generate a gate-on signal. Register SI is a register that stores the integrated value of the change coefficients for use in determining whether or not to generate a gate-off signal. The register SL is a register that stores a threshold value S (Ii) which is compared with the value stored in the register SH to determine whether or not to generate a gate-off signal.

Processing using these registers is carried out by the zero-cross converter 23 or by generating zero-cross data.
This is performed by reading the interwoven routine shown in the figure. That is, in this interwoven routine, first, step S is performed to determine whether or not Kate is currently on.
l), and if the answer is No, gate-on detection processing is performed (step S2), and this interlaced routine is ended. If the answer to step Sl is YES, envelope detection processing is performed (step S3), gate-off detection processing is performed (step S4), and this interwoven routine is ended.

In the gate-on detection process, as shown in FIG.
-1) is transferred to register HN (step S5).

Then, the peak value of the peak hold circuit 27, which is newly holding the peak value of the input waveform signal by being reset in response to the generation of zero cross data, is digitized into digital peak information h(i) or a register. HI (step S6), then register H
1. The stored values of HM are compared (step S7), and if the stored value of register HN is greater than or equal to the stored value of register HI, the stored value of register HN is transferred to register HM (step S8), and the stored value of register HI is If the stored value is larger than the stored value in register HN, the stored value in register H1 is transferred to register HM (step S9). The transfer in step S8.9 is a preliminary preparation for calculation of a change coefficient, which will be described later.

Then, it is determined whether the value stored in the register HI is greater than the threshold value 1 (Iim) stored in the register HL (step 510), and if the answer is NO, this routine is ended. Therefore, for example, in the case of an input waveform signal as shown in FIG.
Gate-on processing is not performed until 3. In addition, in FIG. 4, the threshold value H (Iim) is set to 20.

Next, when the answer to step SIO is YES, that is, when digital information h(1:l) obtained by digitizing the peak value of h13 in FIG. 4 is supplied, the change coefficient R(i ) is calculated (step 5ll).

This calculation is performed by R(i)=(HI-HN)/HM. That is, the difference between the digital peak information h(i) input this time and the digital peak information h(i-1) input last time is determined by calculating the difference between h(i) and h(i-1), whichever is the larger value. The value obtained by dividing the value by 1 indicates the degree of change in the digital peak information.

Then, it is determined whether the coefficient R(i) is larger than the threshold value R(lie) stored in the register RT (step S
12). If the answer is No, this routine is terminated, and if the answer is YES, gate-on processing is performed,
A gate-on signal is generated (step 513). Therefore, in the case of FIG. 4, if the threshold value R(I is) is 0.25, R(i) at h13 is (25-20)/25=
0.2, which is smaller than the threshold R(Iim), so gate-on processing is not performed, but R(1) at h14
is (35-25)/35=0.28, and the threshold R(I
im), gate-on processing is performed.

The envelope detection process is performed as shown in FIG. 9, and in steps S5a, 6a, 7a, 8a, and 9a, which are the same as step S5.6.7.8.9 in FIG. is transferred to register HN, new digital peak information h(i) is stored in register HI, and the one with a larger value among registers H1 and HN is transferred to register HM. Then, the value stored in the register HM is processed as envelope information (step 514), and this routine ends. Therefore, for example, in the case of an input waveform signal as shown in Figure 5(a), if each peak value is used as envelope information, it will have a very large change as shown in Figure 5(b), but this envelope According to the detection process, smooth envelope information can be obtained as shown in FIG. That is, as shown in the figure, the peak values are hlol (-12) and h102 (-17).
, h103(-23), h104(-19)...
In the case of ・, the peak values of hlol, h102, and h103 are larger than the previous peak values, so the envelope information is hlol, h102, as shown in the same figure (c).
, the peak value of h103 is used as is, but h
Since the peak value of 104 is smaller than the previous peak value h103, the previous peak value h103 is used as it is without using h104 as the envelope information. Since similar processing is performed thereafter, smooth envelope information can be obtained.

The gate-off detection process is performed as shown in FIG. 10, and steps S5b, 6b, 7b, 8b, and 9b, which are the same as steps S5, 6, 7, and 8.9 shown in FIG. The stored value of H1 is transferred to register HN, new digital peak information h(i) is stored in register HI, and the one with a larger value among registers H1 and HN is transferred to register HM.

If the current peak value is equal to or smaller than the previous peak value, that is, if the stored value of the register HN is transferred to the register HM, then step S8b is followed by the same step as step Sll in FIG. 5llb is executed to calculate the change coefficient R(i).

However, in this case, the change coefficients R(i) are all negative values. Next, using this R(i), attenuation integration coefficient 5(i
) is calculated based on the following formula (step 515).

5(i) = drx 5(i-1) -R(i) Here, dr is, if the same value of national pic information continues, the value of R(i) becomes 0, and 5(i) To prevent this, the value is determined experimentally, for example, 0.88.
The value can be within the range of 0.99 to 0.99.

Further, if the current peak value is larger than the previous peak value, that is, if the stored value of the register Hl is transferred to the register HM, the value of 5(i) is set to O following step S9b. 516). This is because the fact that the current peak value is larger than the previous peak value indicates that the input waveform signal is increasing, and if this is accumulated as 5(i), gate-off will not be detected promptly.
This is because (i) is calculated only when the input waveform signal shows an attenuation tendency.

Following these steps 316 or 15, it is determined whether 5(i) is larger than the threshold value 5(Ii) set in the register SL (step 517). If the answer is NO, this routine ends; if the answer is YES, gate-off processing is performed and a gate-off signal is generated (step 518).

For example, if the input waveform signal is as shown in Fig. 6, and the dr is set to 0.97 and the threshold value 5 (lie) is set to 1.5, each peak value h201, 202, etc. Assuming the value shown in , h202 (-61) is larger than the previous peak value h201 (=SS), so 5
(i) is O, and the change coefficient R(i) at h20: ] (=57) is -0,0656 [-(57-61)/6
11, and 5(i) becomes 0.0656.

At h204(-57), the change coefficient R(i) is O, and O is subtracted from the value 0.0636 obtained by multiplying the previous 5(i) value 0.0656 by 0.97, and 5(i) is 0.
It becomes 0636. Thereafter, 5(i) is calculated sequentially in the same way, and in this example, when h214(-5), 5(i) exceeds the threshold S(Iim) of 1,5, and a gate off signal is generated. .

As described above, in this embodiment, a gate-on signal, a gate-off signal, and envelope information are generated, and these are supplied to the sound source section, and the sound generation of musical tones is started by the gate-on signal, and the envelope information of the generated musical tones is generated by the envelope information. is controlled, and the tone is muted by the gate-off signal.

In the above embodiment, the change coefficient R(i) is used in the detection process of Kate-on and Kate-off, but instead of this, the previous peak value stored in the register HN and the register H
1 may be calculated and used. Of course, in this case, in the gate-on detection process, the calculated ratio is compared with the threshold R (li), and in the gate-off detection process, the calculated ratio is integrated, and this integrated value is compared with the threshold 5 (li). . In this case, if the current peak value is larger than the previous peak value, as in the above embodiment, the integrated value may be set to 0, or this integrated value may be
may be multiplied by In addition, in the case of gate-off detection, the current peak value stored in register HI is compared with the previous peak value stored in register HN, and if the current peak value is small, a separately provided counter is The gate-off signal may be generated when the count value reaches a preset value. In this case, as in the above example, if the current peak value is larger than the previous peak value,
You may also reset the counter.

[Effects of the Invention] As described above, according to the invention according to claims 1 to 3 of the present invention, the absolute value of the peak value is larger than the predetermined threshold, and the current peak value and the previous peak value are different from each other. Since the configuration is configured to generate a sound generation instruction when the change state or the change exceeds a predetermined threshold value, it is possible that noise is superimposed on the peak value or that the peak value is changed due to a change in the phase of the harmonic. Even if the pronunciation changes, it is possible to give accurate pronunciation instructions.

Further, according to the invention described in claims 4 to 7, when the peak value is changing or the input waveform signal is showing a decreasing tendency,
Since a mute instruction is given, it is possible to reliably give a mute instruction.

According to the invention as set forth in claim 8, when the current peak value is larger than the previous peak value, the current peak value is output as envelope information, and when the current peak value is smaller than the previous peak value, the current peak value is output as envelope information. Since the previous peak value is output as the current envelope information, there is no need for complicated function calculations, and envelope information can be output quickly and smooth envelope information.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an electronic musical instrument according to the present invention, FIG. 2 is a block diagram of an embodiment of the present invention, and FIG. 3 is a diagram showing a part of the working memory of a microcomputer used in the embodiment. FIG. 4 is a diagram showing analog waveform data at the time of gate-on detection in the same embodiment, and FIG. 5 is a diagram showing analog waveform data at the time of envelope curve detection in the same embodiment.
FIG. 6 is a diagram showing analog waveform data at the time of gate-off detection in the same embodiment, FIG. 7 is a flowchart showing the interwoven routine of the same embodiment, FIG. 8 is a flowchart of the gate-on detection processing routine, and FIG. FIG. 10 is a flowchart of the envelope detection processing routine. FIG. 10 is a flowchart of the gate-off detection processing routine. 2...Peak detection means, 4...Change detection means,
6... Pronunciation instruction means, 8... Mute instruction means, 1
o...Envelope information generation means. Patent applicant: Roland Co., Ltd. Representative: Tetsu Shimizu, and 2 others
1 Figure vJ21] 3rd! l] No. 10

Claims (8)

[Claims] (1) peak value detection means for sequentially detecting peak values of an input signal; change detection means for detecting a change state between the peak value and the previously detected peak value each time the peak value is detected; The absolute value of the above peak value is the predetermined first
and a sound generation instruction means for generating a sound generation instruction when the change state changes to a second threshold value or more. (2) In the electronic musical instrument according to claim 1, the change detection means divides the difference between the absolute values of the previous and current peak values by the larger of the absolute values of the previous and current peak values. An electronic musical instrument characterized by calculating a rate of change. (3) The electronic musical instrument according to claim 1, wherein the change detection means calculates a ratio of the absolute values of the previous and current peak values. (4) peak value detection means for sequentially detecting peak values of an input signal; and change detection means for detecting a state of change between the peak value and the previously detected peak value each time the peak value is detected; an electronic musical instrument comprising mute instruction means for generating a mute instruction when each of the sequentially detected change states indicates a tendency of attenuation of the input signal. (5) In the electronic musical instrument according to claim 4, the change detection means divides the difference between the absolute values of the previous and current peak values by the larger one of the absolute values of the previous and current peak values. , the muffling instruction means includes means for accumulating the change rate, and comparison means for generating a muffling instruction when the accumulating capacity exceeds a predetermined threshold. An electronic musical instrument characterized by: (6) The electronic musical instrument according to claim 4, wherein the change detection means is means for calculating a ratio of the absolute values of the previous and current peak values, and the mute instruction means is means for integrating the ratios. 1. An electronic musical instrument comprising: and a comparison means for giving a sound generation instruction when the integrated value is greater than or equal to a predetermined threshold. (7) The electronic musical instrument according to claim 4, wherein the change detecting means comprises a comparing means for generating an output signal when the absolute value of the current peak value is smaller than the absolute value of the previous peak value, and the muting instructing means counts the number of occurrences of the above output,
An electronic musical instrument comprising means for generating a mute instruction when the counted value reaches a predetermined value. (8) peak value detection means for sequentially detecting peak values of the input signal; and change detection means for detecting a change state between the peak value and the previously detected peak value each time the peak value is detected; When the above change state indicates that the currently detected peak value has increased compared to the previously detected peak value, the current peak value is output as the envelope information of the input signal, and the currently detected peak value is envelope information generating means for outputting the previous peak value as the envelope information when the change state indicates that the peak value has decreased from the previously detected peak value.