JP3799969B2 - Resonator - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a resonance apparatus that generates a musical sound signal by applying resonance processing to an input signal having a vibration waveform.
[0002]
[Prior art]
Conventionally, in a musical instrument or computer music, for example, when synthesizing the musical tone of a musical instrument having a resonance tube such as a violin or a wind instrument, a resonator that simulates the physical characteristics of the resonance drum or resonance tube. (Resonance filter) is used. Then, for example, a signal such as a string vibration waveform is input to the resonator, and the signal is resonated by the resonator to be output as a musical sound signal.
[0003]
[Problems to be solved by the invention]
The resonator as described above is typically composed of a loop circuit including a delay, and the frequency characteristic basically has a comb-like characteristic in which a plurality of resonance peaks are arranged. That is, such a resonator functions as a comb filter, and among the input signals, the frequency component corresponding to the resonance peak resonates and increases in level, and the frequency component deviating from the resonance peak attenuates. Therefore, depending on the relationship between the fundamental frequency or overtone frequency of the input string vibration waveform and the frequency of the resonance peak, or the steepness of each resonance peak, that is, the strength of resonance, the volume and tone of the tone waveform that is output change. To do. In some cases, the audible pitch (pitch) may change. For this reason, a very strong resonance effect could not be applied.
[0004]
Further, for example, a resonance algorithm or a sound source using a closed loop circuit including a wave guide technique and delay means disclosed in Japanese Patent Application Laid-Open No. 63-40199 can be exchanged between a resonator and a sound source that generates an input signal. Using an algorithm, the harmonics that have fallen into the peak (the frequency component of the resonance peak) can be attenuated to suppress level variations. However, in such a configuration, the resonators and sound sources that can be used are limited, and the versatility is reduced. There is a problem of lacking. Furthermore, even if such a physical model is used, if the sound source is a continuous sound such as a chord model (for example, a violin model), energy is always injected into the resonator, It was difficult to prevent the variation.
[0005]
It is an object of the present invention to suppress the variation in the volume of the output of the resonator and the sudden change with respect to the characteristics of the input signal and the change thereof, and to improve the versatility of the resonator.
[0006]
[Means for Solving the Problems]
According to a first aspect of the present invention, there is provided a resonance device that detects a level of an input signal in a resonance device that generates a musical sound signal by performing a resonance process on an input signal having a vibration waveform by a loop means including a delay means. Means for detecting the level of the signal after the resonance processing, and a level ratio detecting means for obtaining a ratio between the level detected by the input level detecting means and the level detected by the resonance level detecting means And correcting means for correcting the level of the signal after the resonance processing in accordance with the ratio obtained by the level ratio detecting means and outputting it as a musical sound signal.
[0007]
According to a second aspect of the present invention, there is provided a resonance apparatus having the configuration of the first aspect, wherein the correction means has a variable degree of correction.
[0008]
According to a third aspect of the present invention, there is provided a resonance apparatus having the configuration of the first aspect, wherein the correction means sets a ratio obtained by the level ratio detection means to an initial value set in advance in an initial period of the input signal. It is characterized by replacing with.
[0009]
According to a fourth aspect of the present invention, there is provided a resonance apparatus according to the first aspect, wherein the correction means sets the correction characteristic by a parameter, and the parameter is set according to a tone color of a musical tone signal. And
[0010]
According to a fifth aspect of the present invention, there is provided a resonance apparatus having the configuration according to the first, second, third, or fourth aspect, wherein the correcting means sets the ratio obtained by the level ratio detecting means in time series. Interpolating means for interpolating is provided, and the level of the signal after the resonance processing is corrected based on the ratio interpolated by the interpolating means.
[0011]
A resonance apparatus according to a sixth aspect of the present invention has the configuration according to the fifth aspect, wherein the interpolation means can set the interpolation speed separately when the ratio obtained by the level detection means becomes large and when the ratio becomes small. It is characterized by doing so.
[0012]
According to the resonance device of claim 1, since the level of the signal after the resonance processing is corrected in accordance with the ratio between the level of the input signal and the level of the signal after the resonance processing, the characteristics of the input signal and changes thereof It is possible to suppress variations in the output volume of the resonator and abrupt changes. For example, the volume can be stabilized even for a sound source that outputs a continuous sound, such as a string model that simulates the sound of a violin. That is, the resonator can be easily applied to the characteristics of the input signal and changes thereof, and the versatility of the resonator is enhanced.
[0013]
According to the resonance apparatus of the second aspect, the same effect as that of the first aspect can be obtained, and the degree of correction for correcting the level of the signal after the resonance processing is variable, so that variations of musical sounds are enriched. You can also.
[0014]
According to the resonance device of the third aspect, the same effect as that of the first aspect can be obtained, and a ratio that determines the degree of correction is set in advance in the initial period of the input signal, such as an attack portion of a musical sound. By replacing with the initial value (for example, 1), the rising edge is not lost due to the attenuated sound. In addition, a musical sound signal that reflects the level change of the attack portion as it is at the time of note-on can be output.
[0015]
According to the resonance device of the fourth aspect, the same effect as that of the first aspect can be obtained, and the correction characteristic for correcting the level of the signal after the resonance processing can be changed according to the timbre. A musical sound signal can be output.
[0016]
According to the resonance device of the fifth aspect, the same effect as that of the first, second, third, or fourth aspect is obtained, and the ratio obtained by the level ratio detecting means is interpolated in time series. Thus, since the level of the output signal is corrected based on the interpolated ratio, the degree of correction can be applied smoothly.
[0017]
According to the resonance device of the sixth aspect, the same effect as that of the fifth aspect can be obtained, and the interpolation speed can be set separately for the case where the ratio obtained by the level detection means becomes large and the case where the ratio becomes small. For example, the volume level can be corrected with a natural feeling, such as slowly decreasing the volume or increasing the volume early.
[0018]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a block diagram of the resonance apparatus according to the first embodiment of the present invention. An input signal from a sound source (not shown) is input to the resonator 1. This input signal is, for example, a signal having a vibration waveform of a string of a violin. For example, the vibration of the string is electrically detected by a pickup embedded in a bridge portion of an electric violin or the like, and the peak value is sampled. It is a digital signal composed of a time-series data sequence quantized.
[0019]
The resonator 1 includes an adder 11a, a delay (delay circuit) 11b, an all-pass filter (APF) 11c, a low-pass filter (LPF) 11d and a coefficient multiplier 11e, an adder 12a, and a delay 12b. The second loop circuit 12 includes an all-pass filter 12c, a low-pass filter 12d, and a coefficient multiplier 12e.
[0020]
Each delay time constant that specifies the delay time of the delays 11b and 12b, each filter coefficient that specifies the filter characteristics of the all-pass filters 11c and 12c, each filter coefficient that specifies the filter characteristics of the low-pass filters 11d and 12d, and coefficient multiplication Various parameters such as multiplication coefficients for specifying feedback gains of the resonators 11e and 12e are set by a control circuit (not shown), and the resonator 1 constitutes a comb filter by a two-stage delay loop. .
[0021]
An input signal from a sound source or the like is fed back to the adder 11a via the adder 11a, delay 11b, all-pass filter 11c, low-pass filter 11d, and coefficient multiplier 11e in the first loop circuit 11, and this adder 11a The input signal and the feedback signal are added to form an output signal to the second loop circuit 12. Similarly, in the second loop circuit 12, the signal is fed back to the adder 12a via the adder 12a, the delay 12b, the all-pass filter 12c, the low-pass filter 12d, and the coefficient multiplier 12e. The signals are added and output as an output signal of the resonator 1 to a coefficient multiplier 6 and a constant multiplier 7 constituting correction means described later.
[0022]
Here, the frequency characteristic of the resonator 1 is a comb-shaped output waveform having a peak (resonance peak) in the harmonic component according to the delay time of the delay, and the shape of the peak envelope according to the characteristics of the low-pass filter and the all-pass filter. Is decided. The various parameters that determine the delay time and filter characteristics are appropriately set according to the pitch of the tone signal (and input signal) to be synthesized. It is difficult to always set the frequency characteristics of the resonator 1 to be suitable for the input signal, such as when changing the signal, and the specific frequency component is emphasized by resonance, resulting in a variation in volume. Sometimes. Therefore, the output level is corrected as follows to reduce the variation in volume and output as a musical sound signal.
[0023]
The input signal is input to a first envelope follower (EF) 2 as input level detecting means. The envelope follower 2 detects the envelope of the input signal, and the envelope of the input signal is input to the multiplier 5. The output signal of the resonator 1 is input to a second envelope follower 3 as output level detecting means, and the envelope follower 3 detects the envelope of the output signal. The envelope of the output signal is input to the inverse table 4, converted to the inverse of the envelope of the output signal, and input to the multiplier 5. Then, the multiplier 5 multiplies the envelope of the input signal and the inverse of the envelope of the output signal, and the multiplication result is output to the coefficient multiplier 6.
[0024]
Here, in this embodiment, the level of the input signal is lowered so as not to overflow in the resonator 1, and the envelope followers 2 and 3 change the amplitude level from the maximum output of the resonator 1 to the output zero “1 to 0”. ”Are detected and output as outputs Ld and Lw, respectively. In addition, in the reciprocal table 4, the relationship between the input x and the output y is
y = 1 / (256 · x)
When the input x (output Lw of the envelope follower 3) is 1/256 or less, x = 1/256 is treated.
[0025]
With the above configuration, the multiplier 5
r = Ld / (256 · Lw)
The ratio r is multiplied by the output signal of the resonator 1 as a multiplication coefficient in the coefficient multiplier 6. The constant multiplier 7 multiplies the constant “256”, and the output of the constant multiplier 7 becomes the same level as the input signal.
[0026]
FIG. 2 is a diagram showing an embodiment of an envelope follower. As shown in FIG. 2A, the envelope followers 2 and 3 are full-wave rectified by a full-wave rectifier circuit 10 and an envelope is obtained by a low-pass filter 20. It can be configured to detect. Further, as shown in FIG. 2B, full-wave rectification is performed by the full-wave rectification circuit 10 and input to the low-pass filter 30. In the low-pass filter 30, the sign of the envelope change (difference) is detected by the sign detector 30a, and the multiplication coefficient of the coefficient multiplier 30b is switched between the attack rate ar and the release rate rr in accordance with this sign. Thereby, it can detect as an envelope which has an attack part and a release part.
[0027]
FIG. 3 is a block diagram of the second embodiment of the resonance apparatus. The resonator 1, the envelope followers 2 and 3, the reciprocal table 4, the multiplier 5, the coefficient multiplier 6 and the constant multiplier 7 are the first embodiment of FIG. In the following description, the same elements are denoted by the same reference numerals, and detailed description thereof is omitted.
[0028]
In the second embodiment, the ratio r output from the multiplier 5 is interpolated by a low-pass filter type interpolator 40, and the ratio r ′ interpolated by the interpolator 40 is converted by the coefficient multiplier 6 into the resonator 1. The output signal is multiplied. Since there may be a time delay in the signal passing through the resonator 1, the level ratio becomes maximum (theoretically infinite) immediately after the input signal is input, and suddenly the sound is output from the maximum volume. When the ratio is interpolated as in the second embodiment, the ratio is smoothed to some extent by the interpolation time it set in the multiplier 40a, and it is possible to prevent the sound from being produced from the maximum volume.
[0029]
FIG. 4 is a block diagram of a third embodiment of the resonance apparatus. In this third embodiment, the ratio r is interpolated by the interpolator 50. In this interpolator 50, the sign of the change amount (difference) of the ratio r is detected by the sign detector 50a, and the multiplication coefficient of the coefficient multiplier 50b is switched between the expansion time et and the suppression time st in accordance with this sign. Thereby, the interpolator 50 switches the interpolation speed according to the increasing direction and decreasing direction of the ratio r.
[0030]
That is, when the sign of the change amount of the ratio r is positive, the ratio r is in an increasing state, the level of the output signal of the resonator 1 is decreasing with respect to the level of the input signal, and the coefficient multiplier 6 and the constant multiplier 7 correct the level so as to increase. On the other hand, when the sign of the amount of change of the ratio r is negative, the ratio r is in a decreasing state, the level of the output signal of the resonator 1 is increasing with respect to the level of the input signal, and the coefficient multiplier 6 and the constant multiplier 7 make corrections so as to reduce the level. By switching the interpolation speed of the interpolator 50, the sound volume is lowered slowly and the sound volume is raised (returned) earlier. Thereby, a natural volume change is obtained.
[0031]
FIG. 5 is a block diagram of a fourth embodiment of the resonance apparatus. In the fourth embodiment, the ratio r is interpolated by a low-pass filter type interpolator 60. The interpolator 60 is configured to switch the output to a predetermined value “1/256” when the note-on signal becomes “1”. As a result, the beginning of the musical sound is generated without correcting the volume level, and thereafter the volume level is corrected with a predetermined time constant according to the interpolation time it set in the multiplier 60a. In this way, a natural tone is obtained without a lack of rise due to a decaying sound or the like. In addition, it is possible to output a musical sound signal that directly reflects the change in level of the attack section immediately after note-on. In this embodiment, a low-pass filter type interpolator is used, but it may be applied to the type of interpolator shown in FIG.
[0032]
FIG. 6 is a block diagram of a fifth embodiment of the resonance apparatus. In the fifth embodiment, the degree of volume level correction can be changed. The output of the interpolator 70 is multiplied by the coefficient agc by the coefficient multiplier 80a, the predetermined value “1/256” is multiplied by the coefficient agc by the coefficient multiplier 80b, and the output of the coefficient multiplier 80a is multiplied by the predetermined value “1/2”. 1/256 "is added and the output of the coefficient multiplier 80b is subtracted. Then, the value of the coefficient agc is made variable in the range of “0 to 1”. Thus, by changing the value of the coefficient agc within the range of “0 to 1”, the output of the adder 80c becomes “1/256” when agc = 0, and no correction is made, and when agc = 1. In the adder 80c, the predetermined value “1/256” is canceled and the output becomes the output of the interpolator 70 itself, and 100% correction is effective. Moreover, the effect of the correction can be changed continuously between this non-correction and 100%.
[0033]
FIG. 7 is a block diagram of a sixth embodiment of the resonance apparatus. In the sixth embodiment, the output Ld of the envelope follower 2 is logarithmically converted by the log conversion circuit 90a to log (Ld), and the output Lw of the envelope follower 3 is logarithmically converted by the log conversion circuit 90b to log (Ld). To do. Log (Ld) is subtracted from log (Ld) by adder 90c, and interpolation is performed by interpolator 90d as log (Ld / Lw). The output of the interpolator 90d is multiplied by a coefficient agc by a coefficient multiplier 90e to obtain agc · log (Ld / Lw), which is input to the exponent conversion circuit 90f.
[0034]
In the exponent conversion circuit 90f, the relationship between the input x and the output y is
y = (1/256) · exp (x)
Is set to be
agc · log (Ld / Lw) = log (Ld / Lw) ^agc
For the input of
(1/256) ・ (Ld / Lw) ^agc
Is output to the coefficient multiplier 6. Thereby, the volume level is corrected. In this embodiment, the degree of correction can be controlled by the coefficient agc.
[0035]
In each of the embodiments described above, a value obtained by multiplying the constant of “1/256” is obtained as the ratio between the level of the input signal and the level of the output signal of the resonator. Even if such a constant is multiplied, it corresponds to obtaining the level ratio.
[0036]
2B, attack rate ar and release rate rr, interpolation time it in FIG. 3, expansion time et and suppression time st in FIG. 4, predetermined value “1/256” and interpolation time it in FIG. Parameters relating to correction such as the predetermined value “1/256” and the coefficient agc in FIG. 6 and the coefficient agc in FIG. 7 may be set for each tone color.
[0037]
Further, in the embodiment, a comb filter resonator is used, but as a resonator, a filter such as an equalizer and a waveguide network as disclosed in JP-A-63-40199 are also used. It can be used.
[0038]
The resonator may include means for appropriately mixing the sound that has passed through the resonance and the direct sound, or a means for mixing the resonant sound after the level correction and the direct sound is provided in the subsequent stage. May be.
[0039]
Further, for the level detection, an effective value may be detected instead of a peak detection like an envelope follower.
[0040]
Further, when using a DSP, CPU, or the like provided with a divider, it is also possible to perform the calculation as it is only by considering level alignment. Further, if floating point arithmetic is possible, the level correction range can be made wider.
[0041]
In addition, when the resonator delay is large and the value changes greatly until the sound is generated even if the initial value of the ratio is given, the initial value of the ratio may be held for a certain period of time.
[0042]
Further, the input signal is not limited to the waveform signal of a bowed instrument such as a violin, and is particularly suitable for generating a continuous sound, and may be any input signal that is intended for resonance processing.
[0043]
In addition, the present invention functions as various devices such as various electronic musical instruments, tone generators, sequencers, effectors, and systems that connect these devices using communication means such as MIDI or various networks. Alternatively, it may be incorporated as an application. Further, it may be realized by a personal computer and software.
[0044]
Further, the present invention may be widely applied to a resonance apparatus as a method and method for simulating resonance characteristics of a sound field space or some acoustic medium, in addition to a resonance drum or a resonance tube of a musical instrument.
[0045]
【The invention's effect】
According to the resonance device of the first aspect, it is possible to suppress variations in the volume of the output of the resonator and a sudden change with respect to the characteristics of the input signal and the change thereof, for example, a string model that simulates the sound of a violin. Thus, the sound volume can be stabilized even for a sound source that outputs a continuous sound. Therefore, the resonator can be easily applied to the characteristics of the input signal and changes thereof, and the versatility of the resonator is enhanced.
[0046]
According to the resonance device of the second aspect, the same effect as that of the first aspect can be obtained, and since the degree of correction for correcting the level of the output signal is variable, the variation of the musical sound can be enriched.
[0047]
According to the resonance device of claim 3, the same effect as that of claim 1 can be obtained, and it is possible to prevent a rise from being lost due to a decaying sound or the like, and to output a musical sound signal that directly reflects the level change of the attack part. Can do.
[0048]
According to the resonance device of the fourth aspect, the same effect as that of the first aspect can be obtained, and the correction characteristic for correcting the output signal can be changed according to the tone color, so that a tone signal corresponding to the tone color can be output. it can.
[0049]
According to the resonance device of the fifth aspect, the same effect as that of the first, second, third, or fourth aspect can be obtained, and the degree of correction can be smoothly applied.
[0050]
According to the resonance device of the sixth aspect, the same effect as that of the fifth aspect can be obtained, and the volume level can be corrected with a natural feeling, for example, the volume is slowly decreased or the volume is increased early. Can do.
[Brief description of the drawings]
FIG. 1 is a block diagram of a resonance apparatus according to a first embodiment of the present invention.
FIG. 2 is a diagram showing an example of an envelope follower in each embodiment of the present invention.
FIG. 3 is a block diagram of a resonance apparatus according to a second embodiment of the present invention.
FIG. 4 is a block diagram of a resonance apparatus according to a third embodiment of the present invention.
FIG. 5 is a block diagram of a resonance apparatus according to a fourth embodiment of the present invention.
FIG. 6 is a block diagram of a resonance apparatus according to a fifth embodiment of the present invention.
FIG. 7 is a block diagram of a resonance apparatus according to a sixth embodiment of the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Resonator, 2, 3 ... Envelope follower, 4 ... Reciprocal table, 5 ... Multiplier, 6 ... Coefficient multiplier, 7 ... Constant multiplier, 10 ... Full wave rectifier, 20, 30 ... Low pass filter, 40, 50 , 60, 70, 90d ... interpolator

Claims (6)

In a resonance apparatus that generates a musical sound signal by applying a resonance process to an input signal of a vibration waveform by a loop means including a delay means ,
Input level detection means for detecting the level of the input signal;
Resonance level detection means for detecting the level of the signal after the resonance processing;
Level ratio detection means for obtaining a ratio between the level detected by the input level detection means and the level detected by the resonance level detection means;
Correction means for correcting the level of the signal after the resonance processing according to the ratio obtained by the level ratio detection means and outputting it as a musical sound signal;
A resonance apparatus comprising: The resonance apparatus according to claim 1, wherein the correction means has a variable degree of correction. 2. The resonance apparatus according to claim 1, wherein the correction unit replaces the ratio obtained by the level ratio detection unit with a preset initial value in an initial period of the input signal. 2. The resonance apparatus according to claim 1, wherein the correction means sets the correction characteristic by a parameter, and the parameter is set according to a tone color of a musical tone signal. The correction means includes an interpolation means for interpolating the ratio obtained by the level ratio detection means in time series, and corrects the level of the signal after the resonance processing based on the ratio interpolated by the interpolation means. The resonance apparatus according to claim 1, claim 2, claim 3, or claim 4. 6. The resonance apparatus according to claim 5, wherein the interpolation means can set the interpolation speed separately when the ratio obtained by the level detection means increases and when the ratio decreases.

Images