JP3152210B2 - Music synthesizer - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tone synthesis device using a so-called delayed feedback tone synthesis algorithm for inputting an initial waveform signal to a closed loop including delay means and performing a cyclic processing in the closed loop to synthesize a tone. About.

[0002]

2. Description of the Related Art Conventionally, as a so-called digital sound source used for an electronic musical instrument or the like, a tone is produced by inputting data such as initial waveform data, impulse signal data and nonlinear signal data into a closed loop including a delay circuit and performing regression calculation processing. Tone synthesizers that use a so-called delayed feedback type attenuation sound synthesis algorithm are known (Japanese Patent Publication Nos. 58-58679 and 58-48109).
No., JP-B-59-7396, JP-A-62-14309.
7, JP-A-63-80300, JP-A-63-401
No. 99).

[0003] In these tone synthesizers, a mechanical vibration system of a natural musical instrument such as a wind instrument tube or a string instrument string is physically approximated by an electric circuit. That is, it corresponds to the pitch (pitch) of the musical tone to be pronounced.

In such a tone synthesizer, generally, a waveform memory storing sequentially sampled data for one wavelength of a desired initial waveform is used as initial waveform data input means, and a shift circuit is used as a delay circuit. A register, a BBD, a RAM, and the like are used. The pitch is set by changing the cycle of the clock for driving the delay circuit. At this time, the clock for reading the waveform memory is also variably set. That is, both inversely proportional to the pitch and the delay period of time and the initial waveform data of the delay circuit
(The frequency of the initial waveform data is proportional to the pitch .)

FIGS. 6A and 6B show the spectrum distribution of a musical tone synthesized by such a conventional musical tone synthesizing apparatus.
From FIGS. 7A and 7B, this musical tone is a formant sound, and the pitch and the spectral envelope are such that the higher (or lower) the pitch, the wider (narrower) the spectral envelope. You can see that they are linked. In the following, a formant in which the pitch and the spectrum envelope are linked in this way is referred to as a moving formant.

[0006] A moving formant in which the balance of the line spectrum is constant between the fundamental tone and the overtone is also useful as such. However, as a musical tone, the absolute frequency of the peaks and valleys of the spectral envelope is a constant formant sound (hereinafter referred to as a fixed formant sound) regardless of the pitch, or moves to some extent according to the pitch.
Formant sounds whose frequency spectrum does not always change in proportion to pitch (hereinafter referred to as semi-fixed formant sounds) are also important.

[0007]

SUMMARY OF THE INVENTION It is an object of the present invention to provide a tone synthesizer using a delayed feedback tone synthesis algorithm, which can generate a fixed or semi-fixed formant sound. .

[0008]

In order to achieve this object, according to the present invention, the delay time of the delay means in the delay feedback type tone synthesis algorithm is set in accordance with the pitch of the generated tone. The frequency is controlled so as not to change in proportion to the pitch.

[0009]

Function and Effect The formant sound synthesized in such a musical sound synthesizer has a pitch (fundamental tone) corresponding to the delay time of the delay means and an envelope corresponding to the initial waveform signal.

According to the present invention, the delay time of the delay means and the cycle of the initial waveform signal can be set independently.
Therefore, by changing the delay time of the delay means while keeping the cycle of the initial waveform signal constant, it is possible to variably set only the pitch without changing the envelope (fixed formant), If the period of the initial waveform signal is changed by a desired ratio of the change amount of the delay time,
It is possible to synthesize a formant sound (semi-fixed formant) in which the envelope changes less than the pitch changes. Furthermore, if the change width of the cycle of the initial waveform signal is made larger than the change width of the delay time, it is possible to synthesize a formant sound in which the envelope changes more than the pitch changes. Ma
In addition, the period of the initial waveform signal and the delay time of the delay
Variable so that it is not proportional to
Interpolation when passing the initial waveform signal can be simplified.
You.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail based on embodiments. FIG. 1 is a block diagram showing a basic configuration of a tone synthesizer using a delayed feedback tone synthesis algorithm according to an embodiment of the present invention.

The apparatus shown in FIG. 1 includes a phase accumulator 1, a delayed feedback tone synthesis algorithm 2, a counter 3, a waveform memory 4, and a control circuit 5.

The control circuit 5 sends a reset signal RS to the counter 3 in accordance with the key-on timing of a musical tone to be generated, and outputs an F number (delay time information value) ΔF corresponding to the pitch of the musical tone to the phase accumulator 1. To enter.

The counter 3 is reset by a reset signal supplied from the control circuit 5, and counts a fixed frequency system clock φ sent from a clock generator (not shown). The output of the count value of the counter 3 is given to the waveform memory 4 as an address signal.

The waveform memory 4 stores each sample point data of an initial waveform of a musical tone to be generated, and sequentially reads out the sample point data stored at the address in accordance with the address signal. The read data is input to the delayed feedback tone synthesis algorithm 2. In this case, since the counter 3 for addressing the waveform memory 4 is controlled by the system clock φ having a fixed period φ, the initial waveform is always read from the waveform memory 4 at a constant sample frequency.

Each time the system clock φ is input, the phase accumulator 1 accumulates the F number ΔF in a range where the accumulated value is “0” to “1”, and the accumulated value is 2
reaches the value "1" corresponding to π to generate a carry output C _0. As a result, for example, if the F number ΔF is set to 0.5, the control clock, that is, the period of the carry output C ₀ becomes 2φ. If the F number ΔF is set to 0.66 °, the control clock C ₀ becomes the system clock φ
The second and fourth cycles of every six cycles of
Occurs in the third, fifth and sixth cycles. Therefore, in this case, the average period of the control clock is 6 / 4φ.

The delay feedback type tone synthesis algorithm 2 includes a delay circuit (not shown), a circulating circuit for circulating the output of the delay circuit to the input side, and adding the output of the circulating circuit and the output of the waveform memory 4 to the above. It is composed of an adder and the like input to the delay circuit. In this synthesis algorithm 2, the delay circuit, the phase of the carry output C ₀ sent from the accumulator 1 acts as a control clock, wherein the control data input through the adder clock C time corresponding to ₀ Output with only a delay. Thus, the synthesis algorithm 2 is to be controlled the delay time by carry output C ₀ of the period corresponding to the F number ΔF sent from the phase accumulator 1, to set pitch (fundamental tone) free it can.

FIGS. 2A to 2C show examples of configurations that can be applied as the delayed feedback tone synthesis algorithm 2 in FIG. FIG. 2A is an attenuation sound algorithm described in Japanese Patent Publication No. 58-58679 of the present applicant. This synthesis algorithm can also be used for synthesizing a sustained sound by continuously inputting. FIG. 2B shows a VLPF-type sustained sound algorithm described in Japanese Patent Application Laid-Open No. 61-107298 by the present applicant.
(C) is a continuous sound synthesis algorithm combining the nonlinear signal generation circuit described in Japanese Patent Application No. 1-1115546 by the present applicant.

FIG. 5 shows the configuration of a conventional tone synthesizer using a delayed feedback tone synthesis algorithm.
In this type of algorithm, the pitch F
_P is given by F _P = F _C / τ. Here, F _C is the frequency of the control clock of the algorithm, and τ is the delay time in the algorithm.

[0020] The pitch F _P to variably control, it is necessary to vary the clock frequency F _C or delay time tau. In this case, a shift register or a BBD is used as a delay circuit.
When F _C is fixed, the delay time τ is changed by changing the number of stages of the shift register and the BBD, and can be changed only by an integer value corresponding to the number of stages. For this reason, it is difficult to realize each pitch of the scale constituent sounds. Further, it is impossible to control the delay time τ with high accuracy in accordance with vibrato, pitch bend, and the like. Therefore, the clock frequency F _C is freely changed in proportion to a desired pitch frequency. The counter 3 is for generating an address for reading an initial waveform from the waveform memory 4. Since this counter 3 operates on the same clock as the synthesis algorithm 2, the synthesis algorithm 2
And the synthesis algorithm 2 operate completely synchronously.

In the conventional apparatus shown in FIG. 5, the algorithm shown in FIG. 2 can be used as the synthesizing algorithm 2.
This kind of synthesis algorithm constitutes a comb filter having a peak at a frequency that is almost an integral multiple of the pitch frequency.
Therefore, when this synthesis algorithm is applied to the configuration shown in FIG. 5, the spectrum included in the input waveform is shifted upward by increasing the clock frequency to increase the pitch. As a result, in the apparatus shown in FIG. 5, the spectral structure viewed from the balance between harmonics is constant irrespective of the pitch frequency, but the absolute frequency such as the peak or valley of the spectral envelope shifts according to the pitch frequency. Become a formant.

[0022] In the apparatus of FIG. 1, since the counter 3 for addressing waveform memory 4 is controlled from the system clock φ fixed, the waveform memory 4 the waveform at the same sample frequency F _C regardless of the pitch frequency Is read. On the other hand, since the synthesis algorithm 2 is controlled via the phase accumulator 1, the basic pitch can be freely set. In this case, the carrier output of the phase accumulator 1 frequency F _d of the (control clock) C _O is, F _d = ΔF × F _C, and the basic pitch _{F P, F P = F d} / τ = ΔF × F C / τ Given by As a result, as shown in FIGS. 3A and 3B, the pass-line spectrum of the synthesis algorithm 2 becomes coarser as the pitch increases, but the spectrum structure of the envelope waveform is fixed with respect to the absolute frequency and synthesized. Music is
It has a fixed formant structure.

FIG. 4 shows a configuration of a tone synthesizer according to another embodiment of the present invention. This device adds a second phase accumulator 6 to that of FIG.
Is independent of the phase accumulators 1 and 6 by F
The configuration is such that numbers ΔF ₁ and ΔF ₂ are given, and a reset signal RS is given to the counter 3 at key-on timing. By separately providing the phase accumulator for reading the waveform memory and for the synthesis algorithm, it is possible to control the formant freely. That is, since the reading speed of the waveform can be controlled independently of the pitch on the algorithm side, the spectrum of the output musical sound waveform is, for example, a semi-fixed formant intermediate between the moving formant shown in FIG. 6 and the fixed formant shown in FIG. It is also possible to set the state.

In such a configuration, since the sampling rate of the waveform reading and the sampling rate of the algorithm are different from each other, interpolation is originally required when data is transferred. However, even if this interpolation is omitted, it is equivalent to the zero-order hold type interpolation, and a certain amount of aliasing occurs. However, this aliasing occurs when data is transferred from the waveform memory reading side to the algorithm side, so the initial waveform This is not a big problem, since the spectrum of the A is only distorted to some extent. Therefore, even when interpolation is introduced, a simple one such as linear interpolation is sufficient. When an interpolation circuit is provided, it is inserted between the waveform memory 4 of FIG.

In the above description, the embodiment in the digital system is shown, but the present invention can be applied to the analog system. In this case, it is preferable that the phase accumulator 1 and the counter 3 be digital processing systems and that only the arithmetic circuit be an analog processing system.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a basic configuration of a musical sound synthesizer according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a specific configuration example of a delay feedback algorithm in FIG.

FIG. 3 is an explanatory diagram of a spectrum distribution of a musical tone synthesized by the apparatus of FIG. 1;

FIG. 4 is a block diagram showing a basic configuration of a musical sound synthesizer according to another embodiment of the present invention.

FIG. 5 is a block diagram showing a basic configuration of a conventional tone synthesis device.

FIG. 6 is an explanatory diagram of a spectrum distribution of a musical tone synthesized by the apparatus of FIG. 5;

[Explanation of symbols]

1, 6: phase accumulator, 2: delay feedback type tone synthesis algorithm, 3: counter, 4: waveform memory, 5: control circuit.

──────────────────────────────────────────────────続 き Continued on front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G10H 7/08

Claims (2)

(57) [Claims] 1. A tone synthesizer for inputting an initial waveform signal to a closed loop including a delay unit and performing a cyclic process in the closed loop to synthesize a tone, wherein a delay time of the delay unit and a cycle of the initial waveform signal are determined. independently set, in which the delay time of the delay means is set in accordance with the tone pitch of the musical tones to be generated, the peripheral of the initial waveform signal
A tone synthesizer characterized in that the wave number is not proportional to the pitch. 2. An initial waveform signal in a closed loop including delay means.
, And circulates in the closed loop to synthesize a musical tone
In the musical sound synthesizer, The delay time of the delay means and the cycle of the initial waveform signal
Independently set, the delay time of the delay means is generated
Set according to the pitch of the musical tone, but the
The wave number is variably set so that it is not proportional to the pitch
Music synthesizer.