RU2703895C1 - Digital signal production method for electric musical instruments - Google Patents

Abstract

FIELD: electric musical instruments.

SUBSTANCE: invention relates to electric musical instruments and enables to obtain an audio signal in the form of digital readings using the heterodyning principle. Said technical result is achieved by using a digital phase generator and transcoding the phase into digital samples of the audio signal, wherein phase generator is a clock-pulse counter and a sampling-storage unit connected to it, the counter is clocked from one signal source, and the sampling-storage unit is clocked from the other.

EFFECT: possibility of combining the advantages of digital signal processing techniques with instant response of a musician.

4 cl, 3 dwg

Description

FIELD OF THE INVENTION

The invention relates to electric musical instruments and allows you to receive an audio signal in the form of digital samples using the principle of heterodyning (obtaining the differential frequency). This makes it possible to combine the advantages of digital signal processing methods with the instant response of the instrument to the musician's actions.

State of the art

A known method of producing an analog sound signal using heterodyning - the formation of a differential frequency during the interaction of signals from two generators (Voloshin VI, Fedorchuk LI Electromusical instruments. M., "Energy", 1971, S. 23). As an element producing a difference frequency, a special mixer (beating detector) is used. To implement digital signal processing methods that provide ample opportunities in the field of sound formation, an additional analog-to-digital converter and a high-quality filter at its input are required, which suppresses the remnants of high-frequency components.

Known methods for obtaining a digital signal using heterodyning performed by a D-trigger or logical elements “AND” / “OR” / “EXCLUSIVE OR” (Theremin “Skywave H1” - URL: http://www.thereminworld.com/files/Pages /28/files/SkywaveHi.pdf, access date: 03/20/2019; I. Nechaev. Theremin. Radio., 1986, No. 10, p. 49). The resulting signal is a digital sequence of zeros and ones that require smoothing (filtering), and its spectrum contains dissonant combination frequencies.

Electromusical instruments are known that mimic instruments such as “Theremin” by the method of interaction, and for which an audio signal is obtained using digital generators (OpenTheremin project - URL: http://www.gaudi.ch/OpenTheremin/, access date: 20.03. 2019); Theremini "Theremini" company Moog Music Inc., USA). These generators are controlled by data obtained by measuring the frequency of the tunable generator, and due to the delays inherent in the measurement algorithms, such instruments have a low reaction rate to the actions of the musician.

The closest analogue to the invention is a method of obtaining a digital signal based on a phase generator and a sine table, which in the technique of constructing musical synthesizers is called "table synthesis"("WavetableSynthesis"), and in the transceiver and measuring technique - "DDS"("Direct Digital Synthesis "). With this method, the value of a special register (“phase accumulator”) of sufficiently large bit N increases with a frequency F _s by a certain value Δ, and the higher bits of the register are used to address a table that describes one period of the generated oscillation (in the simplest case, a sinusoidal one). The increase is ignored overflow, so the growth is cyclically repeated from some minimum values. The frequency of the cycles is determined by the width of the register and the increment Δ:

F = F _s ∙ Δ / 2 ^N

Usually, F _s represents the sampling frequency of the audio signal, and so that the latter does not have spurious spectral components, it is necessary (in accordance with the Kotelnikov-Nyquist criterion) that the value of Δ be less than 2 ^N by at least two times. If the table describes a signal with a shape other than sinusoidal, then the requirements for the maximum value of Δ are tightened. Instead of the table, other recoding methods can be used, for example, algorithmic ones.

To simulate the unified sound of musical instruments of the gliding type (trombone, ukulele, "Thereminvox", etc.), the Δ value should be updated quite smoothly and often. This is not always achievable in practice, since the sensors or algorithms used to measure the position of the musician’s hands have a limited speed of outputting results.

Disclosure of the invention

The technical result of the invention is to obtain an audio signal in the form of digital samples (samples) in a heterodyne way - by means of the formation of a difference frequency in the interaction of the frequencies of two signal sources.

The indicated technical result is achieved using a digital phase generator and phase conversion to the sound signal samples, the clock pulse counter and the sampling-storage unit connected to it being used as the phase generator, the counter is clocked from one signal source, and the sampling-storage unit from another. The counter states stored by the sampling and storing unit represent a cyclically changing phase of the difference frequency signal, which is algorithmically or using tables to be transcoded into digital samples of the audio signal.

The process of generating the differential frequency is clearly shown in figure 1. Graph 1 is a stepwise increasing counter value. As soon as the counter reaches its maximum value, the count is repeated first. For a regular binary counter, overflow causes the count to be repeated from zero. Then the period of operation of the counter T will be determined by the period of the clock signal T _clk and the capacity of the counter N:

T = 2 ^N ∙ T _clk .

Accordingly, the frequency of counter overflows will be equal to:

f = 1 / T = 1 / (2 ^N ∙ T _clk ).

With a frequency f _s = 1 / T _s , the state of the counter is memorized (“photographed”), and the results of this process, marked with points 2, form an envelope 3, which has a sawtooth shape. The repetition frequency of the "teeth" of the envelope corresponds to the difference frequency:

F = ff _s .

Provided that f is less than f _s (or when the counter works to decrease), the envelope will take a mirror form.

To obtain a sinusoidal (or other) waveform, the values of samples 2 corresponding to the sawtooth signal 3 are recoded using tables or appropriate algorithms.

Brief Description of the Drawings

The figure 1 shows the process of the formation of the differential frequency between the frequency of the counter overflows f = 1 / T and the frequency of "photographing" its state f _s = 1 / T _s .

Figure 2 shows an embodiment of the invention using a counter and a register-latch, which are clocked by the edges of the pulses coming from two different generators.

Figure 3 shows an embodiment of the invention using a microcontroller incorporating the necessary nodes (circled by a dashed line), and a controlled amplifier and two digital-to-analog converters are used to form the final sound signal: one to receive the signal of the required shape, the other to amplitude envelope.

The implementation of the invention

The proposed method can be implemented as shown in figure 2. The binary counter 6 is clocked by pulses coming from the clock 4. The register latch 7 is connected to the output of the counter 6 and selects and stores along the front of the pulses coming from the tunable generator 5 current state of the counter. The output of the register-latch are digital phase values of the signal, which are encoded using the sine table 8 into an audio signal. This signal is transmitted for further digital processing, or immediately to a digital-to-analog converter for subsequent sound amplification (not shown). Instead of a sine table, a signal table of a different form can be used, recoding can be not tabular, but algorithmic, both hardware and software implementations are possible.

The latch register can also be part of the timer counters that are part of more complex integrated circuits - microcontrollers, digital signal processors, programmable logic chips, etc. Their use reduces the number of components in the final product. For example, figure 3 shows an embodiment of the invention for use in an Theremin type electric musical instrument. It includes a microcontroller (shown by a dashed line) containing:

- counter-timer with latch 10 (“Capture Counter” or “Capture Timer” in English terminology),

- a digital processing device 11,

- clock generator 4,

- two digital-to-analog converters 12 and 13,

- frequency divider 9.

In this embodiment, using the counter-timer 10, which has a built-in sample-storage unit, digital phase samples are obtained, which are transmitted to the digital processing device 11 (implemented programmatically), and the processing results are transmitted via digital-to-analog converters 12 and 13 to an external controlled amplifier 14.

The need to use two digital-to-analog converters can be caused by insufficient capacity of converters available for microcontrollers of general use (usually from 10 to 12 binary digits). Two such transducers, one of which is used to obtain the desired shape of the sound signal, and the other by its amplitude envelope, allows to obtain sound of acceptable quality with a wide range of volume changes.

Also, to implement contactless control, the tunable generator 5 is controlled by the musician’s hand capacity, and the frequency divider 9 provides a lowering of the frequency necessary to improve the linearity of the musical scale when using high operating frequencies (the method described in patent RU 2670397).

Claims (4)

1. A method of obtaining a digital signal from electric musical instruments using a digital phase generator and transcoding the phase into digital samples, characterized in that a clock counter and a sample-storage unit connected to it are used as a phase generator, the counter being clocked from a single signal source, and the fetch-storage node is from another. 2. The method according to p. 1, characterized in that one or both of the signal sources have a variable frequency. 3. The method according to p. 1 or 2, characterized in that they use a counter that works to reduce. 4. The method according to PP. 1, 2 or 3, characterized in that they use a counter with a built-in function of sampling-storage, and the counter can be part of more complex circuits.

Images