JPS62169198A - Noise generation circuit - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to sound p4 composition. Noise generation in a voice conference Fft device that performs predetermined arithmetic processing based on data read out from a memory that stores fc data that has been pulled at predetermined time intervals [91].

[Conventional technology]

Conventionally, in order to add audio to BZ, data sampled at all cycles of the original sound is stored in a memory.

A method is known in which predetermined arithmetic processing is executed based on the data read out from the memory, and finally an analog signal sound is released by a DA converter.

[Problem that the invention seeks to solve]

In the above-mentioned conventional speech sweet and sour method, the speech noise waveform is a random waveform with a frequency lower than that of the tone waveform. Therefore, if the noise waveform is to be sampled faithfully as it is, the sampling interval must be made very small. As a result, a huge amount of sampling data is required, and the memory area allocated to the tone waveform is often limited. Furthermore, the voice noise
As shown in the diagram, the spectrum is a continuous spectrum with peaks at a specific frequency α of the voice type VC. It is normal for the peak frequency of this frequency spectrum to be around 8 KHz for voice noise in the 2nd row, and around 4 kHz for the voice noise in the 4th row. In order to discover a sweet and sour noise waveform with a continuous spectrum, a very complicated pseudo-speech noise generation circuit has conventionally been required. The original sound of the audio noise is sampled, and a part of the friend waveform is stored. , a match P1 path for detecting that a memory address matches a specific address, and an L) Af converter for converting the output of the memory into an analog address.

An audio noise waveform that has a spectrum peak at a specific frequency has a frequency that roughly corresponds to the peak frequency when viewed over an hour.
Waveforms with 1ff components are lined up, and low frequency components and low frequency components are mixed, but their amplitudes are considerably smaller than the peak frequency components. If the waveform of the peak frequency component is repeated to generate a voice noise waveform with such characteristics, a single spectrum will result.In the present invention, several waveforms are extracted from the original sound.However, A line spectrum having a frequency Pl component of the peak frequency of the original sound that repeats these steps as it is is a continuous spectrum, but it is not a continuous spectrum. Therefore, when repeating several waveforms extracted from the original sound, it is possible to generate a waveform with a continuous frequency spectrum by repeating the start position of the waveform at random. For example, a polynomial counter can be used to randomly set the starting position. In the present invention, a memory address is specified using a polynomial counter so that the noise waveform data extracted from the original sound is output to the set start position.

Note that the start position can be randomly set by setting the data obtained by inverting the contents of the polynomial counter for random number generation as the initial value of the polynomial counter for specifying the memory address.

〔Example〕

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a schematic diagram of one embodiment of the present invention. Address input terminals 2 and 3 specify all the upper addresses of the memory 4, and polynomial input terminals 7 specify all the lower addresses of the memory 4. do. By switching the inputs of address input terminals 2 and 3, four types of noise waveforms can be selected. memory 4
The content of the original sound noise as shown in Figure 3 is 25
The sampling data of 31 points sampled every μ8eC is stored. Polynominal counters 7 and 9 in the circuit configuration diagram of FIG. 1 are both 5-bit polynominal counters shown in the second column, and the contents of these polynominal counters are written in the order in which they are written as memory addresses in FIG. 3. changes. When all the outputs of this polynomial counter are inverted by inverters, the data changes to the song written on the fourth vertical axis. 40 KFI to clock input terminal 1 of memory addressing polynomial counter 7
All the clocks of z are input, and the match circuit 8 detects that the value of the polynomial counter 7 matches O or 31.

When a match occurs, the content of the random number generation polynomial counter 9 is moved by one using the matching output of the matching circuit 8, and
f Invert it with an inverter and set it in the memory address designation polynomial carantuff. At this time.

The start address of the original sound noise sampling waveform stored in the memory 4 is set in the memory address designation polynomial counter 7. FIG. 4 shows changes in the position of the start address of the original sound noise sampling waveform corresponding to the movement of the polynomial counter for random number generation. The vertical axis shows the change in the inverted output of the random number generation polynomial counter from top to bottom, and the horizontal axis shows the memory address start position at that time.

It can be seen that the memory address starts changing randomly.

When the start address is set in the memory address designation polynomial counter 7, the coincidence output of the coincidence 10'' path 8 is no longer output, and the polynomial counter 7 becomes 40KF (z
The address of the memory 4 is designated to +IIi by the clock, and the original sound noise sampling waveform inside the memory 4 comes out as the output of the memory 4.

This behavior? When the output of the memory 4 is repeatedly converted into an analog signal by the 1DAi converter 5, a signal having the continuous frequency spectrum shown in FIG. 6, which is close to the frequency spectrum of the original sound noise shown in FIG. 5, is obtained at the output terminal 6. Ru. In practice, the period of the polynomial counter 9 for random number generation has almost no effect on the synthesized spectrum. The coincidence detection circuit detects all the data of O and uses it as the end address of the memory, but it also detects the data of 31 because the inverted output of the random number generation polynomial counter contains the data of 31, but it is not the memory address 1. When the constant polynomial counter receives data of 31, the polynominal cuff/
This is to prevent 41 from being stuck at 31 due to the inherent nature of the situation.

〔Effect of the invention〕

As explained above, according to the present invention, No. 13 having a continuous layer @fi spectrum can be synthesized by extracting a portion of the data sampled from the original sound waveform and repeating the data. The sampling interval is 50 μ8e because the frequency at which the 4f frequency spectrum of voice noise is maximum is 8 KHz 1.
e If the sampling interval is less than
Sampling every 5 μsec will result in σ40,000 sampling points in 1 second, and 40,000 memory addresses will be required, but with the present invention, even if the composite data is output every 25 μsec, the memory capacity t will only need 32 addresses. The effect of reducing memory capacity is remarkable. Also.

Depending on the storage capacity, noise signals with different frequency spectra can be created without increasing the memory capacity.

In particular, since a polynomial counter is used to specify the memory address, the circuit is simpler than using a binary counter, etc., which is a half-body circuit, and the area occupied by the circuit is smaller. @, effective in reducing chip size. Furthermore, since it is possible to create 4 No. 16 signals with a continuous frequency spectrum using all periodic clocks with a simple circuit, it is suitable for digital LSI integration, and since data obtained by sampling the noise waveform of the original sound is used, the frequency spectrum is close to that of the original sound noise. A synthesized noise with n1 is obtained, which is a +1-composition of high quality speech noise. In addition, the number of stages of the polynomial counter, the clock input frequency of the memory addressing polynomial counter, the number of output bits of the memory, D
The number of bits of the A converter, etc. is related to the sound quality of the synthesized noise, and it is obvious that an appropriate value of 1+ffk can be selected.

[Brief explanation of drawings]

FIG.
The figure is a circuit diagram of a specific example of the polynomial counter in the first section.
Is Fig. 3 the original sound noise waveform stored in memory 4 in Fig. 1? Between the sampled data and the inverted output of the random number generation polynomial counter, FIG. 4 shows the memory start address digit Wk. Note 5 is the spectral waveform of the original sound noise.
FIG. 6 is a diagram of the frequency spectrum wave of the fifth generation noise. 1... Clock input terminal, 2-3...
Memory upper address input terminal, 4...Memory,
5...DA converter, 6...Output terminal,
7...Memory address specification polynomial counter, 8...-Fingerprint detection circuit. 9...Random number generation polynomial counter, 10.
...7 rip 70 1% 11...exclusive ORO

Claims (1)

[Claims] It has a memory that stores part of the waveform obtained by sampling the original noise sound, and a polynomial counter that periodically operates to address this memory, and when the content of the polynomial counter that specifies the memory address is 0. 1. A noise generating circuit characterized in that, when , the contents of a polynomial counter for random number generation are set in the polynomial counter for specifying a memory address.