JP5474145B2 - Fade processing device for DSD signal in audio products - Google Patents

Description

  According to the present invention, a DSD (Direct Stream Digital) signal in an audio product is sequentially processed so that an extra memory is not required for the fading process, and the fading of the DSD signal in the audio product can be reduced without switching the fade signal. The present invention relates to a processing apparatus. This DSD system is a simple system in which an A / D converted 1-bit signal is recorded as it is, and reproduction is also passed through an analog low-pass filter. The DSD system adopted for the super audio CD is a system that expresses the size of an audio signal by the density (shading) of a 1-bit digital pulse, unlike a conventional PCM (Pulse Code Modulation) system.

The following digital signal processing apparatus and digital signal processing program (published patent publication) have been proposed as techniques for preventing the occurrence of noise when performing conventional fade processing of an audio signal.
According to the published patent publication, a delay processing means for delaying and outputting a predetermined number of bits of an input signal subjected to sigma delta modulation processing for a predetermined time, and converting the input signal into a signal of a predetermined number of bits of 2 bits or more. A bit length modulation means for modulating, a multiplying means for multiplying the signal of a predetermined number of bits of 2 or more and a coefficient sequentially supplied from the coefficient generating means, and a multiplication result signal output from the multiplying means on the time axis The first time reversing means for inverting and outputting the sine, the sigma delta modulating means for performing the sigma delta modulation processing on the output signal from the first time reversing means, and the sigma delta modulating means A second time reversing means for inverting and outputting the re-sigma delta modulation signal on the time axis, an output signal from the delay processing means, and an output from the second time reversing means. Control means for switching and outputting the output, and further, the sum of signal delay times by the three means of the first time inversion means, the sigma delta modulation means, and the second time inversion means is calculated. There is a digital signal processing apparatus characterized in that the delay processing means is configured to be the same as a delay time for delaying the input signal, and the coefficient generating means is means for supplying a coefficient every predetermined time. JP, 2007-267091, A).

JP2007-267091A

  The above-mentioned patent document describes that an effect that it is possible to realize a digital signal processing device and a program that do not generate noise when performing fade processing is obtained.

  However, the invention of Patent Document 1 requires a memory for time reversal and is generally difficult to realize in an embedded system with a limited memory capacity. Further, in the above document, since modules such as signal selection and mute pattern generation exist, there are difficulties such as an increase in the scale of digital signal processing.

The present invention has been made in order to solve the above-described problems, and an object of the present invention is to realize a fade function by sequentially processing input signals, so that no extra memory is required and the memory capacity is reduced. It can be realized with a small system, mute pattern, switching of source signal and fade signal is not required, processing scale can be reduced, output signal after fade-in matches input signal, and output signal after fade-out is 0 An object of the present invention is to provide a fade processing apparatus for a DSD signal in an audio product in which 1 is a mute pattern that repeats in the same number.

A fade processing apparatus for a DSD signal in an audio product according to the present invention includes a fader unit having output level adjustment and setting in a device handling the DSD system, and one of the output sides of the fader unit as a multi-bit signal on the input side. A second-order ΔΣ modulator connected to the output of the fader unit, and a mixing feedback value calculation unit branched from the output side of the fader unit and connected to the input side, and a mixing feedback from the output side of the mixing feedback value calculation unit The value DITER is added to the feedback value of the first stage of the secondary ΔΣ modulator, and the branch signal on the output side of the secondary ΔΣ modulator is connected to the input side of the feedback value calculation unit for mixing , The input X is taken out as a 1-bit output Y from the output side of the secondary ΔΣ modulator .

In claim 1 mixes feedback value calculation section according period from after completion fade-before fade starts, if the second-order ΔΣ modulator output does not match the fixed value C, enter the secondary ΔΣ modulator When the outputs match, or during fading or muting, the mixing feedback value is set to 0.

  Since the DSD signal fade processing apparatus in the audio product according to the present invention sequentially processes the input signal and obtains the output signal, the time reversal operation is unnecessary, and the corresponding memory becomes unnecessary.

  In addition, since the present invention sequentially processes the input signal and obtains the output signal, it is not necessary to perform a large amount of calculations all at once even when real-time processing is assumed, and a highly practical product can be obtained.

Furthermore, the present invention does not require switching of output signals, and does not require a delay device or a switch, so that the amount of calculation can be reduced.
In the present invention, it is not necessary to switch the mute pattern, source signal, and fade signal, and the processing scale can be reduced.

  Since the output signal after the fade-in coincides with the input signal, a switch for that purpose is unnecessary.

  The output signal after the fade-out becomes a mute pattern in which 0 and 1 are repeated in the same number, and a satisfactory sound can be provided to the listener.

It is a block diagram which shows one Example of this invention. It is explanatory drawing which shows the fader part which shows one Example of this invention. It is explanatory drawing which shows the secondary delta-sigma modulator which shows one Example of this invention. It is a relationship diagram of the normalized frequency and noise transfer frequency which show one Example of this invention.

Hereinafter, the present invention will be specifically described with reference to the drawings.
FIG. 1 is a block diagram showing an embodiment of the present invention.
Although not shown in FIG. 1, the audio data used in the present invention is a DSD system. For example, a super audio CD or the like is used as the recording medium. In the conventional PCM system, various filters are applied, and sampling is performed after correction and thinning, so that the number of processes in between is increased. On the other hand, in the DSD method, the sampling frequency is 2822.4 kHz, which is 64 times as fine as the CD, and this is recorded as it is as 1-bit data. Similarly, one bit is processed at the time of reproduction. For this reason, the DSD system has a simple structure in which there is little conversion that only passes through an analog low-pass filter. Therefore, a sound close to the original sound can be obtained. That is, the CD cuts over 22 KHz, which cannot be heard by human ears, but the frequency of the super audio CD covers 100 KHz.

1 and 2, an input signal X is input as a 1-bit audio signal to a bit length converter 2 that is a digital-analog converter (DAC) of the fader unit 1. This converter 2 is not different from existing technology. That is, when the input signal is X and the output signal is Y, the bit length converter 2 is Y = 1 (X = 1), Y = −1 (X = 0), and the output signal Y becomes multibit. In this embodiment, a 16-digit binary system is adopted, but the present invention is not limited to this, and it may be 24 bits expressed in a 24-digit binary system. In FIG. 2, the coefficient processing after the DA conversion of the X signal is performed with a ₁ = 0.99 and a ₂ = 0.95.

  The multi-bit signal on the output side of the fader unit 1 is sent to the input side of the secondary ΔΣ modulator 3. The output side of the secondary ΔΣ modulator 3 is taken out as a 1-bit audio signal Y. The multi-bit signal on the output side of the fader unit 1 is also sent to the input side of the feedback value calculation unit 4 for mixing. The output side of the mixing feedback value calculation unit 4 is transmitted to the input side of the secondary ΔΣ modulator 3. The 1-bit audio signal Y on the output side of the secondary ΔΣ modulator 3 is fed back to the input side of the feedback value calculation unit 4 for mixing.

The details of the D / A conversion by the secondary ΔΣ modulator 3 are shown in FIG.
The principle of ΔΣ is that a difference between a signal obtained by integrating an analog input and a signal obtained by D / A conversion of a digital output is obtained and quantized by comparison with a midpoint voltage. The digital signal is oversampled to a sufficiently high frequency to distribute the requantization noise over a wide band, and the second-order ΔΣ modulator 3 is used to shape the requantization noise. In the figure, 5 indicates a quantizer. When high-speed low-bit D / A conversion is performed and re-quantization noise is removed by an analog low-pass filter, an analog signal in which an S / N ratio is secured can be obtained.

  In the mixing feedback value calculation unit 4, when the input / output of the secondary ΔΣ modulator 3 does not match between the completion of fade-in and before the start of fade-out, the fixed value C is used. It is set to 0 when the input / output of the secondary ΔΣ modulator 3 is coincident, or during fading or muting. In this embodiment, the lower 10 bits of the 16-bit fixed point are the decimal part, and C = 1 (= 1/1024 = 0.000976...).

FIG. 3 is an explanatory diagram showing a second-order ΔΣ modulator showing an embodiment of the present invention.
In the figure, 2 is a DAC (Digital Analog Converter), and 5 is a quantizer. Quantization is a technique for approximately expressing a continuous amount of an analog signal or the like with a discrete value such as an integer. In the quantizer 5, when the input is X and the output is Y, Y = 1 (X ≧ 1) and Y = 0 (X <0). In the figure, DITHER is a digital signal processing technique and has the effect of increasing the apparent bit depth. It is used to suppress signal deterioration when an A / D converter that takes in an analog signal and digitizes it, or when a digital signal processed in 24 or 32 bits is output in 16 bits inside the device. In this embodiment, this is a process for ensuring that the input and output signals of the second-order ΔΣ modulator 3 quickly match, and is calculated by the mixing feedback value calculation unit 4.

When the quantizer 5 in FIG. 3 is regarded as a noise source (E), the transfer function of the second-order ΔΣ modulator 3 is as follows.

Y = X + (1-a ₁ Z ⁻¹ ) (1−a ₂ Z ⁻¹ ) E

Normally, a ₁ and a ₂ are 1, but noise is generated in the vicinity of the moment when the gain becomes 1 by fading in. In order to suppress this behavior, a ₁ = 0.99 and a ₂ = 0.95. These values are obtained as the optimum values by repeating the experiment, and do not need to match exactly at the time of implementation.
By setting a ₁ and a ₂ to the above-described values, the data is smoothly switched to original data (input data) after fading in, and smoothly converges to a sufficiently small value after fading out. However, it may not be switched to the original data after fading in. It is known that this phenomenon occurs frequently and does not converge to the original data as time passes. Therefore, in the present invention, in order to realize reliable and smooth switching, a value (DITER shown in FIG. 3) in the first-stage feedback value is added. It has been obtained according to the present invention that the DITER value (mix feedback value) plays a role of gradually changing the feedback value and finally converging the noise to zero.

FIG. 4 is a graph showing the relationship between the normalized frequency and the noise transfer frequency according to an embodiment of the present invention.
For NTF (Noise Transfer function), the solid line indicates the NTF when a ₁ = 0.99 and a ₂ = 0.95.
In the figure, 1st order DSM is indicated by a one-dot chain line, 2nd order DSM is indicated by a dotted line (broken line), and a proposal 2nd order DSM according to the present invention is indicated by a solid line. As shown in FIG. 4, the characteristics in the low frequency region during the fade are deteriorated, but it is understood that the value is sufficiently low in terms of hearing.

  The present invention provides a DSD signal fade processing apparatus for audio products, which can provide a sound manufacturer that handles audio products to a sound product distributor or a consumer who is an enthusiast of audio products at a relatively inexpensive price. Can be expected.

1 Fader section 2 DAC (Digital Analog Converter)
3 Second-order ΔΣ modulator 4 Mix feedback value calculation unit 5 Quantizer

Claims (2)

In a device that handles the DSD system, a fader unit having output level adjustment and setting, a secondary ΔΣ modulator that connects one of the output sides of the fader unit to the input side as a multi-bit signal , and an output of the fader unit And a mix feedback value calculation unit that is branched from the input side and connected to the input side, and DITHER , which is a mix feedback value, is output from the output side of the mix feedback value calculation unit to the first stage of the secondary ΔΣ modulator. And a branch signal on the output side of the secondary ΔΣ modulator is connected to the input side of the mixing feedback value calculation unit , and the input X is input to the 1-bit input from the output side of the secondary ΔΣ modulator. A fade processing apparatus for a DSD signal in an audio product, which is extracted as an output Y. 2. The mixing feedback value calculation unit according to claim 1, wherein when the input / output of the secondary ΔΣ modulator does not match between the completion of fade-in and before the start of fade-out, the mixing feedback value is set to a fixed value C. 2. A DSD signal fade processing apparatus in an audio product according to claim 1, wherein the feedback value for mixing is set to 0 when the inputs and outputs of the [Delta] [Sigma] modulator match, or during fading or muting. .

Images