Bit extension processing device, bit extension processing method, and storage medium

Abstract

A frame processing section (1) frames a first digital music signal of a first number of quantization bits. A difference signal calculating section (2) calculates a first difference signal that takes a difference value between two adjacent samples as a sample. A flat region detecting section (3) detects a flat region in the first difference signal. A flat region correcting section (4) generates a second difference signal that makes the flat region non-flat. A difference signal averaging section (5) calculates a difference average value for each frame, and generates a third difference signal by subtracting the difference average value from each sample value of the second difference signal. A re-quantization error generating section (6) generates a re-quantization error signal for each frame based on the third difference signal. An adding section (7) adds the re-quantization error signal to the first digital music signal included in each frame, and outputs a second digital music signal of a second number of quantization bits.

Description

Technical Field

[0001] This invention relates to a bit extension processing apparatus, a bit extension processing method, and a bit extension processing program for extending the quantization bit depth of digital music signals. Background Technology

[0002] There are cases where a digital music signal with an m-bit quantization bit depth can be extended to n bits using a bit extension processing device, where m and n are positive integers, and n > m. A simple bit extension processing device generates an n-bit digital music signal by adding (nm) bits of zero data to the lower bits of the m-bit digital music signal. Typically, bit extension processing tools installed in audio editing software employ this simple bit extension method.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent document 1: Japanese Patent Application Publication No. 2004-180017. Summary of the Invention

[0006] Patent Document 1 describes an improved bit-expansion processing procedure. According to the bit-expansion processing procedure described in Patent Document 1, when processing digital music signals of quiet pieces with consecutive sample values ​​as the object of bit-expansion processing, the quality (sound quality) of the bit-expansion processing can be improved. However, when processing digital music signals of pieces with rapidly changing sample values, the quality cannot be significantly improved.

[0007] One or more embodiments aim to provide a bit extension processing apparatus, bit extension processing method, and bit extension processing program that can perform high-quality bit extension processing on digital music signals regardless of the music genre.

[0008] According to a first aspect of one or more embodiments, a bit extension processing apparatus may be provided, comprising: a framing processing unit, a differential signal calculation unit, a flat region detection unit, a flat region correction unit, a differential signal averaging unit, a requantization error generation unit, and an addition unit.

[0009] The framing processing unit divides and frames the samples of the first digital music signal quantized with the first quantization bit depth into multiple sample sizes. The differential signal calculation unit calculates a first differential signal, which uses the difference between two adjacent samples of the first digital music signal contained in each frame as differential samples. The flat region detection unit detects the start and end positions of two or more consecutive flat regions with the same sample value in the first differential signal, as well as the number of differential samples in the flat region.

[0010] The flat region correction unit uses one or more differential samples in the flat region as differential samples to be corrected, based on whether the sample values ​​increase or decrease from the previous differential sample in the flat region to the differential sample at the starting position, and whether the sample values ​​increase or decrease from the differential sample at the ending position to the next differential sample in the flat region. The unit corrects the sample values ​​of the correction targets by adding or subtracting correction values, thereby generating a second differential signal that makes the flat region non-flat.

[0011] The differential signal averaging unit sums the sample values ​​of the second differential signal contained in each frame, divides them by the number of samples of the second differential signal contained in each frame to calculate the differential average value, and subtracts the differential average value from each sample value of the second differential signal to generate a third differential signal.

[0012] The requantization error generation unit generates a requantization error signal for each frame based on the third differential signal, represented by the difference between the first quantization bit depth and a second quantization bit depth that is greater than the first quantization bit depth. The addition unit adds the requantization error signal to the first digital music signal contained in each frame and outputs a second digital music signal with the second quantization bit depth.

[0013] According to a second aspect of one or more embodiments of the present invention, a bit extension processing method is provided that performs the following processing: In this bit extension processing method, samples of a first digital music signal quantized with a first quantization bit depth are segmented and framed according to a plurality of sample numbers; a first differential signal is calculated, wherein the difference value between two adjacent samples of the first digital music signal contained in each frame is used as a differential sample; and the start and end positions of two or more consecutive flat regions with the same sample value in the first differential signal, and the number of differential samples in the flat regions are detected.

[0014] The bit-expansion processing method, based on whether the sample value increases or decreases from the previous differential sample in the flat region to the differential sample at the starting position, and whether the sample value increases or decreases from the differential sample at the ending position to the next differential sample in the flat region, takes one or more differential samples in the flat region as differential samples to be corrected, and corrects the sample value of the correction object by adding or subtracting a correction value, thereby generating a second differential signal that makes the flat region non-flat.

[0015] In the bit-expansion processing method, the sample values ​​of the second differential signal contained in each frame are summed, divided by the number of samples of the second differential signal contained in each frame to calculate the differential average value, and the differential average value is subtracted from each sample value of the second differential signal to generate a third differential signal.

[0016] In the bit-expansion processing method, based on the third differential signal, a requantization error signal for each frame is generated, which is represented by the difference between the first quantization bit depth and a second quantization bit depth that is greater than the first quantization bit depth; and the requantization error signal is added to the first digital music signal contained in each frame to generate a second digital music signal with the second quantization bit depth.

[0017] According to a second aspect of one or more embodiments of the present invention, a computer-readable storage medium is provided storing a bit-expansion processing program that causes a computer to perform the following steps. The bit-expansion processing program causes a computer to perform the following steps: segmenting and framing samples of a first digital music signal quantized with a first quantization bit depth according to a plurality of sample numbers; calculating a first differential signal, wherein the first differential signal uses the difference value between two adjacent samples of the first digital music signal contained in each frame as differential samples; and detecting the start and end positions of two or more consecutive flat regions with the same sample value in the first differential signal, and the number of differential samples in the flat regions.

[0018] The bit extension processing program causes the computer to perform the following steps: based on the pattern of whether the sample value increases or decreases from the previous differential sample in the flat region to the differential sample at the start position, and whether the sample value increases or decreases from the differential sample at the end position to the next differential sample in the flat region, taking one or more differential samples in the flat region as differential samples to be corrected, correcting the sample value of the correction object by adding or subtracting a correction value, and generating a second differential signal that makes the flat region non-flat.

[0019] The bit extension processing program causes the computer to perform the following steps: summing the sample values ​​of the second differential signal contained in each frame, dividing by the number of samples of the second differential signal contained in each frame to calculate the differential average value, and subtracting the differential average value from each sample value of the second differential signal to generate a third differential signal.

[0020] The bit extension processing procedure causes the computer to perform the following steps: generating a requantization error signal for each frame based on a third differential signal, represented by the number of bits representing the difference between the first quantization bit depth and the second quantization bit depth, which is greater than the first quantization bit depth; and adding the requantization error signal to the first digital music signal contained in each frame to generate a second digital music signal with the second quantization bit depth.

[0021] Bit extension processing apparatus, bit extension processing method, and bit extension processing program according to one or more embodiments are capable of performing high-quality bit extension processing on digital music signals, regardless of the type of music. Attached Figure Description

[0022] Figure 1 This is a block diagram illustrating a bit extension processing apparatus according to one or more embodiments.

[0023] Figure 2 This is a diagram conceptually illustrating the operation of the differential signal calculation unit and the flat region detection unit in a bit extension processing apparatus according to one or more embodiments.

[0024] Figure 3 It is a waveform diagram representing the requantization error signal of two digital music signals that were quantized and recorded in two formats: 24-bit quantization and 16-bit quantization.

[0025] Figure 4 This is a waveform diagram showing the difference between two adjacent samples in a 16-bit quantized digital music signal.

[0026] Figure 5A This is a diagram illustrating a rising-flattening-falling pattern with a sample size of 2 for the flat region, and the first example of the correction method for the flat region in this case.

[0027] Figure 5B This is a diagram illustrating the first example of a rising-flat-rising pattern with a sample size of 2 for the flat region, and the correction method for the flat region in this case.

[0028] Figure 5C This is a figure illustrating a first example of a descending-flattening-ascending pattern with a sample size of 2 for the flat region, and the correction method for the flat region in this case.

[0029] Figure 5D This is a figure illustrating the first example of a descending-flattening-descending pattern with a sample size of 2 for the flat region, and the correction method for the flat region at this time.

[0030] Figure 6A This is a diagram illustrating the first example of a rising-flattening-falling pattern with 3 samples in the flat region and the correction method for the flat region at this time.

[0031] Figure 6B This is a diagram illustrating the first example of a rising-flat-rising pattern with 3 samples in the flat region and the correction method for the flat region at this time.

[0032] Figure 6C This is a figure illustrating the first example of a falling-flattening-rising pattern with 3 samples in a flat region and the correction method for the flat region at this time.

[0033] Figure 6D This is a figure illustrating the first example of a descending-flattening-descending pattern with 3 samples in the flat region and the correction method for the flat region at this time.

[0034] Figure 7A This is a diagram illustrating the first example of a rising-flattening-falling pattern with a sample size of 4 for the flat region, and the correction method for the flat region at this time.

[0035] Figure 7B This is a figure illustrating the first example of a rising-flat-rising pattern with 4 samples in the flat region and the correction method for the flat region at this time.

[0036] Figure 7C This is a figure illustrating a first example of a descending-flattening-ascending pattern with a sample size of 4 for the flat region, and the correction method for the flat region at this time.

[0037] Figure 7D This is a figure illustrating the first example of a descending-flattening-descending pattern with a sample size of 4 for the flat region, and the correction method for the flat region at this time.

[0038] Figure 8A This is a diagram illustrating the first example of a rising-flattening-falling pattern with 5 samples in the flat region and the correction method for the flat region at this time.

[0039] Figure 8B This is a diagram illustrating the first example of a rising-flat-rising pattern with 5 samples in the flat region and the correction method for the flat region at this time.

[0040] Figure 8C This is a diagram illustrating a first example of a descending-flattening-ascending pattern with 5 samples in a flat region, and the correction method for the flat region in this case.

[0041] Figure 8D This is a figure illustrating the first example of a descending-flattening-descending pattern with 5 samples in the flat region and the correction method for the flat region at this time.

[0042] Figure 9 This is a second example of a falling-flat-falling pattern with 2 samples in the flat region and the correction method for the flat region at this time.

[0043] Figure 10A This is a second example of a rising-flattening-falling pattern with 3 samples in the flat region, and the correction method for the flat region in this case.

[0044] Figure 10B This is a second example of a rising-flat-rising pattern with 3 samples in the flat region and the correction method for the flat region in this case.

[0045] Figure 11A This is a second example of a rising-flattening-falling pattern with 4 samples in the flat region, and the correction method for the flat region in this case.

[0046] Figure 11B This is a second example of a rising-flat-rising pattern with 4 samples in the flat region and the correction method for the flat region in this case.

[0047] Figure 11C This is a second example of a falling-flattening-rising pattern with 4 samples in a flat region, and the correction method for the flat region in this case.

[0048] Figure 11D This is a second example of a falling-flat-falling pattern with 4 samples in the flat region and the correction method for the flat region at this time.

[0049] Figure 12A This is a second example of a rising-flattening-falling pattern with 5 samples representing a flat region, and the correction method for the flat region in this case.

[0050] Figure 12B This is a second example of a rising-flat-rising pattern with 5 samples representing a flat region, and the correction method for the flat region in this case.

[0051] Figure 12C This is a second example of a falling-flattening-rising pattern with 5 samples representing a flat region, and the correction method for the flat region in this case.

[0052] Figure 12D This is a second example of a falling-flat-falling pattern with 5 samples representing a flat region, and the correction method for the flat region in this case.

[0053] Figure 13 It is a waveform diagram of the requantization error signal of a digital music signal recorded by quantizing a performance of a specified piece of music in a 24-bit quantization format.

[0054] Figure 14 This is a waveform diagram showing the requantization error signal of a digital music signal, which is... Figure 13The digital music signal of the same music is obtained by undergoing bit-expansion processing with a quantization bit depth of 24 bits in one or more embodiments of the bit-expansion processing apparatus.

[0055] Figure 15 This is a block diagram illustrating an example configuration of a computer executing a bit extension processor according to one or more embodiments.

[0056] Figure 16 This is a flowchart illustrating the operation of a bit extension processing apparatus according to one or more embodiments, the processing of a bit extension processing method according to one or more embodiments, and the processing of causing a computer to execute a bit extension processing program according to one or more embodiments. Detailed Implementation

[0057] Hereinafter, a bit extension processing apparatus, a bit extension processing method, and a bit extension processing program according to one or more embodiments will be described with reference to the accompanying drawings.

[0058] Figure 1 The bit extension processing apparatus 100 of one or more embodiments shown performs bit extension processing on an input digital music signal quantized with an input quantization bit depth of m bits (first quantization bit depth) and outputs a digital music signal quantized with a quantization bit depth of n bits (second quantization bit depth), where m and n are positive integers and n > m. As an example, the sample frequency (fs) of the input digital music signal is 192 kHz and the quantization bit depth is 16 bits, and the sample frequency of the output digital music signal is 192 kHz and the quantization bit depth is 24 bits.

[0059] The bit extension processing device 100 includes: a framing processing unit 1, a differential signal calculation unit 2, a flat region detection unit 3, a flat region correction unit 4, a differential signal averaging unit 5, a requantization error generation unit 6, an adder unit 7, and a delay unit 8.

[0060] The framing processing unit 1 divides and frames the samples input sequentially in the digital music signal according to a predetermined time (i.e., a predetermined number of samples). For example, the framing processing unit 1 divides a series of samples of the input digital music signal into 128 samples to form a frame containing 128 samples. The differential signal calculation unit 2 calculates the difference between the current sample and the previous sample of each sample input sequentially in each frame as the differential signal (first differential signal). To distinguish between the samples of the input digital music signal and the samples of the differential signal, it is assumed that the samples of the differential signal are called differential samples.

[0061] use Figure 2 The operation of the differential signal calculation unit 2 and the flat region detection unit 3 is explained in detail. Figure 2In the diagram, (a) represents an example of the variation in sample values ​​of samples S0 to S8 of the digital music signal provided from the framing processing unit 1 to the differential signal calculation unit 2. Assume that the sample values ​​of samples S0 to S8 are 0, 1, 4, 5, 6, 7, 3, 2, and 4, respectively.

[0062] like Figure 2 As shown in (b), the differential signal calculation unit 2 calculates the difference value with the previous samples S0 to S7 when each of the samples S1 to S8 is used as the current sample. Figure 2 In the waveform shown in (a), the difference values ​​1, 3, 1, 1, 1, -4, -1, 2 are obtained corresponding to samples S1 to S8. Figure 2 (c) indicates based on Figure 2 The difference signal shown in (b) is a difference value. The difference signal calculated by the difference signal calculation unit 2 is a numerical sequence of difference values ​​obtained corresponding to each sample. Figure 2 The differential signal shown in (c) contains differential samples D1 to D8.

[0063] Furthermore, if sample S0 is the first sample in a frame, then there is no preceding sample to sample S0. As described later, the differential signal calculated by the differential signal calculation unit 2 is provided to the flat region detection unit 3, which detects flat regions in the differential signal where the difference values ​​(sample values ​​of the differential samples) are the same. If the differential signal calculation unit 2 incorrectly outputs a difference value at the time of sample S0 that is the same as the difference value obtained at the time of sample S1, the flat region detection unit 3 will falsely detect a flat region. Therefore, the differential signal calculation unit 2 can be configured to output a maximum value that will not be generated after sample S1 as the difference value at the time of the first sample S0.

[0064] The differential signal calculation unit 2 may also omit the differential value in the initial sample S0, and calculate the differential value after sample S1. In one or more embodiments, such as... Figure 2 As shown in (c), the differential signal calculation unit 2 outputs the differential signal formed by the differential sample after sample S1.

[0065] The differential signal calculated by the differential signal calculation unit 2 is provided to the flat region detection unit 3. The flat region detection unit 3 detects the start and end positions of two or more consecutive flat regions with the same differential value in the differential signal, as well as the number of samples in the flat region (the number of consecutive samples with the same value). Figure 2 In the example shown in (c), the flat region detection unit 3 detects the differential sample D3 as the start position of the flat region, detects the differential sample D5 as the end position of the flat region, and detects the number of samples in the flat region as 3.

[0066] Figure 3This represents the requantization error signal obtained from two digital music signals. These two signals were obtained by quantizing and recording a single performance of a given piece of music at the same time using the same equipment, at a sample frequency of 192kHz, and with quantization bits of 16 bits and 24 bits respectively. The calculation is as follows. Figure 3 The requantization error signal is shown.

[0067] To compare a first digital music signal with 16 quantization bits with a second digital music signal with 24 quantization bits, the second digital music signal is multiplied by 1 / 256, and the lower 8 bits of the second digital music signal are represented by a decimal point. The beginning positions of the first digital music signal and the second digital music signal (represented by the decimal point in the lower 8 bits) are aligned, and the difference between the two sample values ​​is calculated as the requantization error for each sample.

[0068] When the first digital music signal with a quantization bit depth of 16 bits is set as M1, the second digital music signal with a quantization bit depth of 24 bits is set as M2, the requantization error signal is set as ERROR (a decimal value), and i is set as the sample number, the requantization error signal ERROR[i] in each sample number is represented by equation (1). If equation (1) is rewritten, it becomes equation (2) and equation (3). In equation (3), ROUND represents rounding to the nearest decimal.

[0069] ERROR[i]=M2[i] / 256-M1[i]……(1)

[0070] M2[i] / 256=M1[i]+ERROR[i]……(2)

[0071] M2[i]=ROUND[(M1[i]×256)+(ERROR[i]×256)]……(3)

[0072] According to equation (3), if a requantization error signal is added to the first digital music signal M1 quantized with 16 bits, the second digital music signal M2 quantized with 24 bits can be reproduced. Here, the coefficient of M2 divided by equation (1) is determined based on the difference in the number of quantization bits between the two signals (nm).

[0073] Figure 4 The difference between two adjacent samples in the first digital music signal M1 is shown. Assuming that the difference between the current sample pcm[i] and the previous sample pcm[i-1] in the first digital music signal M1 is DIFF(M1)[i], then DIFF(M1)[i] is represented by equation (4).

[0074] DIFF(M1)[i]=pcm[i]-pcm[i-1]……(4)

[0075] Compare Figure 3 and Figure 4 It can be seen that the requantization error signals of the first digital music signal M1 and the second digital music signal M2 are highly correlated with the difference values ​​between two adjacent samples in the first digital music signal M1. However, in Figure 3 The requantization error signal does not exist in Figure 4 The flat waveform portion with the same difference value, enclosed by a dashed ellipse, is shown in the image. Correcting this flat waveform portion to approximate the requantization error signal can further improve the correlation between the requantization error signal and the difference value.

[0076] Return to Figure 1 The flat region correction unit 4 provides the differential signal calculated by the differential signal calculation unit 2, the start and end positions of the flat region detected by the flat region detection unit 3, and the number of samples in the flat region to the flat region correction unit 4. The flat region correction unit 4 corrects the flat region of the differential signal in a way that makes the waveform of the differential signal approximate the waveform of the requantization error signal.

[0077] use Figures 5A to 5D , Figures 6A to 6D , Figures 7A to 7D , Figures 8A to 8D This illustrates the first example of a flat region correction method performed by the flat region correction unit 4 when the number of samples in the flat region is 2 to 5. In each figure, the dashed circle represents the difference sample before correction, and the solid circle represents the difference sample after correction. Figures 5A to 5D , Figures 6A to 6D , Figures 7A to 7D , Figures 8A to 8D This indicates the sample number of a flat region with a sample number of 2 to 5 when the digital music signal of the specified piece of music is input into the local expansion processing device 100.

[0078] exist Figure 5A In the differential signal shown, the two differential samples D74 and D75 with sample numbers 74 and 75 form a flat region. Figure 5A The diagram illustrates a pattern where the difference value increases from the preceding sample (73) with a flat region to the difference sample (D74), and decreases from the difference sample (D75) to the following sample (76) with a flat region. This pattern, where the difference value increases before the flat region and decreases after the flat region, is called the rise-flatten-fall pattern. The uncorrected difference values ​​for difference samples D73 to D76 are 2, 3, 3, and 2, respectively.

[0079] The flat region correction unit 4 adds a correction value of 0.5 to the differential sample D74, correcting the differential value to 3.5, and subtracts the correction value of 0.5 from the differential sample D75, correcting the differential value to 2.5. Figure 5A In the diagram, difference samples D74 and D75 are the difference samples to be corrected. Difference sample D74 is corrected to a corrected difference sample D74' with a difference value larger than that of difference sample D74. Difference sample D75 is corrected to a corrected difference sample D75' with the difference value between difference sample D75 and difference sample D76.

[0080] Adding the correction value of 0.5 is equivalent to treating the current difference sample as a value advanced towards the positive side and adding a decimal part. Subtracting the correction value of 0.5 is equivalent to treating the current difference sample as a value delayed towards the negative side and adding a decimal part.

[0081] The flat region correction unit 4, as described above, deflates the flat regions of sample number 2 in the differential signal that have a rising-flat-falling pattern.

[0082] exist Figure 5B In the differential signal shown, two differential samples D999 and D1000 with sample numbers 999 and 1000 form a flat region. Figure 5B This indicates a pattern where the difference value increases from the previous sample number 998 (with a flat region) to the difference sample D999, and then increases from the difference sample D1000 to the next sample number 1001 (with a flat region). This pattern, where the difference value increases before and after the flat region, is called the rise-flatten-rise pattern. The uncorrected difference values ​​for samples D998 to D1001 are 0, 3, 3, and 4, respectively.

[0083] The flat region correction unit 4 subtracts a correction value of 0.5 from the differential sample D999 to correct the differential value to 2.5, and adds a correction value of 0.5 to the differential sample D1000 to correct the differential value to 3.5. Figure 5B In the diagram, difference samples D999 and D1000 are the difference samples to be corrected. Difference sample D999 is corrected to a corrected difference sample D999' having the difference value between difference sample D998 and difference sample D999. Difference sample D1000 is corrected to a corrected difference sample D1000' having the difference value between difference sample D1000 and difference sample D1001.

[0084] The flat region correction unit 4, as described above, deflates the flat regions of sample number 2 in the differential signal that have a rising-flat-rising pattern.

[0085] exist Figure 5CIn the differential signal shown, two differential samples, D168 and D169, with sample numbers 168 and 169, form a flat region. Figure 5C This indicates a pattern where the difference value decreases from the preceding sample (D167, which has a flat region) to the next sample (D168, which has a flat region), and increases from the preceding sample (D169, which has a flat region) to the next sample (D170, which has a flat region). This pattern, where the difference value decreases before the flat region and increases after the flat region, is called a decreasing-flattening-increasing pattern. The uncorrected difference values ​​for samples D167 to D170 are 3, 2, 2, and 4, respectively.

[0086] The flat region correction unit 4 subtracts a correction value of 0.5 from the differential sample D168 to correct the differential value to 1.5, and adds a correction value of 0.5 to the differential sample D169 to correct the differential value to 2.5. Figure 5C In the diagram, difference samples D168 and D169 are the difference samples to be corrected. Difference sample D168 is corrected to a corrected difference sample D168' with a smaller difference value than difference sample D168. Difference sample D169 is corrected to a corrected difference sample D169' with the difference value between difference sample D169 and difference sample D170.

[0087] The flat region correction unit 4, as described above, deflates the flat regions of sample number 2 in the differential signal that have a falling-flat-rising pattern.

[0088] exist Figure 5D In the differential signal shown, two differential samples, D574 and D575, with sample numbers 574 and 575, form a flat region. Figure 5D This indicates a pattern where the difference value decreases from the preceding sample (573) with a flat region to the difference sample (D574), and from the difference sample (D575) to the following sample (D576) with a flat region. This pattern, where the difference value decreases before and after the flat region, is called a decreasing-flattening-decreasing pattern. The uncorrected difference values ​​for samples D573 to D576 are 0, -2, -2, and -4, respectively.

[0089] The flat region correction unit 4 subtracts a correction value of 0.5 from the differential sample D575, correcting the differential value to -2.5. Figure 5D In this context, difference sample D575 is the difference sample of the object to be corrected. Difference sample D575 is corrected to a corrected difference sample D575' having the difference value between difference sample D575 and difference sample D576.

[0090] The flat region correction unit 4, as described above, deflatens the flat regions of sample number 2 in the differential signal that have a falling-flat-falling pattern.

[0091] exist Figure 6A In the differential signal shown, the three differential samples D591 to D593 with sample numbers 591 to 593 form a flat region. Figure 6A This represents an ascending-flattening-descending pattern where the difference value rises from the previous sample number 590 (with a flat region) to the difference sample D591, and then falls from the difference sample D593 to the next sample number 594 (with a flat region). The uncorrected difference values ​​for samples D590 to D594 are -3, -2, -2, -2, and -3, respectively.

[0092] The flat region correction unit 4 adds a correction value of 0.5 to the differential sample D592 to correct the difference value to -1.5, and subtracts the correction value of 0.5 from the differential sample D593 to correct the difference value to -2.5. Figure 6A In the diagram, difference samples D592 and D593 are the difference samples to be corrected. Difference sample D592 is corrected to a corrected difference sample D592' with a larger difference value than difference sample D592. Difference sample D593 is corrected to a corrected difference sample D593' with the difference value between difference sample D593 and difference sample D594.

[0093] The flat region correction unit 4, as described above, deflates the flat regions of the differential signal with the sample number 3 having the rising-flat-falling pattern.

[0094] exist Figure 6B In the differential signal shown, the three differential samples D493 to D495 with sample numbers 493 to 495 form a flat region. Figure 6B This indicates an ascending-flattening-ascending pattern where the difference value rises from the previous sample number 492 (which has a flat region) to the difference sample D493, and the difference value rises from the difference sample D495 to the next sample number 496 (which has a flat region). The uncorrected difference values ​​of the difference samples D492 to D496 are -7, -6, -6, -6, and -5, respectively.

[0095] The flat region correction unit 4 subtracts a correction value of 0.5 from the difference sample D494 to correct the difference value to -6.5, and adds a correction value of 0.5 to the difference sample D495 to correct the difference value to -5.5. Figure 6BIn the diagram, difference samples D494 and D495 are the difference samples to be corrected. Difference sample D494 is corrected to become a corrected difference sample D494' having the difference value between difference sample D492 and difference sample D494. Difference sample D495 is corrected to become a corrected difference sample D495' having the difference value between difference sample D495 and difference sample D496.

[0096] The flat region correction unit 4, as described above, deflates the flat regions of the differential signal with the sample number 3 having the rising-flat-rising pattern.

[0097] exist Figure 6C In the differential signal shown, the three differential samples D314 to D316 with sample numbers 314 to 316 form a flat region. Figure 6C This represents a decreasing-flattening-increasing pattern, where the difference value decreases from the previous sample (313) with a flat region to the next sample (317) with a flat region, and then increases from the previous sample (316) to the next sample (317) with a flat region. The uncorrected difference values ​​for samples D313 to D317 are 9, 8, 8, 8, and 9, respectively.

[0098] The flat region correction unit 4 subtracts a correction value of 0.5 from the differential sample D314 to correct the differential value to 7.5, and adds a correction value of 0.5 to the differential sample D316 to correct the differential value to 8.5. Figure 6C In the diagram, difference samples D314 and D316 are the difference samples to be corrected. Difference sample D314 is corrected to a corrected difference sample D314' with a smaller difference value than difference sample D314. Difference sample D316 is corrected to a corrected difference sample D316' with the difference value between difference sample D316 and difference sample D317.

[0099] The flat region correction unit 4, as described above, deflates the flat regions of the differential signal with a falling-flat-rising pattern of 3 samples.

[0100] exist Figure 6D In the differential signal shown, the three differential samples D135 to D137 with sample numbers 135 to 137 form a flat region. Figure 6D This represents a decreasing-flattening-decreasing pattern, where the difference value decreases from the previous sample (D134) with a flat region to the next sample (D135), and from the previous sample (D137) to the next sample (D138) with a flat region. The uncorrected difference values ​​for samples D134 to D138 are 4, 3, 3, 3, and 2, respectively.

[0101] The flat region correction unit 4 adds a correction value of 0.5 to the differential sample D135 to correct the differential value to 3.5, and subtracts the correction value of 0.5 from the differential sample D137 to correct the differential value to 2.5. Figure 6D In the diagram, difference samples D135 and D137 are the difference samples to be corrected. Difference sample D135 is corrected to a corrected difference sample D135' having the difference value between difference sample D134 and difference sample D135. Difference sample D137 is corrected to a corrected difference sample D137' having the difference value between difference sample D136 and difference sample D138.

[0102] The flat region correction unit 4, as described above, deflates the flat regions of the differential signal with a falling-flat-falling pattern of 3 samples.

[0103] exist Figure 7A In the differential signal shown, the four differential samples D822 to D825 with sample numbers 822 to 825 form a flat region. Figure 7A This represents an ascending-flattening-descending pattern where the difference value rises from the previous sample number 821 (with a flat region) to the difference sample D822, and then falls from the difference sample D825 to the next sample number 826 (with a flat region). The uncorrected difference values ​​for samples D821 and D826 are both 7, while the uncorrected difference values ​​for samples D822 to D825 are 8.

[0104] The flat region correction unit 4 subtracts a correction value of 0.5 from the difference sample D823 to correct the difference value to 7.5, and adds a correction value of 0.5 to the difference sample D824 to correct the difference value to 8.5. Additionally, the flat region correction unit 4 subtracts a correction value of 0.5 from the difference sample D825. Figure 7A In the text, the differential samples D823 to D825 are the differential samples of the calibration object.

[0105] Differential sample D823 is corrected to a corrected differential sample D823' having the difference between differential sample D821 and differential sample D823. Differential sample D824 is corrected to a corrected differential sample D824' having a larger difference than differential sample D824. Differential sample D825 is corrected to a corrected differential sample D825' having the difference between differential sample D825 and differential sample D826.

[0106] The flat region correction unit 4, as described above, deflates the flat regions of the differential signal with the rising-flat-falling pattern of sample number 4.

[0107] exist Figure 7BIn the differential signal shown, the four differential samples D655 to D658 with sample numbers 655 to 658 form a flat region. Figure 7B This indicates an ascending-flattening-ascending pattern where the difference value increases from the previous sample number 654 (with a flat region) to the difference sample D655, and from the difference sample D658 to the next sample number 659 (with a flat region). The uncorrected difference values ​​for difference samples D654 and D659 are 7 and 11, respectively, while the uncorrected difference values ​​for difference samples D655 to D658 are 9.

[0108] The flat region correction unit 4 subtracts a correction value of 0.5 from the difference sample D655 to correct the difference value to 8.5, and subtracts a correction value of 0.5 from the difference sample D657 to correct the difference value to 8.5. Additionally, the flat region correction unit 4 adds a correction value of 0.5 to the difference sample D658 to correct the difference value to 9.5. Figure 7B In the diagram, the differential samples D655, D657, and D658 are the differential samples of the calibration object.

[0109] Differential sample D655 is corrected to a corrected differential sample D655' having the difference value between differential sample D654 and differential sample D655. Differential sample D657 is corrected to a corrected differential sample D657' having the difference value between differential sample D654 and differential sample D657. Differential sample D658 is corrected to a corrected differential sample D658' having the difference value between differential sample D658 and differential sample D659.

[0110] The flat region correction unit 4, as described above, deflates the flat regions of the differential signal with a rising-flat-rising pattern of 4 samples.

[0111] exist Figure 7C In the differential signal shown, the four differential samples D253 to D256 with sample numbers 253 to 256 form a flat region. Figure 7C This represents a decreasing-flattening-increasing pattern, where the difference value decreases from the previous sample number 252 (with a flat region) to the difference sample D253, and the difference value increases from the difference sample D256 to the next sample number 257 (with a flat region). The uncorrected difference values ​​for difference samples D252 and D257 are both 12, and the uncorrected difference values ​​for difference samples D253 to D256 are 11.

[0112] The flat region correction unit 4 subtracts a correction value of 0.5 from the difference sample D253 to correct the difference value to 10.5, and subtracts a correction value of 0.5 from the difference sample D255 to correct the difference value to 10.5. Additionally, the flat region correction unit 4 adds a correction value of 0.5 to the difference sample D256 to correct the difference value to 11.5. Figure 7C In the diagram, difference samples D253, D255, and D256 are the difference samples of the calibration object.

[0113] Differential sample D253 is corrected to a corrected differential sample D253' with a smaller difference value than differential sample D253. Differential sample D255 is corrected to a corrected differential sample D255' with a smaller difference value than differential sample D255. Differential sample D256 is corrected to a corrected differential sample D256' with the difference value between differential sample D256 and differential sample D257.

[0114] The flat region correction unit 4, as described above, deflates the flat regions of the sample number 4 in the differential signal that have a falling-flat-rising pattern.

[0115] exist Figure 7D In the differential signal shown, the four differential samples D342 to D345 with sample numbers 342 to 345 form a flat region. Figure 7D This represents a decreasing-flattening-decreasing pattern, where the difference value decreases from the previous sample number 341 (with a flat region) to the difference sample D342, and the difference value decreases from the difference sample D345 to the next sample number 346 (with a flat region). The uncorrected difference values ​​for difference samples D341 and D346 are 12 and 10, respectively, while the uncorrected difference values ​​for difference samples D342 to D345 are 11.

[0116] The flat region correction unit 4 adds a correction value of 0.5 to the differential sample D343 to correct the differential value to 11.5, and subtracts the correction value of 0.5 from the differential sample D345 to correct the differential value to 10.5. Figure 7D In the diagram, difference samples D343 and D345 are the difference samples to be corrected. Difference sample D343 is corrected to a corrected difference sample D343' having the difference value between difference sample D341 and difference sample D343. Difference sample D345 is corrected to a corrected difference sample D345' having the difference value between difference sample D345 and difference sample D346.

[0117] The flat region correction unit 4, as described above, deflates the flat regions of the differential signal with a falling-flat-falling pattern of 4 samples.

[0118] exist Figure 8AIn the differential signal shown, the five differential samples D282 to D286 with sample numbers 282 to 286 form a flat region. Figure 8A This represents an ascending-flattening-descending pattern where the difference value rises from the previous sample number 281 (with a flat region) to the difference sample D282, and then falls from the difference sample D286 to the next sample number 287 (with a flat region). The uncorrected difference values ​​for samples D281 and D287 are both -7, while the uncorrected difference values ​​for samples D282 to D286 are -6.

[0119] The flat region correction unit 4 subtracts a correction value of 0.5 from the differential sample D283 to correct the difference value to -6.5, and adds a correction value of 0.5 to the differential sample D285 to correct the difference value to -5.5. Figure 8A In the diagram, difference samples D283 and D285 are the difference samples to be corrected. Difference sample D283 is corrected to become a corrected difference sample D283' with the difference value between difference sample D281 and difference sample D283. Difference sample D285 is corrected to become a corrected difference sample D285' with a larger difference value than difference sample D285.

[0120] The flat region correction unit 4, as described above, deflates the flat regions of the sample number 5 in the differential signal that have a rising-flat-falling pattern.

[0121] exist Figure 8B In the differential signal shown, the five differential samples D446 to D450 with sample numbers 446 to 450 form a flat region. Figure 8B This represents an ascending-flattening-ascending pattern where the difference value rises from the previous sample number 445 (with a flat region) to the difference sample D446, ​​and from the difference sample D450 to the next sample number 451 (with a flat region). The uncorrected difference values ​​for difference samples D445 and D451 are -7 and -5, respectively, while the uncorrected difference values ​​for difference samples D446 to D450 are -6.

[0122] The flat region correction unit 4 subtracts a correction value of 0.5 from the difference sample D447 to correct the difference value to -6.5, and subtracts a correction value of 0.5 from the difference sample D449 to correct the difference value to -6.5. Figure 8B In the diagram, difference samples D447 and D449 are the difference samples to be corrected. Difference sample D447 is corrected to a corrected difference sample D447' having the difference value between difference sample D445 and difference sample D447. Difference sample D449 is corrected to a corrected difference sample D449' having the difference value between difference sample D445 and difference sample D449.

[0123] The flat region correction unit 4, as described above, deflates the flat regions of the differential signal with the sample number 5 having the rising-flat-rising pattern.

[0124] exist Figure 8C In the differential signal shown, the five differential samples D618 to D622 with sample numbers 618 to 622 form a flat region. Figure 8C This represents a decreasing-flattening-increasing pattern, where the difference value decreases from the previous sample number 617 (with a flat region) to the difference sample D618, and the difference value increases from the difference sample D622 to the next sample number 623 (with a flat region). The uncorrected difference values ​​for samples D617 and D623 are both -5, while the uncorrected difference values ​​for samples D618 to D622 are -6.

[0125] The flat region correction unit 4 adds a correction value of 0.5 to the differential sample D619 to correct the differential value to -5.5, and subtracts the correction value of 0.5 from the differential sample D621 to correct the differential value to -6.5. Figure 8C In the diagram, difference samples D619 and D621 are the difference samples to be corrected. Difference sample D619 is corrected to a corrected difference sample D619' having the difference value between difference sample D617 and difference sample D619. Difference sample D621 is corrected to a corrected difference sample D621' having a smaller difference value than difference sample D621.

[0126] The flat region correction unit 4, as described above, deflates the flat regions of the differential signal with a falling-flat-rising pattern of 5 samples.

[0127] exist Figure 8D In the differential signal shown, the five differential samples D875 to D879 with sample numbers 875 to 879 form a flat region. Figure 8D This represents a decreasing-flattening-decreasing pattern, where the difference value decreases from the previous sample number 874 (with a flat region) to the difference sample D875, and the difference value decreases from the difference sample D879 to the next sample number 880 (with a flat region). The uncorrected difference values ​​for difference samples D874 and D880 are -5 and -7, respectively, while the uncorrected difference values ​​for difference samples D875 to D879 are -6.

[0128] The flat region correction unit 4 adds a correction value of 0.5 to the differential sample D876, correcting the differential value to -5.5, and adds a correction value of 0.5 to the differential sample D878, correcting the differential value to -5.5. Figure 8DIn the diagram, difference samples D876 and D878 are the difference samples to be corrected. Difference sample D876 is corrected to a corrected difference sample D876' having the difference value between difference sample D874 and difference sample D876. Difference sample D878 is corrected to a corrected difference sample D878' having the difference value between difference sample D874 and difference sample D878.

[0129] The flat region correction unit 4, as described above, deflatens the flat regions of the differential signal with a falling-flat-falling pattern of 5 samples.

[0130] However, the flat regions generated in the differential signal calculated based on the digital music signal are at most 5 samples, and flat regions with more than 6 samples are almost never generated. Therefore, the flat region correction unit 4 only needs to be compatible with... Figures 5A to 5D The sample size shown is 2. Figures 6A to 6D The sample size shown is 3. Figures 7A to 7D The sample size shown is 4. Figures 8A to 8D For each of the 5 sample sizes shown, the position for correcting the differential samples in the flat area and the correction value for performing addition or subtraction operations can be set accordingly.

[0131] In the flat region correction unit 4, up to multiple sampling numbers, the position and correction value of the differential samples in the flat region are set arbitrarily.

[0132] The method for correcting the flat area of ​​the flat area correction unit 4 is not limited to Figures 5A to 5D , Figures 6A to 6D , Figures 7A to 7D , Figures 8A to 8D The first example shown. The flat region correction unit 4 can also correct the flat region by using the center point of the flat region as a reference, decreasing the difference value of the difference sample at the beginning of the flat region and increasing the difference value of the difference sample at the end of the flat region. Alternatively, the flat region correction unit 4 can also correct the flat region by using the center point of the flat region as a reference, increasing the difference value of the difference sample at the beginning of the flat region and decreasing the difference value of the difference sample at the end of the flat region. Figure 9 , Figure 10A , Figure 10B , Figures 11A to 11D , Figures 12A-12D This is a second example of a correction method for flat regions, illustrating a correction in a way that causes the flat region to rise or fall as a whole.

[0133] exist Figures 5A to 5C In the correction of the rising-flat-falling, rising-flat-rising, and falling-flat-rising patterns for sample number 2, the center point of the flat region is used as a reference to correct the flat region by making one rise and the other fall, or by making one fall and the other rise. Therefore, it is possible to... Figures 5A to 5C The correction method shown in the first example is directly used as the correction method in the second example.

[0134] In the second example, as a correction method for the decreasing-flattening-decreasing pattern with 2 samples, the flat region correction unit 4 is used instead. Figure 5D And use Figure 9 The correction method shown.

[0135] The flat region correction unit 4 adds a correction value of 0.5 to the differential sample D574 to correct the difference value to -1.5, and subtracts the correction value of 0.5 from the differential sample D575 to correct the difference value to -2.5. Figure 9 In the diagram, difference samples D574 and D575 are the difference samples to be corrected. Difference sample D574 is corrected to a corrected difference sample D574' having the difference value between difference sample D573 and difference sample D574. Difference sample D575 is corrected to a corrected difference sample D575' having the difference value between difference sample D575 and difference sample D576.

[0136] In the second example, as a correction method for the rising-flat-falling pattern and the rising-flat-rising pattern with a sample size of 3, the flat region correction unit 4 replaces... Figure 6A and Figure 6B And use Figure 10A and Figure 10B The correction method is shown. In Figure 6C as well as Figure 6D In the shown sample number 3, the decreasing-flattening-increasing pattern and the correction of the decreasing-flattening-decreasing pattern are corrected by using the center point of the flat region as a reference, and adjusting one aspect by increasing the other, or vice versa. Therefore, it is possible to... Figure 6C and Figure 6D The correction method shown in the first example is directly used as the correction method in the second example.

[0137] like Figure 10A As shown, in the rising-flattening-falling mode, the flat area correction unit 4 adds a correction value of 0.5 to the differential sample D591 to correct the differential value to -1.5, and subtracts the correction value of 0.5 from the differential sample D593 to correct the differential value to -2.5. Figure 10A In the diagram, difference samples D591 and D593 are the difference samples to be corrected. Difference sample D591 is corrected to a corrected difference sample D591' with a larger difference value than difference sample D591. Difference sample D593 is corrected to a corrected difference sample D593' with the difference value between difference sample D593 and difference sample D594.

[0138] like Figure 10BAs shown, in the rising-flattening-rising mode, the flat area correction unit 4 subtracts a correction value of 0.5 from the differential sample D493 to correct the differential value to -6.5, and adds a correction value of 0.5 to the differential sample D495 to correct the differential value to -5.5. Figure 10B In the diagram, difference samples D493 and D495 are the difference samples to be corrected. Difference sample D493 is corrected to become a corrected difference sample D493' having the difference value between difference sample D492 and difference sample D493. Difference sample D495 is corrected to become a corrected difference sample D495' having the difference value between difference sample D495 and difference sample D496.

[0139] In the second example, as a correction method for all patterns with a sample size of 4, the flat region correction unit 4 uses... Figures 11A to 11D The correction method shown is to replace Figures 7A to 7D .

[0140] like Figure 11A As shown, in the rising-flattening-falling mode, the flat area correction unit 4 adds a correction value of 0.5 to the difference sample D822 to correct the difference value to 8.5, and adds a correction value of 0.25 to the difference sample D823 to correct the difference value to 8.25. Additionally, the flat area correction unit 4 subtracts a correction value of 0.25 from the difference sample D824 to correct the difference value to 7.75, and subtracts a correction value of 0.5 from the difference sample D825 to correct the difference value to 7.5. Figure 11A In the text, the differential samples D822 to D825 are the differential samples of the calibration object.

[0141] Differential sample D822 is corrected to a corrected differential sample D822' with a larger difference value than differential sample D822. Differential sample D823 is corrected to a corrected differential sample D823' with the difference value between corrected differential sample D822' and differential sample D823. Differential sample D825 is corrected to a corrected differential sample D825' with the difference value between differential sample D825 and differential sample D826. Differential sample D824 is corrected to a corrected differential sample D824' with the difference value between differential sample D824 and corrected differential sample D825'.

[0142] Thus, the correction value added to the difference sample and the correction value subtracted from the difference sample are not limited to 0.5; 0.25 can also be used.

[0143] like Figure 11BAs shown, in the rising-flattening-rising mode, the flat area correction unit 4 subtracts a correction value of 0.5 from the difference sample D655 to correct the difference value to 8.5, and subtracts a correction value of 0.25 from the difference sample D656 to correct the difference value to 8.75. Additionally, the flat area correction unit 4 adds a correction value of 0.25 to the difference sample D657 to correct the difference value to 9.25, and adds a correction value of 0.5 to the difference sample D658 to correct the difference value to 9.5. Figure 11B In the text, the differential samples D655 to D658 are the differential samples of the calibration object.

[0144] Differential sample D655 is corrected to a corrected differential sample D655' having the difference value between differential sample D654 and differential sample D655. Differential sample D656 is corrected to a corrected differential sample D656' having the difference value between corrected differential sample D655' and differential sample D656. Differential sample D658 is corrected to a corrected differential sample D658' having the difference value between differential sample D658 and differential sample D659. Differential sample D657 is corrected to a corrected differential sample D657' having the difference value between differential sample D657 and corrected differential sample D658'.

[0145] like Figure 11C As shown, in the descending-flattening-ascending mode, the flat region correction unit 4 subtracts a correction value of 0.5 from the difference sample D253 to correct the difference value to 10.5, and subtracts a correction value of 0.25 from the difference sample D254 to correct the difference value to 10.75. Additionally, the flat region correction unit 4 adds a correction value of 0.25 to the difference sample D255 to correct the difference value to 11.25, and adds a correction value of 0.5 to the difference sample D256 to correct the difference value to 11.5. Figure 11C In the text, the differential samples D253 to D256 are the differential samples of the correction object.

[0146] Differential sample D253 is corrected to a corrected differential sample D253' with a smaller difference value than differential sample D253. Differential sample D254 is corrected to a corrected differential sample D254' with the difference value between corrected differential sample D253' and differential sample D254. Differential sample D256 is corrected to a corrected differential sample D256' with the difference value between differential sample D256 and differential sample D257. Differential sample D255 is corrected to a corrected differential sample D255' with the difference value between differential sample D255 and corrected differential sample D256'.

[0147] like Figure 11DAs shown, in the falling-flattening-falling mode, the flat region correction unit 4 adds a correction value of 0.5 to the difference sample D342 to correct the difference value to 11.5, and adds a correction value of 0.25 to the difference sample D343 to correct the difference value to 11.25. Additionally, the flat region correction unit 4 subtracts a correction value of 0.25 from the difference sample D344 to correct the difference value to 10.75, and subtracts a correction value of 0.5 from the difference sample D345 to correct the difference value to 10.5. Figure 11D In the text, the differential samples D342 to D345 are the differential samples of the correction object.

[0148] Differential sample D342 is corrected to a corrected differential sample D342' having the difference value between differential sample D341 and differential sample D342. Differential sample D343 is corrected to a corrected differential sample D343' having the difference value between corrected differential sample D342' and differential sample D343. Differential sample D345 is corrected to a corrected differential sample D345' having the difference value between differential sample D345 and differential sample D346. Differential sample D344 is corrected to a corrected differential sample D344' having the difference value between differential sample D344 and corrected differential sample D345'.

[0149] In the second example, as a correction method for all patterns with a sample size of 5, the flat region correction unit 4 uses... Figures 12A-12D The correction method shown is to replace Figures 8A to 8D .

[0150] like Figure 12A As shown, in the rising-flattening-falling mode, the flat area correction unit 4 adds a correction value of 0.5 to the differential sample D282, correcting the differential value to -5.5, and adds a correction value of 0.25 to the differential sample D283, correcting the differential value to -5.75. Additionally, the flat area correction unit 4 subtracts a correction value of 0.25 from the differential sample D285, correcting the differential value to -6.25, and subtracts a correction value of 0.5 from the differential sample D286, correcting the differential value to -6.5. Figure 12A In the diagram, the differential samples D282, D283, D285, and D286 are the differential samples of the calibration object.

[0151] Differential sample D282 is corrected to a corrected differential sample D282' with a larger difference value than differential sample D282. Differential sample D283 is corrected to a corrected differential sample D283' with the difference value between corrected differential sample D282' and differential sample D283. Differential sample D286 is corrected to a corrected differential sample D286' with the difference value between differential sample D286 and differential sample D287. ​​Differential sample D285 is corrected to a corrected differential sample D285' with the difference value between differential sample D285 and corrected differential sample D286'.

[0152] like Figure 12B As shown, in the rising-flattening-rising mode, the flat area correction unit 4 subtracts a correction value of 0.5 from the difference sample D446 to correct the difference value to -6.5, and subtracts a correction value of 0.25 from the difference sample D447 to correct the difference value to -6.25. Additionally, the flat area correction unit 4 adds a correction value of 0.25 to the difference sample D449 to correct the difference value to -5.75, and adds a correction value of 0.5 to the difference sample D450 to correct the difference value to -5.5. Figure 12B In the diagram, the differential samples D446, ​​D447, D449, and D450 are the differential samples of the calibration target.

[0153] Differential sample D446 is corrected to a corrected differential sample D446' having the difference value between differential sample D445 and differential sample D446. ​​Differential sample D447 is corrected to a corrected differential sample D447' having the difference value between corrected differential sample D446' and differential sample D447. Differential sample D450 is corrected to a corrected differential sample D450' having the difference value between differential sample D450 and differential sample D451. Differential sample D449 is corrected to a corrected differential sample D449' having the difference value between differential sample D449 and corrected differential sample D450'.

[0154] like Figure 12C As shown, in the descending-flattening-ascending mode, the flat region correction unit 4 subtracts a correction value of 0.5 from the difference sample D618 to correct the difference value to -6.5, and subtracts a correction value of 0.25 from the difference sample D619 to correct the difference value to -6.25. Additionally, the flat region correction unit 4 adds a correction value of 0.25 to the difference sample D621 to correct the difference value to -5.75, and adds a correction value of 0.5 to the difference sample D622 to correct the difference value to -5.5. Figure 12C In the diagram, differential samples D618, D619, D621, and D622 are differential samples of the calibration target.

[0155] Differential sample D618 is corrected to a corrected differential sample D618' with a smaller difference value than differential sample D618. Differential sample D619 is corrected to a corrected differential sample D619' with the difference value between corrected differential sample D618' and differential sample D619. Differential sample D622 is corrected to a corrected differential sample D622' with the difference value between differential sample D622 and differential sample D623. Differential sample D621 is corrected to a corrected differential sample D621' with the difference value between differential sample D621 and corrected differential sample D622'.

[0156] like Figure 12DAs shown, in the falling-flattening-falling mode, the flat region correction unit 4 adds a correction value of 0.5 to the differential sample D875 to correct the differential value to -5.5, and adds a correction value of 0.25 to the differential sample D876 to correct the differential value to -5.75. Additionally, the flat region correction unit 4 subtracts a correction value of 0.25 from the differential sample D878 to correct the differential value to -6.25, and subtracts a correction value of 0.5 from the differential sample D879 to correct the differential value to -6.5. Figure 12D In the diagram, the differential samples D875, D876, D878, and D879 are the differential samples of the calibration target.

[0157] Differential sample D875 is corrected to a corrected differential sample D875' having the difference value between differential sample D874 and differential sample D875. Differential sample D876 is corrected to a corrected differential sample D876' having the difference value between corrected differential sample D875' and differential sample D876. Differential sample D879 is corrected to a corrected differential sample D879' having the difference value between differential sample D879 and differential sample D880. Differential sample D878 is corrected to a corrected differential sample D878' having the difference value between differential sample D878 and corrected differential sample D879'.

[0158] In the first and second examples described above, the flat region correction unit 4 needs to deflatten the flat region so that adjacent difference samples in the flat region do not have the same difference value after correction. Deflating the flat region by the flat region correction unit 4 means correcting the difference samples in the flat region without leaving two or more remaining flat regions.

[0159] In the second example, when the number of differential signals is 3 and 5, and there is a central differential sample in the flat region, the flat region correction unit 4 can set the correction value of the central differential sample to 0.

[0160] In the second example, if the differential signal in the rising-flat-falling mode and the differential signal in the falling-flat-rising mode are waveforms that are symmetrical in the time axis direction, then the flat region correction unit 4 can also make the correction method for the flat region symmetrical. Furthermore, if the differential signal in the rising-flat-rising mode and the differential signal in the falling-flat-falling mode are waveforms that are symmetrical in the time axis direction, then the flat region correction unit 4 can also make the correction method for the flat region symmetrical.

[0161] return Figure 1The differential signal (second differential signal) after the flat region is deflattened by the flat region correction unit 4 is provided to the differential signal averaging unit 5. The differential signal averaging unit 5 sums the values ​​of the differential samples contained in one frame in each frame and divides them by the number of differential samples contained in one frame to calculate the differential average value.

[0162] If the differential average value is zero, then the change on the positive side of the differential signal is equal to the change on the negative side in each frame. The differential signal averaging unit 5 directly provides the differential signal provided by the flat region correction unit 4 as the third differential signal to the requantization error generation unit 6.

[0163] When the differential average value is a predetermined value that is non-zero, the differential signal averaging unit 5 subtracts the differential average value from the differential value of each differential sample of the differential signal. As a result, the differential average value of the differential signal becomes zero, enabling the change on the positive side of the differential signal to be equal to the change on the negative side within each frame. The differential signal averaging unit 5 provides the corrected differential signal (third differential signal) to the requantization error generation unit 6, ensuring that the change on the positive side of the differential signal is equal to the change on the negative side.

[0164] exist Figure 1 In this context, a first digital music signal with a quantization bit depth of 16 bits and a second digital music signal with a quantization bit depth of 24 bits are used as the objects. Therefore, the requantization error generation unit 6 multiplies the input differential signal by a coefficient to generate a requantization error signal that should be added to 8 bits (nm). If the requantization error signal that should be added to 8 bits is set as Qerror[i], and the difference value of the input differential signal is set as diff[i], then the requantization error signal Qerror[i] is represented by equation (5).

[0165] Qerror[i]=(512 / 131070)·diff[i]……(5)

[0166] The requantization error signal needs to be represented using 8 bits plus a coding bit (1 bit), so there are 0 to 255 on the positive side and -1 to -256 on the negative side, for a total of 512 levels. Therefore, the difference value is divided into 512 levels.

[0167] By subtracting -32768, the maximum value of the negative 16-bit quantization, from the maximum value of the positive 16-bit quantization (32767), the maximum difference component of the first digital music signal with 16-bit quantization becomes 65535 when it changes from negative to positive.

[0168] Therefore, it is possible to generate a maximum value on the positive side and a maximum value on the negative side, so the variation amplitude of the difference component becomes 131070 according to |65535-(-65535)|.

[0169] The coefficient multiplied by the difference value diff[i] in equation (5) is the reciprocal of the segment amplitude obtained by dividing the variation amplitude of the difference component by the number of segments. Furthermore, the coefficient multiplied by the difference value diff[i] is determined based on the quantization bit depth of the first digital music signal and the bit extension degree of the first digital music signal (i.e., the value of (nm)).

[0170] The requantization error generation unit 6 provides the requantization error signal generated as described above to the addition unit 7. The delay unit 8 delays the framed digital music signal output from the framing processing unit 1 by one sample and provides the result to the addition unit 7. This is because, as described above, the difference value is not calculated in the first sample of each frame.

[0171] As shown in equation (3), the adder 7 generates an 8-bit space by multiplying the input digital music signal by 256, and then adds a requantization error signal. The adder 7 rounds the value after the decimal point and expands the output bits to a 24-bit digital music signal.

[0172] Figure 13 Indicates that will be with Figure 3 The requantization error signal is obtained from two digital music signals that were quantized and recorded at the same time using the same equipment, with a sample frequency of 192kHz and two quantization bit depths of 16 bits and 24 bits, respectively, based on different prescribed musical pieces.

[0173] exist Figure 13 In, with Figure 3 Similarly, to compare a first digital music signal with 16 quantization bits with a second digital music signal with 24 quantization bits, the second digital music signal is multiplied by 1 / 256, and the lower 8 bits of the second digital music signal are represented with a decimal point. The beginning positions of the first digital music signal and the second digital music signal (represented by the decimal point in the lower 8 bits) are aligned, and the difference between the two sample values ​​is calculated as the requantization error for each sample.

[0174] Figure 14 The quantization error signal of the second digital music signal is shown, which is obtained by... Figure 1 The bit extension processing device 100 shown will be used as a... Figure 13The first digital music signal of the same piece of music was obtained by bit-expanding the quantization bit depth to 24 bits. The second digital music signal was multiplied by 1 / 256, and the lower 8 bits of the second digital music signal were represented by a decimal point. The beginning positions of the first digital music signal and the second digital music signal (represented by the decimal point in the lower 8 bits) were aligned, and the difference between the two sample values ​​was calculated as the requantization error for each sample.

[0175] Compare Figure 13 The requantization error signal shown and Figure 14 As can be seen from the requantization error signals shown, the two are extremely similar. High-quality bit-expansion processing of digital music signals can be performed using the bit-expansion processing device 100 and the bit-expansion processing method executed by it. This effect can be obtained regardless of the type of music. Even if the correction methods for the differential samples that deflatten flat regions differ, as long as the flat regions that do not originally exist are deflattened, the correlation between the requantization error signal and the differential value can be improved. Therefore, the correction method for the differential samples is not limited.

[0176] Figure 1 The bit extension processing device 100 shown can be constructed by hardware using circuitry or by software. Figure 1 The bit extension processing device 100 shown can be implemented by executing a bit extension processing program through the central processing unit (CPU) of a microcomputer.

[0177] exist Figure 15 In this configuration, CPU 10, main memory 15, and storage medium 20 are connected via a bus. Storage medium 20 can be any non-transitory storage medium such as a hard disk drive, optical disk, or semiconductor memory. Bit extension processing programs are stored in storage medium 20. These bit extension processing programs can also be sent from an external server via communication lines such as the Internet and stored in storage medium 20.

[0178] CPU 10 loads the bit extension processing program stored in storage medium 20 into main memory 15. CPU 10 executes the instructions described in the bit extension processing program loaded into main memory 15. Figure 16 The processing shown.

[0179] use Figure 16 The flowchart shown illustrates the processing that the bit extension handler causes CPU10 to execute. Figure 16 It shows Figure 1 The operation of the bit extension processing device 100 shown and the processing of the bit extension processing method executed by the bit extension processing device 100.

[0180] exist Figure 16In step S1, CPU 10 frames the input first digital music signal. In step S2, CPU 10 calculates the differential signal within the frame. In step S3, CPU 10 detects the start position, end position, and number of differential samples of the flat region in the differential signal. In step S4, CPU 10 deflattens the flat region according to the pattern of the differential signal. In step S5, CPU 10 calculates the average differential value within the frame.

[0181] In step S6, CPU 10 determines whether the differential average value is zero. If the differential average value is zero (yes), CPU 10 transfers the processing to step S8. If the differential average value is not zero ("no"), CPU 10 corrects the differential signal in step S7 so that the change on the positive side of the differential signal is equal to the change on the negative side, and transfers the processing to step S8.

[0182] In step S8, CPU 10 generates the quantization error within the frame. In step S9, CPU 10 adds the quantization error to the first digital music signal in the frame and outputs the result as the second digital music signal.

[0183] In step S10, CPU 10 determines whether the first digital music signal has been continuously input. If the first digital music signal continues to be input (yes), CPU 10 repeats the processing of steps S1 to S10. If the first digital music signal is not continuously input (no), CPU 10 ends the processing.

[0184] This invention is not limited to one or more of the embodiments described above, and various modifications can be made without departing from the spirit of this invention.

[0185] This application claims priority to Japanese Patent Application No. 2021-101324, filed with the Japan Patent Office on June 18, 2021, the entire disclosure of which is incorporated herein by reference.

Claims

1. A bit extension processing apparatus, comprising: The framing processing unit divides the samples of the first digital music signal quantized with the first quantization bit depth into multiple sample numbers and frames them. The differential signal calculation unit calculates a first differential signal, which takes the difference between two adjacent samples of the first digital music signal contained in each frame as the differential sample. The flat region detection unit detects the start and end positions of the flat region in the first differential signal, as well as the number of differential samples in the flat region. The flat region is a region with two or more consecutive differential samples having the same sample value. The flat region correction unit, based on whether the sample value from the previous differential sample in the flat region to the differential sample at the starting position increases or decreases, and whether the sample value from the differential sample at the ending position to the next differential sample in the flat region increases or decreases, takes one or more differential samples in the flat region as differential samples to be corrected, and corrects the sample value of the correction object by adding or subtracting a correction value, thereby generating a second differential signal that deflattens the flat region; The differential signal averaging unit sums the sample values ​​of the second differential signal contained in each frame, divides them by the number of samples of the second differential signal contained in each frame to calculate the differential average value, and subtracts the differential average value from each sample value of the second differential signal to generate a third differential signal. The requantization error generation unit generates a requantization error signal for each frame based on the third differential signal. The requantization error signal is represented by the difference in the number of bits between the first quantization bit and the second quantization bit, wherein the second quantization bit is more than the first quantization bit. as well as The addition unit adds the requantization error signal to the first digital music signal contained in each frame, and outputs the second digital music signal as the second quantization bit depth.

2. A bit extension processing method, wherein, The samples of the first digital music signal, quantized with the first quantization bit depth, are divided into multiple sample numbers and then framed. Calculate the first differential signal, which takes the difference between two adjacent samples of the first digital music signal contained in each frame as the differential sample. The start and end positions of flat regions in the first differential signal, as well as the number of differential samples in the flat regions, are detected. A flat region is a region with two or more consecutive differential samples having the same sample value. Based on the pattern of whether the sample values ​​from the previous differential sample in the flat region to the starting differential sample, and whether the sample values ​​from the ending differential sample to the next differential sample in the flat region increase or decrease, one or more differential samples in the flat region are used as differential samples for correction. The sample values ​​of the correction targets are corrected by adding or subtracting correction values, thereby generating a second differential signal that deflattens the flat region. The sample values ​​of the second differential signal contained in each frame are summed, and the result is divided by the number of samples of the second differential signal contained in each frame to calculate the differential average value. The third differential signal is generated by subtracting the differential average value from each sample value of the second differential signal. The requantization error signal for each frame is generated based on the third differential signal. This requantization error signal is represented by the difference in bit depth between the first quantization bit depth and the second quantization bit depth, wherein the second quantization bit depth is greater than the first quantization bit depth. The requantization error signal is added to the first digital music signal contained in each frame to generate a second digital music signal with the second quantization bit depth.

3. A computer-readable storage medium storing a bit-extended processing program that causes a computer to perform the following steps: The samples of the first digital music signal, quantized with the first quantization bit depth, are divided into multiple sample numbers and then framed. Calculate the first differential signal, which takes the difference between two adjacent samples of the first digital music signal contained in each frame as the differential sample. The start and end positions of flat regions in the first differential signal, as well as the number of differential samples in the flat regions, are detected. A flat region is a region with two or more consecutive differential samples having the same sample value. Based on the pattern of whether the sample values ​​from the previous differential sample in the flat region to the starting differential sample, and whether the sample values ​​from the ending differential sample to the next differential sample in the flat region increase or decrease, one or more differential samples in the flat region are used as differential samples for correction. The sample values ​​of the correction targets are corrected by adding or subtracting correction values, thereby generating a second differential signal that deflattens the flat region. The sample values ​​of the second differential signal contained in each frame are summed, and the result is divided by the number of samples of the second differential signal contained in each frame to calculate the differential average value. The third differential signal is generated by subtracting the differential average value from each sample value of the second differential signal. The requantization error signal for each frame is generated based on the third differential signal. This requantization error signal is represented by the difference in bit depth between the first quantization bit depth and the second quantization bit depth, wherein the second quantization bit depth is greater than the first quantization bit depth. The requantization error signal is added to the first digital music signal contained in each frame to generate a second digital music signal with the second quantization bit depth.

Images