JP6314803B2 - Signal processing apparatus, signal processing method, and program - Google Patents

Description

  The present disclosure relates to a signal processing device, a signal processing method, and a program.

  In recent years, signal processing apparatuses for suppressing specific sounds from acoustic signals have been developed. As an example, many signal processing apparatuses that realize a so-called karaoke function that reproduces music while suppressing vocals have been developed. In the vocal suppression technology, it is a basic policy to suppress vocals by focusing on the fact that the position where the vocals are localized is generally in the center. Specifically, since many of the music pieces are created so that the vocal is localized in the center, as a result, the vocal is recorded in the left channel and the right channel in the same way. For this reason, if a difference is taken between the signals of both channels of the stereo signal, vocals recorded in the same way on both channels are suppressed. However, in such vocal suppression technology, there is a case where auditory noise may occur, and thus a technology for reducing noise is required.

  For example, in Patent Document 1 below, there is a technique in which after a sound signal is once expressed in the frequency domain, a difference calculation for suppressing vocals is performed in the frequency domain, and a frequency band having a low signal level is complemented with the original sound signal. It is disclosed.

Japanese Patent No. 5365380

  However, in the technique described in Patent Document 1, the performance of suppressing vocals has been reduced as a price for reducing noise. Specifically, the frequency band with a low signal level has been supplemented by the original acoustic signal including vocals.

  Therefore, the present disclosure proposes a new and improved signal processing apparatus, signal processing method, and program capable of both suppressing a specific sound from an acoustic signal and reducing auditory noise. To do.

  According to the present disclosure, the difference signal calculation unit that calculates the difference signal between the first channel acoustic signal and the second channel acoustic signal that form the input acoustic signal, and the difference signal calculation unit calculates the difference signal calculation unit. And a processing unit that adds a signal obtained by processing the difference signal to the difference signal.

  In addition, according to the present disclosure, the difference signal between the acoustic signal of the first channel and the acoustic signal of the second channel forming the input acoustic signal is calculated, and the difference signal is added to the calculated difference signal. A signal processing method comprising: adding a processed signal by a processor.

  In addition, according to the present disclosure, the difference signal calculation unit that calculates a difference signal between the first channel acoustic signal and the second channel acoustic signal that form the input acoustic signal, and the difference signal calculation. And a processing unit that adds a signal obtained by processing the difference signal to the difference signal calculated by a unit.

  As described above, according to the present disclosure, it is possible to achieve both suppression of a specific sound from an acoustic signal and reduction of auditory noise. Note that the above effects are not necessarily limited, and any of the effects shown in the present specification, or other effects that can be grasped from the present specification, together with or in place of the above effects. May be played.

It is a block diagram which shows an example of the logical structure of the signal processing apparatus which concerns on 1st Embodiment. It is a figure which shows an example of the signal flow of the difference signal calculation part which concerns on 1st Embodiment. It is a figure which shows an example of the signal flow of the difference signal calculation part which concerns on 1st Embodiment. It is a figure which shows an example of the signal flow of the difference signal calculation part which concerns on 1st Embodiment. It is a figure which shows an example of the signal flow of the difference signal calculation part which concerns on 1st Embodiment. It is a figure which shows an example of the signal flow of the blurring process part which concerns on 1st Embodiment. It is a figure which shows an example of the signal flow of the blurring process part which concerns on 1st Embodiment. It is a flowchart which shows an example of the flow of the signal processing performed in the signal processing apparatus which concerns on 1st Embodiment. It is a flowchart which shows an example of the flow of the update process of delay buffer DB performed in the blurring process part which concerns on 1st Embodiment. It is a figure for demonstrating the signal processing which concerns on a 1st comparative example. It is a figure for demonstrating the signal processing which concerns on a 1st comparative example. It is a figure for demonstrating the signal processing which concerns on a 1st comparative example. It is a figure for demonstrating the signal processing which concerns on a 2nd comparative example. It is a figure for demonstrating the signal processing which concerns on a 2nd comparative example. It is a figure for demonstrating the effect of the signal processing apparatus which concerns on 1st Embodiment. It is a block diagram which shows an example of the logical structure of the signal processing apparatus which concerns on 1st Embodiment. It is a block diagram which shows an example of the logical structure of the signal processing apparatus which concerns on 1st Embodiment. It is a figure for demonstrating the degree to which the input acoustic signal is near monaural. It is a block diagram which shows an example of the logical structure of the signal processing apparatus which concerns on 1st Embodiment. It is a block diagram which shows an example of the logical structure of the signal processing apparatus which concerns on 2nd Embodiment. It is a flowchart which shows an example of the flow of the signal processing performed in the signal processing apparatus which concerns on 2nd Embodiment. It is a figure for demonstrating the effect of the signal processing apparatus which concerns on 2nd Embodiment. It is a block diagram which shows an example of the logical structure of the signal processing apparatus which concerns on 2nd Embodiment. It is a block diagram which shows an example of the logical structure of the signal processing apparatus which concerns on 3rd Embodiment. It is a flowchart which shows an example of the flow of the signal processing performed in the signal processing apparatus which concerns on 3rd Embodiment. It is a block diagram which shows an example of the logical structure of the signal processing apparatus which concerns on 4th Embodiment. It is a figure which shows an example of the signal flow of the blurring process part which concerns on 4th Embodiment. It is a block diagram which shows an example of the hardware constitutions of the information processing apparatus which concerns on this embodiment.

  Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

The description will be made in the following order.
1. Overview 2. First embodiment 2-1. First configuration example 2-2. Operation processing example 2-3. Effect 2-4. Second configuration example 2-5. Third configuration example 2-6. 4. Fourth configuration example Second embodiment 3-1. First configuration example 3-2. Example of operation processing 3-3. Effect 3-4. Second configuration example 4. Third embodiment 4-1. Configuration example 4-2. Example of operation processing 4. Fourth embodiment Hardware configuration Summary

<1. Overview>
First, an overview of a signal processing device according to an embodiment of the present disclosure will be described.

  The signal processing apparatus according to the present embodiment performs signal processing that suppresses a specific sound from an input acoustic signal. The specific sound to be suppressed may be, for example, a sound localized at the center of the acoustic signal. Examples of such specific sounds include vocals and bass sounds. Hereinafter, as an example, the specific sound suppressed by the signal processing apparatus 100 according to the present embodiment will be described as being vocal. In addition, the process for suppressing the specific sound is hereinafter referred to as a blur process.

  The signal processing apparatus according to the present embodiment first suppresses a specific sound localized at the center of the acoustic signal by generating a differential signal. Subsequently, the signal processing apparatus according to the present embodiment reduces the auditory noise generated in the process of generating the difference signal by performing the blurring process.

  Hereinafter, this embodiment will be described in detail with reference to FIGS.

<2. First Embodiment>
[2-1. First Configuration Example]
FIG. 1 is a block diagram illustrating an example of a logical configuration of the signal processing apparatus 100 according to the present embodiment. Hereinafter, the configuration example illustrated in FIG. 1 is also referred to as a first configuration example. As illustrated in FIG. 1, the signal processing device 100 according to this configuration example includes a differential signal calculation unit 110 and a blur processing unit 120. The signal processing apparatus 100 performs signal processing on the input audio signal (acoustic signal) and outputs the processed acoustic signal.

(1) Difference signal calculation unit 110
The differential signal calculation unit 110 has a function of calculating a differential signal between the acoustic signal of the first channel and the acoustic signal of the second channel that form the input acoustic signal. For example, the input acoustic signal is a stereo signal, the first channel acoustic signal is a left channel acoustic signal, and the second channel acoustic signal is a right channel acoustic signal. Hereinafter, the acoustic signal of the left channel is also referred to as Lch, and the acoustic signal of the right channel is also referred to as Rch. The acoustic signal output from the difference signal calculation unit 110 may be a stereo signal. Hereinafter, the output left channel acoustic signal is also referred to as L'ch, and the output right channel acoustic signal is also referred to as R'ch.

  The difference signal calculation unit 110 calculates a difference signal in the time domain. For example, the difference signal calculation unit 110 calculates the difference signal by taking the difference between the Rch signal and the Lch signal, which are time domain signals. Hereinafter, an example of the signal flow of the difference signal calculation unit 110 for calculating the difference signal in the time domain will be described with reference to FIGS.

FIG. 2 is a diagram illustrating an example of a signal flow of the difference signal calculation unit 110 according to the present embodiment. In the example illustrated in FIG. 2, the difference signal calculation unit 110 obtains the difference signal S (i) by subtracting Rch from Lch and multiplying by 0.5. This signal flow is expressed by the following mathematical formula.
S (i) = (L (i) −R (i)) × 0.5 (Formula 1)

  Here, L (i) is an Lch signal, and R (i) is an Rch signal. i represents the sample time. The difference signal calculation unit 110 multiplies the signal after subtraction by 0.5 for the purpose of maintaining the signal level before and after processing.

  FIG. 3 is a diagram illustrating an example of a signal flow of the difference signal calculation unit 110 according to the present embodiment. In the example shown in FIG. 3, the difference signal calculation unit 110 calculates the difference signal S (i) in the same manner as in the example shown in FIG. Then, the difference signal calculation unit 110 outputs the same difference signal S (i) for L′ ch and R′ch. The output signal by this signal flow is substantially equivalent to a monaural signal.

  FIG. 4 is a diagram illustrating an example of a signal flow of the difference signal calculation unit 110 according to the present embodiment. In the example shown in FIG. 4, the difference signal calculation unit 110 calculates the difference signal S (i) in the same manner as in the example shown in FIG. Then, the difference signal calculation unit 110 outputs a signal obtained by inverting the phase of L′ ch as R′ch. Compared with the example shown in FIG. 3, the output signal by this signal flow can make the user feel a sense of spread. However, the output signal by this signal flow can give the user a sense of discomfort due to the inversion of this phase.

  FIG. 5 is a diagram illustrating an example of a signal flow of the difference signal calculation unit 110 according to the present embodiment. In the example shown in FIG. 5, the differential signal calculation unit 110 first extracts a vocal localized at the center by adding the Lch and Rch of the input signal to make it monaural. Next, the difference signal calculation unit 110 multiplies the monaural signal by 0.5, maintains the signal level, and subtracts from each of Lch and Rch to obtain L′ ch and R′ch. The output signal by this signal flow is the same as the example shown in FIG.

(2) Blur processing unit 120
The blur processing unit 120 performs blur processing. Specifically, the blur processing unit 120 has a function as a processing unit that adds a signal obtained by processing the difference signal to the difference signal calculated by the difference signal calculation unit 110. There are various signals that can be processed from the differential signal. The blurring processing unit 120 according to the present embodiment generates a delayed signal obtained by delaying the differential signal as a signal obtained by processing the differential signal. Then, the blurring processing unit 120 obtains an output signal by adding the generated delay signal to the difference signal. Note that the process of adding the delay signal to the difference signal may be simple addition or weighted addition, or addition (ie, subtraction) with one of the signs reversed. It may be. Hereinafter, the output signal from the blur processing unit 120 is also referred to as a blur signal F (i).

  The blurring processing unit 120 may generate a delay signal using an IIR (Infinite impulse response) filter. Here, with reference to FIG. 6, a signal flow for generating a delay signal using an IIR filter and obtaining a blur signal F (i) will be described.

FIG. 6 is a diagram illustrating an example of a signal flow of the blur processing unit 120 according to the present embodiment. As illustrated in FIG. 6, the blurring processing unit 120 obtains the blurring signal F (i) by adding the delay signal D (i) accumulated in the delay buffer DB121 to the differential signal S (i). The delayed signal D (i) is a signal obtained by delaying the blur signal F (i) by n samples. At the time of addition, the blurring processing unit 120 weights and adds the difference signal S (i) and the delayed signal D (i) using the weighting coefficient r related to the addition. The weighting coefficient r can also be regarded as a mixing ratio of the differential signal S (i) and the delayed signal D (i). This signal flow is expressed by the following mathematical formula.
F (i) = (1−r) × S (i) + r × D (i) (Formula 2)

Here, the weighting coefficient r takes a value in the following range.
0 <r <1 (Formula 3)

  The blurring processing unit 120 may generate a delay signal using an FIR (Finite impulse response) filter. Here, a signal flow for generating a delay signal using an FIR filter and obtaining a blur signal F (i) will be described with reference to FIG.

FIG. 7 is a diagram illustrating an example of a signal flow of the blur processing unit 120 according to the present embodiment. As shown in FIG. 7, the blurring processing unit 120 has m delay units 122 that delay the input signal by one sample, and weights and adds up to a delay signal delayed by a maximum of m samples to the differential signal S (i). Thus, the blur signal F (i) is obtained. The delayed signal here is a signal obtained by delaying the differential signal S (i). This signal flow is expressed by the following mathematical formula.
F (i) = r ₀ × S (i) + r ₁ × S (i−1) +... + R _m × S (im)
(Formula 4)

Here, S (i−m) represents a differential signal in the past of m samples. Further, weighting coefficient _r 0 ~r _m, respectively satisfy the above equation 3.

  Note that the blurring processing unit 120 may use either the IIR filter or the FIR filter, may use both in combination, or may generate a delay signal by any other method.

  The first configuration example has been described above. Subsequently, an operation process of the signal processing apparatus 100 according to the present embodiment will be described.

[2-2. Operation processing example]
FIG. 8 is a flowchart showing an example of the flow of signal processing executed in the signal processing apparatus 100 according to the present embodiment. In this flowchart, an example in which the blur processing unit 120 generates a delay signal using an IIR filter will be described.

  As shown in FIG. 8, first, in step S102, the difference signal calculation unit 110 accepts input of an i-th Lch signal L (i) and an Rch signal R (i).

  Next, in step S104, the difference signal calculation unit 110 calculates the difference signal S (i). For example, the difference signal calculation unit 110 calculates the difference signal S (i) using the above Equation 1.

  Next, in step S106, the blurring processing unit 120 calculates the blurring signal F (i) from the difference signal S (i) and the delay signal D (i). For example, the blur processing unit 120 calculates the blur signal F (i) using Equation 2 above.

  Next, in step S108, the blurring processing unit 120 updates the delay buffer DB 121. Since this process will be described in detail later, a description thereof is omitted here.

  In step S110, the blur processing unit 120 outputs the calculated blur signal F (i).

  The signal processing example by the signal processing apparatus 100 has been described above. Next, the process in step S108 will be described with reference to FIG.

  FIG. 9 is a flowchart showing an example of a flow of update processing of the delay buffer DB 121 executed in the blur processing unit 120 according to the present embodiment.

  As shown in FIG. 9, the blurring processing unit 120 first sets j = 0 in step S202. j is a variable used for the update process.

  Next, in step S204, the blurring processing unit 120 determines whether j <n−1 is satisfied. Here, n is the size of the delay buffer DB 121 and represents the delay amount.

  When it is determined that j <n−1 (S204 / YES), in step S206, the blurring processing unit 120 copies the delay buffer DB [j + 1] to the delay buffer DB [j]. Here, the delay buffer DB [j] represents the jth data stored in the delay buffer DB121.

  Next, in step S208, the blurring processing unit 120 increments the variable j by setting j = j + 1.

  Thereafter, the process returns to step S204 again. In this way, the processes in steps S206 and S208 are repeated until j <n−1 is not satisfied.

  When it is determined that j <n−1 is not satisfied (S204 / NO), in step S210, the blurring processing unit 120 copies the blurring signal F (i) to the delay buffer DB [n−1].

  Through the processing described above, a signal delayed by n samples is stored in the delay buffer DB [0]. The blurring processing unit 120 uses the delay buffer DB [0] as the delay signal D (i). The example of the update processing of the delay buffer DB 121 by the blur processing unit 120 has been described above.

[2-3. effect]
Below, the effect of the signal processing apparatus 100 according to the present embodiment will be described in comparison with the comparative example.

(Prerequisite knowledge)
As one of the compression coding techniques, there is a joint stereo coding method in which the coding is performed using the correlation between channels. The joint stereo coding method includes a middle side stereo coding method and an intensity stereo coding method. The middle-side stereo encoding method is a method of encoding separately for a sum signal (Lch + Rch) and a difference signal (Lch-Rch), and encoding the sum signal (Lch + Rch) with a weight increases the encoding efficiency. This is an encoding method that can be improved. The intensity stereo coding method is a coding method capable of improving the coding efficiency by coding the power ratio between the sum signal (Lch + Rch) and the left and right channels. The joint stereo coding scheme improves compression efficiency and enables compression with a smaller bit rate, or enables compression with higher sound quality at the same bit rate.

(First comparative example)
First, as a first comparative example, consider a signal processing apparatus that suppresses vocals recorded in the same way on both channels by taking the difference between the signals of both channels of the stereo signal described above. Below, with reference to FIGS. 10-12, the case where the signal processing apparatus which concerns on a 1st comparative example processes about the sound source compressed using the joint stereo encoding system is demonstrated.

  10 to 12 are diagrams for explaining signal processing according to the first comparative example. Specifically, FIG. 10 is an example of a power spectrogram when a sound source compressed using the joint stereo coding method is processed by the signal processing apparatus according to this comparative example. In FIG. 10, the horizontal axis represents time, the vertical axis represents frequency, the colored part indicates that the signal level (power) is high, and the colorless part indicates that the signal level is low. Referring to FIG. 10, a portion with a high signal level and a portion with a low signal level are formed in a block shape in which the time direction is a frame unit and the frequency direction is a scale factor band unit, and are mixed. Such an unpleasant auditory noise is caused by the fact that the high signal level and the low signal level are formed in a block shape.

  FIG. 11 is a graph obtained by extracting the power spectrogram between the sections AB in FIG. 10 and shows a change in the frequency direction at a certain time. In FIG. 11, the horizontal axis is frequency and the vertical axis is power. The scale on the horizontal axis is scaled in units of scale factor bands. In actual compression coding, the width of the low-scale scale factor band is set to be narrower than that of the high frequency, but in FIG. 11, they are schematically depicted with the same width. Referring to FIG. 11, the power spectrum rises and falls sharply for each scale factor band. Such a steep change is attributed to the fact that the sound source is compressed using the joint stereo encoding method.

  More specifically, first, in the joint stereo encoding method, determination as to whether middle-side stereo encoding is performed or intensity stereo encoding is performed for each scale factor band. When the bit rate assigned to the difference signal (Lch-Rch) of the scale factor band compressed by the middle side stereo coding method is very small, the scale factor band portion in the compressed acoustic signal is substantially monaural. Close to the signal. Therefore, in the processing by the signal processing apparatus according to this comparative example, the level of the scale factor band portion that is substantially close to a monaural signal can be a value close to zero. Similarly, when the power ratio of the left and right channels of the scale factor band compressed by the intensity stereo coding method is close to 1, the scale factor band portion in the compressed acoustic signal is substantially close to a monaural signal. Become. Therefore, in the processing by the signal processing apparatus according to this comparative example, the level of the scale factor band portion that is substantially close to a monaural signal can be a value close to zero. As described above, due to the sound source being compressed using the joint stereo encoding method, the steep level change in the frequency direction shown in FIG. 11 may occur. Such a sharp level change in the frequency direction is one of the causes of annoying auditory noise.

  FIG. 12 is a graph obtained by extracting a power spectrogram between the section CDs of FIG. 10, and shows a change in the time direction at a certain frequency. In FIG. 12, the horizontal axis is time, and the vertical axis is power. The scale on the horizontal axis is swung in units of frames. Referring to FIG. 12, the power spectrum sharply rises and falls every frame. Such a steep level change in the time direction may occur in each scale factor band due to the sound source being compressed using a joint stereo coding method for each frame. Such a steep level change in the time direction is one of the major causes of annoying auditory noise.

(Second comparative example)
Next, as a second comparative example, consider a signal processing apparatus using the technique described in Patent Document 1 that has an effect of preventing the generation of auditory noise even with a compressed sound source. As described above, the signal processing apparatus according to this comparative example once expresses an acoustic signal in the frequency domain, and then performs difference calculation for suppressing vocals, that is, Lch-Rch in the frequency domain. Hereinafter, a case where the signal processing apparatus according to the second comparative example processes a sound source compressed using the joint stereo coding scheme will be described with reference to FIGS. 13 and 14.

  13 and 14 are diagrams for explaining signal processing according to the second comparative example. Specifically, reference numeral 200 in FIG. 13 denotes power (Pl) for each Lch scale factor band. Reference numeral 210 in FIG. 13 represents power (Pr) for each Rch scale factor band. The code | symbol 220 of FIG. 13 is the power (Pd) for every scale factor band of difference signal Lch-Rch. When the Lch power and the Rch power are at the same level in the same scale factor band, the power of the differential signal is close to zero. For example, reference numerals 201 and 211, reference numerals 202 and 212, reference numerals 203 and 213, and reference numerals 204 and 214 are at the same level. For this reason, in the differential signal indicated by reference numeral 220, the power of the scale factor band corresponding to these is close to zero. Such a state is the same as the example shown in FIG.

  Therefore, as shown in FIG. 14, the signal processing apparatus according to the present comparative example alleviates such a steep level change by complementing the portion that is close to zero with the original signal. Yes. For example, the signal processing apparatus according to this comparative example detects a steep level decrease such as the section 11, the section 12, and the section 13 as the first step. Then, as a second step, the signal processing apparatus according to this comparative example uses the original Lch signal to complement the section 11, the section 12, and the section 13, thereby preventing a steep level decrease.

  Specifically, as indicated by reference numeral 240 in FIG. 14, the signal processing apparatus according to this comparative example uses the difference signal sections 11, 12, and 13 indicated by reference numeral 220 in FIG. The scale factor band power 201, 202, 203 and 204 corresponding to each section is copied. The signal processing apparatus according to this comparative example can multiply an arbitrary coefficient during copying. Comparing the reference numeral 220 in FIG. 13 and the reference numeral 240 in FIG. 14, the sections other than the sections 11, 12, and 13 are the same. As indicated by reference numeral 240 in FIG. 14, the signal processing device according to this comparative example can prevent a steep level change in the frequency direction. Along with this, the signal processing apparatus according to this comparative example is expected to prevent a steep level change in the time direction to some extent, so that it is possible to prevent the generation of auditory noise.

  However, the signal processing apparatus according to this comparative example has a reduced performance of suppressing vocals as a price for reducing noise. This is due to the fact that the signal processing apparatus according to this comparative example supplemented the section in which the steep level decrease was detected using the Lch signal including the vocal in order to prevent the steep level change.

  In addition, when the signal processing apparatus according to this comparative example fails in the first step described above, the failed section cannot be complemented. Furthermore, since the first step is performed in the frequency domain, when the signal to be processed is a time domain signal, the signal processing apparatus according to this comparative example performs a time domain and frequency domain conversion process before and after the vocal suppression process. Had gone. For example, the signal processing apparatus according to this comparative example converts the signal into a frequency domain signal using FFT (Fast Fourier Transform) or the like before vocal suppression processing, and converts it into a time domain signal using IFFT (Inverse FFT) or the like after vocal suppression processing. obtain. The amount of calculation for such conversion processing is not small. Also, the amount of calculation for the detection process in the first step is not small.

(Effect of this embodiment)
Below, with reference to FIG. 15, the effect of the signal processing apparatus 100 which concerns on this embodiment is demonstrated.

  FIG. 15 is a diagram for explaining the effect of the signal processing apparatus 100 according to the present embodiment. Specifically, reference numeral 300 in FIG. 15 represents a change in the time direction at a certain frequency of a power spectrogram when a sound source compressed using the joint stereo encoding method is processed by the signal processing apparatus 100 according to the present embodiment. The state of is shown. Moreover, the code | symbol 310 of FIG. 15 is a mode of the change of the power spectrogram shown in FIG. 12 and FIG. 15 indicate the same section.

  Referring to FIG. 15, when processed by the signal processing apparatus 100 according to the present embodiment, a steep change in level is alleviated. For example, in the section CD2, a steep level drop is recognized in the reference numeral 310, whereas a steep level drop is not recognized in the reference numeral 300, and the change is gradually made. This is due to the fact that a delayed signal that does not have a level drop is added to a section in which a steep level drop has occurred. As shown in FIG. 15, the signal processing apparatus 100 according to the present embodiment can alleviate a steep level change in the time direction, and thus can prevent generation of annoying auditory noise.

  Also, the blurring processing unit 120 according to the present embodiment generates a delay signal using the differential signal in which vocals are suppressed, and relaxes a steep level change using the delay signal. Therefore, in the present embodiment, unlike the second comparative example, the vocal suppression performance is not compensated, and high vocal suppression performance can be realized.

  Further, the blurring processing unit 120 according to the present embodiment can process the time domain signal output from the differential signal calculation unit 110 without converting it to the frequency domain. For this reason, the signal processing apparatus 100 according to the present embodiment can reduce the amount of calculation for the conversion process as compared with the signal processing apparatus according to the second comparative example.

  Further, since the blurring processing unit 120 according to the present embodiment generates a delay signal using IIR, FIR, or the like, it is possible to alleviate a steep level change with a small amount of calculation. Furthermore, since the blurring processing unit 120 according to the present embodiment does not detect a steep decrease in level, it is possible to avoid a failure in complementation due to a failure in detection processing, as compared with the second comparative example. It is possible to reduce the amount of calculation for the detection process.

  The effects according to the present embodiment have been described above. Hereinafter, another configuration example according to the present embodiment will be described. Note that the effects described above are similarly achieved in other configuration examples described below.

[2-4. Second configuration example]
This configuration example is a configuration example in which the delay amount n and the weighting coefficient r used by the blurring processing unit 120 are appropriately set. Hereinafter, this configuration example will be described with reference to FIG.

  FIG. 16 is a block diagram illustrating an example of a logical configuration of the signal processing apparatus 100 according to the present embodiment. Hereinafter, the configuration example illustrated in FIG. 16 is also referred to as a second configuration example. As illustrated in FIG. 16, the signal processing device 100 according to this configuration example includes a differential signal calculation unit 110, a blur processing unit 120, a delay amount setting unit 123, and a coefficient setting unit 124.

  The difference signal calculation unit 110 outputs a difference signal S (i). The blurring processing unit 120 obtains the blurring signal F (i) using the delay amount n set by the delay amount setting unit 123 and the weighting coefficient r set by the coefficient setting unit 124 in Equation 2 above. Since the internal processing of the difference signal calculation unit 110 and the blur processing unit 120 is as described above, detailed description thereof is omitted here.

(1) Delay amount setting unit 123
The delay amount setting unit 123 has a function of setting the delay amount n of the delay signal. The delay amount setting unit 123 can alleviate a sharp level change in the time direction by setting an appropriate delay amount n.

  The size of each block of the block spectrogram shown in FIG. 10 that has occurred in the first comparative example depends on the compression coding information (audio codec). Specifically, the size of the block in the time direction is approximately equal to the frame width of the audio codec, and the size of the block in the frequency direction is approximately equal to the scale factor bandwidth of the audio codec. As shown in the example of the level fluctuation in the time axis direction shown in FIG. 12 that has occurred in the first comparative example, the time width when the level suddenly approaches zero or returns to a certain level is It almost matches the integer multiple of the audio codec frame width. For example, the section CD2 in FIG. 12 has a width of one frame, and the width between the sections CD2 and CD3 is also one frame.

As described above, since the steep level fluctuation in the time direction in the first comparative example occurs in units of frames of the audio codec, the delay amount setting unit 123 uses the compression encoding information of the input acoustic signal to delay the amount of delay. Set n. In the present embodiment, the current frame level of the blur signal F (i) output from the signal processing device 100 is added to the difference signal S (i) in order to prevent the current frame level from dropping sharply compared to the immediately preceding frame. It is desirable that the level of the delayed signal D (i) is a certain level. That is, in Equation 2 above, when the level of the differential signal S (i) is close to zero and the delay signal D (i) has a certain level, a steep drop in the blur signal F (i) is prevented. . Therefore, the delay amount setting unit 123 sets the delay amount n of the delay signal D (i) to be equal to or less than the frame width indicated by the audio codec, as shown in the following equation.
0 <Delay amount n <= Audio codec frame width (Formula 5)

  In this case, at the timing when the level of the differential signal S (i) approaches zero, the immediately preceding non-zero differential signal S (i) component is included in the delayed signal D (i). Therefore, even when the level of the differential signal S (i) is close to zero, it is realized that the level of the delayed signal D (i) is some, and a sharp level drop of the blur signal F (i) is prevented. It is.

Empirically, it is desirable to set the delay amount n within the range of the following mathematical formula.
70% of audio codec frame width <delay amount n
<Audio codec frame width (Formula 6)

(2) Coefficient setting unit 124
The coefficient setting unit 124 has a function of setting a weighting coefficient r related to the addition performed by the blurring processing unit 120. The coefficient setting unit 124 can adjust the strength of the blurring process by setting an appropriate weighting coefficient r. For example, the coefficient setting unit 124 sets the weighting coefficient r based on the audio codec of the input acoustic signal.

  When the bit rate of the audio codec is low, the block spectrogram shown in FIG. 10 is likely to occur in the first comparative example. This is because joint stereo coding is more actively used when the bit rate of the audio codec is low. Therefore, the coefficient setting unit 124 sets the weighting coefficient r based on the bit rate of the audio codec. More specifically, the coefficient setting unit 124 sets the weighting coefficient r so as to perform the blurring process more strongly when the bit rate of the audio codec is low. That is, in Equation 2, when the audio codec bit rate is low, the coefficient setting unit 124 sets the weighting coefficient r to the 1 side, and when the audio codec bit rate is high, the coefficient setting unit 124 moves the weighting coefficient r to the zero side. To set. In addition, the coefficient setting unit 124 may set the weighting coefficient r according to the use situation of joint stereo coding. With this setting, the signal processing apparatus 100 can perform strong blurring processing when there is a high possibility of auditory noise, and can use the original sound by weakening the blurring processing when the possibility of auditory noise is low. Become.

Empirically, the coefficient setting unit 124 desirably sets the weighting coefficient r within the range of the following mathematical formula.
0.0 <r <0.4 (Formula 7)

(3) Others The delay amount setting unit 123 and the coefficient setting unit 124 may change the delay amount n and the weighting coefficient r over time. In this case, the delay amount setting unit 123 and the coefficient setting unit 124 can support automatic switching of a plurality of frame widths and a variable bit rate audio codec. When it is determined from the audio codec information that joint stereo encoding is not used, the coefficient setting unit 124 may set the weighting coefficient r to zero and turn off the blurring process.

  As described above, according to the present configuration example, the signal processing apparatus 100 can reliably mitigate a sharp level change in the time direction by setting the delay amount n. Further, according to the present configuration example, the signal processing device 100 can achieve both reducing auditory noise and utilizing the original sound by setting the weighting coefficient r.

[2-5. Third configuration example]
This configuration example is a configuration example in which a parameter for the coefficient setting unit 124 to set the weighting coefficient r is introduced. Hereinafter, this configuration example will be described with reference to FIGS. 17 and 18.

  FIG. 17 is a block diagram illustrating an example of a logical configuration of the signal processing apparatus 100 according to the present embodiment. Hereinafter, the configuration example illustrated in FIG. 17 is also referred to as a third configuration example. As illustrated in FIG. 17, the signal processing device 100 according to this configuration example includes a difference signal calculation unit 110, a blur processing unit 120, a coefficient setting unit 124, and a blur level calculation unit 125.

  The difference signal calculation unit 110 outputs a difference signal S (i). The coefficient setting unit 124 according to the present embodiment sets the weighting coefficient r according to the blur level f (i) calculated by the blur level calculation unit 125. The blurring processing unit 120 obtains the blurring signal F (i) by using the weighting coefficient r set by the coefficient setting unit 124 in Equation 2 above. Since the internal processing of the difference signal calculation unit 110, the blurring processing unit 120, and the coefficient setting unit 124 is as described above, detailed description thereof is omitted here.

  The blur level calculation unit 125 sets the blur level f (i) in accordance with the conspicuousness of the auditory noise of the input acoustic signal. Hereinafter, an example will be described in which the degree to which the input acoustic signal is close to monaural is adopted as an example of a measure of the conspicuousness of auditory noise.

  The degree of auditory noise caused by the block spectrogram shown in FIG. 10 that has occurred in the first comparative example can vary during the music. For this reason, it is desirable to change the intensity | strength of a blurring process according to the conspicuousness of auditory noise. The conspicuousness of auditory noise can be roughly measured by, for example, how similar the input audio signal Lch and Rch are, in other words, how close to monaural. Auditory noise tends to be conspicuous in a part that is close to monaural in the input acoustic signal, that is, a part in which most sounds are localized in the center. For example, vocal solo parts are often close to monaural, and auditory noise tends to stand out. Conversely, the part that is not close to monaural, that is, the part that has few sounds localized in the center, is less noticeable. This is due to the fact that joint stereo coding itself is mainly used in parts close to monaural. For this reason, it is desirable that the blurring process be performed more strongly when the input acoustic signal is close to monaural.

  Therefore, the coefficient setting unit 124 sets the weighting coefficient r based on the degree to which the input acoustic signal is close to monaural. Therefore, the blur level calculation unit 125 calculates the blur level f (i) based on the degree to which the input acoustic signal is close to monaural. For example, the blur level calculation unit 125 sets the blur level f (i) to be large when the input acoustic signal is close to monaural, and sets the blur level blur level f (i) to be small when it is not close to monaural. Then, the coefficient setting unit 124 sets the weighting coefficient r according to the blur level f (i). For example, the coefficient setting unit 124 sets the weighting coefficient r closer to the 1 side as the blur level f (i) is higher, and sets the weighting coefficient r closer to the zero side as the blur level f (i) is lower.

Whether or not it is close to monaural can be determined by a scale t (i) indicating how close to monaural is shown in the following mathematical formula.
Peak _S (i) = (1−k) × Peak _S (i−1)
+ K × (| L (i) −R (i) |) (Formula 8)
Peak _M (i) = (1−k) × Peak _M (i−1)
+ K × (| L (i) + R (i) |) (Equation 9)
t (i) = Peak _S (i) / Peak _M (i) (Formula 10)

Here, the coefficient k is a time constant. Also, Peak _M (i) is assumed not to be zero. Peak _S (i) is a peak level of a signal obtained by subtracting Rch from Lch. Peak _M (i) is a peak level of a signal obtained by adding Rch to Lch. In addition, although the absolute value is used in the above mathematical formulas 8 and 9, square may be used.

When the input acoustic signal is close to monaural, Peak _S (i) is small and Peak _M (i) is large. On the other hand, if it is not close to monaural, Peak _S (i) increases and Peak _M (i) decreases. Therefore, the scale t (i) is small when it is close to monaural and is large when it is not close to monaural. This point will be described in more detail with reference to FIG.

FIG. 18 is a diagram for explaining the degree to which the input acoustic signal is close to monaural. Specifically, in FIG. 18, an example of the temporal change of the peak level Peak _M (i) indicated by reference numeral 401 and an example of the peak level Peak _S (i) and the temporal change indicated by reference numeral 402 are respectively shown. . The sections 21 and 22 are parts in which the input acoustic signal is close to monaural, and at the same time, auditory noise is easily noticeable. In these sections, the peak level Peak _S (i) indicated by reference numeral 402 decreases, and the peak level Peak _M (i) indicated by reference numeral 401 increases, so the scale t (i) decreases. In other sections, the scale t (i) is larger than that in the sections 21 and 22.

  The blur level calculation unit 125 calculates the blur level f (i) according to the scale t (i). For example, the blur level calculation unit 125 sets the blur level f (i) to be large when the scale t (i) is small. For this reason, the coefficient setting unit 124 sets the weighting coefficient r large for the difference signal S (i) corresponding to the sections 21 and 22 shown in FIG. 18, and the blurring processing unit 120 performs the blurring process strongly. On the other hand, the blur level calculation unit 125 sets the blur level f (i) to be small when the scale t (i) is large. For this reason, the coefficient setting unit 124 sets the weighting coefficient r small for the difference signal S (i) corresponding to the sections other than the sections 21 and 22 shown in FIG. 18, and the blurring processing unit 120 performs the blurring process weakly. . As described above, the signal processing apparatus 100 according to the present configuration example can perform the strong blurring process by focusing on the part where the auditory noise is conspicuous by changing the intensity of the blur level depending on the conspicuousness of the auditory noise. It is possible to prevent auditory noise more effectively.

Note that it is difficult to determine whether the input acoustic signal is close to monaural or the level of the acoustic signal itself is small only by the magnitude of Peak _S (i). Further, the blur level calculation unit 125 may use the correlation between Lch and Rch as the scale t (i). However, in that case, the magnitude relationship of the scale t (i) is reversed.

  As described above, according to the present configuration example, the signal processing apparatus 100 can more effectively prevent the auditory noise by performing the strong blurring process focusing on the part where the auditory noise is conspicuous.

[2-6. Fourth configuration example]
This configuration example is a configuration example in which a band in which auditory noise is generated is extracted from the difference signal and blur processing is performed. Hereinafter, this configuration example will be described with reference to FIG.

  FIG. 19 is a block diagram illustrating an example of a logical configuration of the signal processing apparatus 100 according to the present embodiment. Hereinafter, the configuration example illustrated in FIG. 19 is also referred to as a fourth configuration example. As illustrated in FIG. 19, the signal processing device 100 according to this configuration example includes a difference signal calculation unit 110, a blur processing unit 120, a band division unit 130, and a synthesis unit 131.

  The difference signal calculation unit 110 outputs a difference signal. Next, the band dividing unit 130 divides the differential signal into a plurality of bands. Next, the blurring processing unit 120 performs blurring processing in at least one band among the plurality of bands divided by the band dividing unit 130. The synthesizing unit 131 then synthesizes the signal that has not been subjected to the blurring processing by the blurring processing unit 120 and the signal that has not been subjected to blurring processing, to obtain a blur signal. Since the internal processing of the difference signal calculation unit 110 and the blur processing unit 120 is as described above, detailed description thereof is omitted here.

(1) Band division unit 130
The band dividing unit 130 has a function of dividing the difference signal output from the difference signal calculating unit 110 into a plurality of bands. For example, the band dividing unit 130 divides the band into the band to be blurred by the blur processing unit 120 and the band to be out of the band. The band to be subjected to the blurring process may be one continuous band or an aggregate of a plurality of discontinuous bands. The same applies to bands that are not subject to blur processing.

  The conspicuousness of the auditory noise caused by the block-shaped spectrogram shown in FIG. 10 that occurs in the first comparative example differs depending on which frequency band the block-shaped spectrogram is generated. This is considered to depend on the frequency band bias targeted for joint stereo coding and the characteristics of human hearing. The frequency band in which auditory noise is conspicuous is empirically 1 kHz to 10 kHz. For this reason, it is desirable to perform the blurring process mainly in a band in which auditory noise is conspicuous. Therefore, the band dividing unit 130 outputs the band in which the auditory noise is conspicuous to the blurring processing unit 120, and outputs the other band to the synthesizing unit 131.

  For example, the band dividing unit 130 extracts a signal of the band to be output to the blurring processing unit 120 using a bandpass filter in which the lower cutoff frequency is Fc1 and the upper cutoff frequency is Fc2. obtain. The cut-off frequency is empirically effective to be about Fc1 = 1 kHz and Fc2 = 10 kHz. The band dividing unit 130 outputs the band signal extracted by the band-pass filter to the blurring processing unit 120, thereby realizing a focused blurring process regarding the band. The band dividing unit 130 may include a high-pass filter whose cutoff frequency is Fc1 instead of the band-pass filter. In this case, the amount of calculation can be suppressed.

  Note that the band dividing unit 130 may be provided before the differential signal calculating unit 110. In that case, the band dividing unit 130 narrows the band for obtaining the difference signal to a frequency band in which vocal sounds mainly exist, thereby suppressing the bass sound that is often localized in the center and reducing the low frequency band. It can be avoided that there are few light sounds.

(2) Synthesis unit 131
The combining unit 131 has a function of combining a plurality of difference signals divided by the band dividing unit 130. Specifically, the synthesizing unit 131 uses the difference signal of the band subjected to the blurring process by the blurring processing unit 120 and the difference signal of the band that has not been subjected to the blurring process by the blurring processing unit 120 among the plurality of bands divided by the blurring processing unit 120. And synthesize. The synthesizer 131 can synthesize these signals simply by adding them.

  As described above, according to the present configuration example, the signal processing apparatus 100 can more effectively prevent the auditory noise by performing the blurring process mainly in the band where the auditory noise is conspicuous.

<3. Second Embodiment>
In the present embodiment, auditory noise is reduced by gain control. First, the basic configuration of the present embodiment will be described with reference to FIG.

[3-1. First Configuration Example]
FIG. 20 is a block diagram illustrating an example of a logical configuration of the signal processing apparatus 100 according to the present embodiment. Hereinafter, the configuration example illustrated in FIG. 20 is also referred to as a first configuration example. As illustrated in FIG. 20, the signal processing device 100 according to this configuration example includes a differential signal calculation unit 110, a gain level setting unit 140, and a gain control unit 141.

  The difference signal calculation unit 110 outputs a difference signal. Next, the gain level setting unit 140 sets a gain level. Then, the gain control unit 141 controls the gain of the differential signal using the gain level set by the gain level setting unit 140. The signal processing apparatus 100 according to the present embodiment receives a time-domain acoustic signal that is a song in which the vocal is localized in the center, and outputs a time-domain acoustic signal in which the vocal is suppressed. Since the internal processing of the difference signal calculation unit 110 is as described above, a detailed description thereof is omitted here.

(1) Gain level setting unit 140
The gain level setting unit 140 has a function of setting the gain level of the differential signal. For example, the gain level setting unit 140 sets the gain level according to the conspicuousness of the auditory noise of the input acoustic signal.

  The degree of auditory noise caused by the block spectrogram shown in FIG. 10 that has occurred in the first comparative example can vary during the music. For this reason, it is desirable to change the gain level of a differential signal according to the conspicuousness of auditory noise. As described above, the input sound signal is close to monaural, for example, a part in which most sounds such as vocal solo parts are localized in the center, auditory noise is easily noticeable, and other parts are not easily noticeable. Therefore, it is desirable to change the gain level of the differential signal when the input acoustic signal is close to monaural.

Therefore, the gain level setting unit 140 uses the scale t (i) shown in the above Equations 8 to 10 as an example of a measure of the conspicuousness of auditory noise, and the degree to which the input acoustic signal is close to monaural. Set the gain level based on. Specifically, the gain level setting unit 140 sets the gain level g (i) small when the scale t (i) is small, and sets the gain level g (i) large when the scale t (i) is large. To do. For example, the gain level setting unit 140 sets the gain level g (i) within the range of the following mathematical formula.
0.0 <= g (i) <= 1.0 (Formula 11)

Empirically, it is desirable to set the gain level g (i) within the range of the following formula.
0.25 <g (i) <= 1.0 (Formula 12)

(2) Gain control unit 141
The gain control unit 141 has a function of controlling the gain of the differential signal using the gain level set by the gain level setting unit 140. For example, the gain control unit 141 can control the gain level based on the setting by the gain level setting unit 140, thereby reducing the gain in a section where the vocal is conspicuous and outputting a time domain acoustic signal in which the vocal is suppressed. Is possible. When the gain level set by the gain level setting unit 140 is g (i), the gain control unit 141 calculates a signal G (i) whose gain is controlled by the following mathematical formula.
G (i) = g (i) × S (i) (Formula 13)

  The first configuration example has been described above. Subsequently, an operation process of the signal processing apparatus 100 according to the present embodiment will be described.

[3-2. Operation processing example]
FIG. 21 is a flowchart illustrating an example of the flow of signal processing executed in the signal processing apparatus 100 according to the present embodiment.

  As shown in FIG. 21, first, in step S302, the difference signal calculation unit 110 accepts input of an i-th Lch signal L (i) and an Rch signal R (i).

  Next, in step S304, the difference signal calculation unit 110 calculates the difference signal S (i). For example, the difference signal calculation unit 110 calculates the difference signal S (i) using the above Equation 1.

  Next, in step S306, the gain level setting unit 140 calculates the gain level g (i). For example, the gain level setting unit 140 calculates the gain level g (i) using the above formulas 8 to 12.

  Next, in step S308, the gain control unit 141 calculates a signal G (i) whose gain is controlled. For example, the gain control unit 141 calculates the signal G (i) whose gain is controlled using the above Equation 13.

  In step S310, the gain control unit 141 outputs a signal G (i) in which the calculated gain is controlled.

[3-3. effect]
Below, with reference to FIG. 22, the effect of the signal processing apparatus 100 which concerns on this embodiment is demonstrated.

  FIG. 22 is a diagram for explaining the effect of the signal processing apparatus 100 according to the present embodiment. The solid line in FIG. 22 is an example of the time change of the power of the acoustic signal processed by the signal processing apparatus according to the first comparative example. For example, the section 31 and the section 32 are sections in which the input acoustic signal is close to monaural, such as a vocal solo part. Such a section is a section in which the power of the differential signal is reduced by suppressing a signal close to monaural and is a portion in which auditory noise is easily noticeable. The sections other than the section 31 and the section 32 are sections in which the input acoustic signal is not close to monaural, such as a part in which various musical instruments exist. Such a section is a section in which the power of the differential signal is larger than that of the sections 31 and 32 and is a portion in which auditory noise is not noticeable.

  The broken line in FIG. 22 is an example of time change of the power of the acoustic signal processed by the signal processing apparatus 100 according to the present embodiment. As indicated by the broken lines in the section 31 and the section 32, the signal processing apparatus 100 according to the present embodiment can lower the level by performing gain control mainly on the portion where the auditory noise is conspicuous. Since the signal processing apparatus 100 can lower the level of each auditory noise in a portion where the auditory noise is conspicuous, it is possible to reduce the discomfort of the auditory noise given to the user. Further, since the signal processing apparatus 100 according to the present embodiment does not perform processing in the frequency domain as in the second comparative example, it is possible to perform processing with a small amount of calculation.

  The effects according to the present embodiment have been described above. Hereinafter, another configuration example according to the present embodiment will be described. Note that the effects described above are similarly achieved in other configuration examples described below.

[3-4. Second configuration example]
This configuration example is a configuration example in which gain control is performed by extracting a band in which auditory noise occurs from the differential signal. Hereinafter, this configuration example will be described with reference to FIG.

  FIG. 23 is a block diagram illustrating an example of a logical configuration of the signal processing apparatus 100 according to the present embodiment. Hereinafter, the configuration example illustrated in FIG. 23 is also referred to as a second configuration example. As shown in FIG. 23, the signal processing apparatus 100 according to this configuration example includes a differential signal calculation unit 110, a band division unit 130, a synthesis unit 131, a gain level setting unit 140, and a gain control unit 141.

  The difference signal calculation unit 110 outputs a difference signal. Next, the band dividing unit 130 divides the differential signal into a plurality of bands. Specifically, the band dividing unit 130 divides a band that is a target of gain control by the gain control unit 141 and a band that is not a target. Here, for the same reason as in the fourth configuration example in the first embodiment, it is desirable that gain control is performed mainly in a band in which auditory noise is conspicuous. Therefore, the band dividing unit 130 outputs to the gain control unit 141 the band where auditory noise is conspicuous, and outputs the other band to the combining unit 131.

  Next, the gain level setting unit 140 sets a gain level. Then, the gain control unit 141 controls the gain of the differential signal using the gain level set by the gain level setting unit 140. Specifically, the gain control unit 141 controls the gain of the differential signal using the gain level set by the gain level setting unit 140 in at least one of the plurality of bands divided by the band dividing unit 130.

  Then, the synthesis unit 131 obtains an acoustic signal to be output by synthesizing the signal output from the gain control unit 141 and the signal output directly from the band dividing unit 130 to the synthesis unit 131. Specifically, the synthesizing unit 131 includes a difference signal in a band that is gain-controlled by the gain control unit 141 and a difference signal in a band that is not subjected to gain control by the gain control unit 141 among a plurality of bands divided by the band dividing unit 130. And synthesize.

  As described above, according to this configuration example, the signal processing apparatus 100 efficiently reduces the discomfort of the auditory noise given to the user by performing gain control mainly in a band in which the auditory noise is conspicuous. It is possible. Moreover, since the signal processing apparatus 100 according to the present configuration example performs gain control in a part of the band, it is possible to prevent the volume of the entire output acoustic signal from being excessively reduced.

<4. Third Embodiment>
This embodiment is a combination of the first embodiment and the second embodiment described above. Hereinafter, a configuration example of the signal processing apparatus 100 according to the present embodiment will be described with reference to FIG.

[4-1. Configuration example]
FIG. 24 is a block diagram illustrating an example of a logical configuration of the signal processing apparatus 100 according to the present embodiment. As illustrated in FIG. 24, the signal processing device 100 according to the present embodiment includes a difference signal calculation unit 110, a band division unit 130, a blur processing unit 120, a delay amount setting unit 123, a coefficient setting unit 124, and a blur level calculation unit 125. , A gain level setting unit 140, a gain control unit 141, and a synthesis unit 131.

  The difference signal calculation unit 110 outputs a difference signal. Next, the band dividing unit 130 divides the differential signal into a plurality of bands. Specifically, the band dividing unit 130 divides a band to be subjected to blurring processing by the blurring processing unit 120 and gain control by the gain control unit 141 and a band to be excluded. For example, the band dividing unit 130 outputs a band in which auditory noise is conspicuous to the blurring processing unit 120, and outputs the other band to the synthesizing unit 131.

  Next, the blurring processing unit 120 performs a blurring process on the band difference signal output from the band dividing unit 130. Specifically, the blurring processing unit 120 performs the blurring processing in at least one band among a plurality of bands divided by the band dividing unit 130. At this time, the blurring processing unit 120 uses the delay amount n set by the delay amount setting unit 123 and the weighting factor r set by the coefficient setting unit 124 to obtain the blur signal F (i) according to the above equation 2.

Here, the coefficient setting unit 124 may perform the processing described in the second configuration example of the first embodiment, or may perform the processing described in the third configuration example of the second embodiment. Good. That is, the coefficient setting unit 124 may set the weighting coefficient r based on the audio codec of the input acoustic signal, or the weighting coefficient according to the blur level f (i) calculated by the blur level calculation unit 125. r may be set. For example, if the former weighting coefficient is r1 and the latter weighting coefficient is r2, the coefficient setting unit 124 may adopt the maximum value as the weighting coefficient r as shown in the following equation.
r (i) = MAX (r1 (i), r2 (i)) (Formula 14)

  The coefficient setting unit 124 may set the weighting coefficient r by combining r1 and r2. For example, the coefficient setting unit 124 may set the weighting coefficient r by the average value of r1 and r2. That is, the magnitude relationship between r1 and r2 may be reflected in the weighting coefficient r.

The gain control unit 141 performs gain control of the blur signal output from the blur processing unit 120. Specifically, the gain control unit 141 uses the gain level set by the gain level setting unit 140 to control the gain of the signal subjected to the blur processing by the blur processing unit 120. For example, the gain control unit 141 obtains a signal G (i) whose gain is controlled using the following mathematical formula.
G (i) = g (i) × F (i) (Formula 15)

  Then, the synthesis unit 131 obtains an acoustic signal to be output by synthesizing the signal output from the gain control unit 141 and the signal output directly from the band dividing unit 130 to the synthesis unit 131. Specifically, the synthesis unit 131 synthesizes the signal gain-controlled by the gain control unit 141 and the difference signal in the band that has not been gain-controlled by the gain control unit 141 among the plurality of bands divided by the band division unit 130. To do.

  Note that the blur level calculation unit 125 and the gain level setting unit 140 may use the scale t (i) shown in the above formulas 8 to 10 in common as the scale of the visibility of auditory noise. Different scales may be employed.

  The order of processing of the blur processing unit 120 and the gain level setting unit 140 may be reversed.

  The configuration example of the signal processing device 100 according to the present embodiment has been described above. Subsequently, an operation process of the signal processing apparatus 100 according to the present embodiment will be described.

[4-2. Operation processing example]
FIG. 25 is a flowchart illustrating an example of the flow of signal processing executed in the signal processing device 100 according to the present embodiment.

  As shown in FIG. 25, first, in step S <b> 402, the difference signal calculation unit 110 accepts input of an i-th Lch signal L (i) and an Rch signal R (i).

  Next, in step S404, the difference signal calculation unit 110 calculates the difference signal S (i). For example, the difference signal calculation unit 110 calculates the difference signal S (i) using the above Equation 1.

  Next, in step S406, the gain level setting unit 140 calculates the gain level g (i). For example, the gain level setting unit 140 calculates the gain level g (i) using the above formulas 8 to 12.

  Next, in step S408, the delay amount setting unit 123 calculates the delay amount n, and the coefficient setting unit 124 calculates the weighting coefficient r. For example, the delay amount setting unit 123 calculates the delay amount n using Equations 5 and 6 above. For example, the coefficient setting unit 124 calculates the weighting coefficient r using the above formula 14.

  Next, in step S410, the band dividing unit 130 divides the difference signal S (i) into a band to be processed and a band not to be processed. The processing target here refers to a target of blur processing by the blur processing unit 120 and gain control by the gain control unit 141. For example, the band dividing unit 130 divides the differential signal S (i) into a band in which auditory noise is conspicuous and a band in which auditory noise is not conspicuous. To do.

  Next, in step S412, the blur processing unit 120 calculates the blur signal F (i) in the processing target band. For example, the blurring processing unit 120 calculates the blurring signal F (i) using the above Equation 2 for a difference signal in a band in which auditory noise is conspicuous among a plurality of bands divided by the band dividing unit 130.

  Next, in step S414, the gain control unit 141 calculates a signal G (i) whose gain is controlled in the band to be processed. For example, the gain control unit 141 calculates a signal G (i) whose gain is controlled using the formula 15 for the blur signal F (i) output from the blur processing unit 120.

  Next, in step S416, the synthesizer 131 synthesizes the signal after processing in steps S412 and S414 and the signal not to be processed. For example, the synthesizing unit 131 uses the signal G (i) whose gain in the processing target band whose gain is controlled in step S414 and the difference signal S (i) in the non-processing band divided in step S410. ) And.

  In step S418, the synthesis unit 131 outputs the signal synthesized in step S416.

  As described above, according to this embodiment, the signal processing apparatus 100 can achieve the effects of the first embodiment and the second embodiment, and more effectively prevent auditory noise. Can do.

<5. Fourth Embodiment>
In the present embodiment, the signal processing apparatus 100 performs vocal suppression processing on a frequency domain signal. Hereinafter, the present embodiment will be described with reference to FIGS. 26 and 27.

  FIG. 26 is a block diagram illustrating an example of a logical configuration of the signal processing apparatus 100 according to the present embodiment. As shown in FIG. 26, the signal processing apparatus 100 according to the present embodiment includes an FFT unit 150, a difference signal calculation unit 110, a blurring processing unit 120, and an IFFT unit 151.

(1) FFT unit 150
The FFT unit 150 has a function of converting an input time domain signal into a frequency domain signal. For example, the FFT unit 150 converts a time domain signal into a frequency domain signal by FFT. Any method other than FFT may be employed for this conversion processing. Further, when the input acoustic signal is a frequency domain signal, the FFT unit 150 may be omitted. When the input acoustic signal is a frequency domain signal, the first to third embodiments require a step of converting the frequency domain signal into a time domain signal. On the other hand, since the signal processing apparatus 100 according to the present embodiment can omit this step, the processing is made efficient.

(2) Difference signal calculation unit 110
The difference signal calculation unit 110 according to the present embodiment calculates a difference signal in the frequency domain. For example, the difference signal calculation unit 110 calculates a difference signal by subtracting the power of the corresponding scale factor band for Lch and Rch. The difference signal calculation unit 110 may subtract Rch from Lch or subtract Lch from Rch.

(3) Blur processing unit 120
The blurring processing unit 120 according to the present embodiment adds a frequency domain signal obtained by processing the differential signal to the frequency domain differential signal calculated by the differential signal calculation unit 110. For example, the blurring processing unit 120 generates a delayed signal obtained by delaying the differential signal as a signal obtained by processing the differential signal. Then, the blurring processing unit 120 adds the frequency domain delay signal to the frequency domain differential signal. Hereinafter, an example of the signal flow of the blur processing unit 120 according to the present embodiment will be described with reference to FIG.

  FIG. 27 is a diagram illustrating an example of a signal flow of the blur processing unit 120 according to the present embodiment. FIG. 27 illustrates an example in which blurring processing is performed using the spectra of two temporally continuous frames. As shown in FIG. 27, the blurring processing unit 120 includes a delay unit 122 that delays an input signal by one frame, and blurs the difference signal S (i) by weighting and adding a delayed signal delayed by one frame. A signal F (i) is obtained. Reference numerals 401 and 402 indicate the power for each scale factor band of the differential signal S (i). For example, reference numeral 401 is the power for each scale factor band of the h-th frame of the difference signal, and reference numeral 402 is the power for each scale factor band of the (h−1) -th frame of the difference signal. Reference numeral 403 indicates the power for each scale factor band of the blur signal F (i). Specifically, reference numeral 403 denotes power for each scale factor band of a signal obtained by weighted averaging the signal indicated by reference numeral 401 and the signal indicated by reference numeral 402 with a weight of 0.5. As indicated by reference numeral 403, the steepness of the change in the time direction of the power for each scale factor band of the output signal F (i) is suppressed, and as a result, auditory noise is suppressed.

  In the example shown in FIG. 27, for simplification of explanation, the blurring processing unit 120 has one delay unit 122 and the weighting coefficient r = 0.5, but other arbitrary settings may be used. May be. Further, in FIG. 27, the example in which the delay signal is generated using the FIR filter has been described, but an IIR filter may be used.

(4) IFFT unit 151
The IFFT unit 151 has a function of converting an input frequency domain signal into a time domain signal. For example, the IFFT unit 151 converts a time-domain signal into a frequency-domain signal by IFFT. Any method other than IFFT may be employed for this conversion processing. If the output signal is a frequency domain signal, IFFT unit 151 may be omitted.

  As described above, according to the present embodiment, it is possible to prevent the generation of auditory noise while suppressing the specific sound for the acoustic signal in the frequency domain.

<6. Hardware configuration>
Finally, the hardware configuration of the information processing apparatus according to the present embodiment will be described with reference to FIG. FIG. 28 is a block diagram illustrating an example of a hardware configuration of the information processing apparatus according to the present embodiment. Note that the information processing apparatus 900 illustrated in FIG. 28 is, for example, the signal processing apparatus 100 according to each embodiment illustrated in FIGS. 1, 16, 17, 19, 20, 23, 24, and 26. Can be realized. Information processing by the signal processing apparatus 100 according to each embodiment is realized by cooperation of software and hardware described below.

  As shown in FIG. 28, the information processing apparatus 900 includes a CPU (Central Processing Unit) 901, a ROM (Read Only Memory) 902, a RAM (Random Access Memory) 903, and a host bus 904a. The information processing apparatus 900 includes a bridge 904, an external bus 904b, an interface 905, an input device 906, an output device 907, a storage device 908, a drive 909, a connection port 911, a communication device 913, and a sensor 915. The information processing apparatus 900 may include a processing circuit such as a DSP or an ASIC in place of or in addition to the CPU 901.

  The CPU 901 functions as an arithmetic processing device and a control device, and controls the overall operation in the information processing device 900 according to various programs. Further, the CPU 901 may be a microprocessor. The ROM 902 stores programs used by the CPU 901, calculation parameters, and the like. The RAM 903 temporarily stores programs used in the execution of the CPU 901, parameters that change as appropriate during the execution, and the like. The CPU 901 forms, for example, each component included in the signal processing device 100 according to each embodiment illustrated in FIGS. 1, 16, 17, 19, 20, 23, 24, and 26. obtain.

  The CPU 901, ROM 902 and RAM 903 are connected to each other by a host bus 904a including a CPU bus. The host bus 904a is connected to an external bus 904b such as a PCI (Peripheral Component Interconnect / Interface) bus via a bridge 904. Note that the host bus 904a, the bridge 904, and the external bus 904b do not necessarily have to be configured separately, and these functions may be mounted on one bus.

  The input device 906 is realized by a device to which information is input by a user, such as a mouse, a keyboard, a touch panel, a button, a microphone, a switch, and a lever. The input device 906 may be, for example, a remote control device using infrared rays or other radio waves, or may be an external connection device such as a mobile phone or a PDA that supports the operation of the information processing device 900. . Furthermore, the input device 906 may include, for example, an input control circuit that generates an input signal based on information input by the user using the above-described input means and outputs the input signal to the CPU 901. A user of the information processing apparatus 900 can input various data and instruct a processing operation to the information processing apparatus 900 by operating the input device 906.

  The output device 907 is formed of a device capable of visually or audibly notifying acquired information to the user. Examples of such devices include CRT display devices, liquid crystal display devices, plasma display devices, EL display devices, display devices such as lamps, audio output devices such as speakers and headphones, printer devices, and the like. For example, the output device 907 outputs results obtained by various processes performed by the information processing device 900. Specifically, the display device visually displays results obtained by various processes performed by the information processing device 900 in various formats such as text, images, tables, and graphs. On the other hand, the audio output device converts an audio signal composed of reproduced audio data, acoustic data, and the like into an analog signal and outputs it aurally.

  The storage device 908 is a data storage device formed as an example of a storage unit of the information processing device 900. The storage apparatus 908 is realized by, for example, a magnetic storage device such as an HDD, a semiconductor storage device, an optical storage device, a magneto-optical storage device, or the like. The storage device 908 may include a storage medium, a recording device that records data on the storage medium, a reading device that reads data from the storage medium, a deletion device that deletes data recorded on the storage medium, and the like. The storage device 908 stores programs executed by the CPU 901, various data, various data acquired from the outside, and the like.

  The drive 909 is a storage medium reader / writer, and is built in or externally attached to the information processing apparatus 900. The drive 909 reads information recorded on a removable storage medium such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and outputs the information to the RAM 903. The drive 909 can also write information to a removable storage medium.

  The connection port 911 is an interface connected to an external device, and is a connection port with an external device capable of transmitting data by, for example, USB (Universal Serial Bus).

  The communication device 913 is a communication interface formed by a communication device for connecting to the network 920, for example. The communication device 913 is, for example, a communication card for wired or wireless LAN (Local Area Network), LTE (Long Term Evolution), Bluetooth (registered trademark), or WUSB (Wireless USB). The communication device 913 may be a router for optical communication, a router for ADSL (Asymmetric Digital Subscriber Line), a modem for various communication, or the like. The communication device 913 can transmit and receive signals and the like according to a predetermined protocol such as TCP / IP, for example, with the Internet and other communication devices.

  The network 920 is a wired or wireless transmission path for information transmitted from a device connected to the network 920. For example, the network 920 may include a public line network such as the Internet, a telephone line network, a satellite communication network, various LANs (Local Area Network) including Ethernet (registered trademark), a WAN (Wide Area Network), and the like. The network 920 may also include a dedicated line network such as an IP-VPN (Internet Protocol-Virtual Private Network).

  Heretofore, an example of the hardware configuration capable of realizing the functions of the information processing apparatus 900 according to the present embodiment has been shown. Each of the above components may be realized using a general-purpose member, or may be realized by hardware specialized for the function of each component. Therefore, it is possible to change the hardware configuration to be used as appropriate according to the technical level at the time of carrying out this embodiment.

  Note that a computer program for realizing each function of the information processing apparatus 900 according to the present embodiment as described above can be produced and mounted on a PC or the like. In addition, a computer-readable recording medium storing such a computer program can be provided. The recording medium is, for example, a magnetic disk, an optical disk, a magneto-optical disk, a flash memory, or the like. Further, the above computer program may be distributed via a network, for example, without using a recording medium.

<7. Summary>
The embodiment of the present disclosure has been described in detail above with reference to FIGS. As described above, the signal processing apparatus 100 according to the present embodiment calculates the difference signal between the acoustic signal of the first channel and the acoustic signal of the second channel that forms the input acoustic signal, and generates the difference signal. The signal obtained by processing the difference signal is added. By adding the signal obtained by processing the difference signal to the difference signal, the signal processing apparatus 100 can alleviate a sharp level change in the time direction and prevent generation of annoying auditory noise. This effect is remarkably obtained when the input acoustic signal is compressed by a joint stereo encoding method or the like. According to the present embodiment, it is possible to directly mitigate a steep level change in the time direction, which is one of the major causes of auditory noise. For this reason, the signal processing apparatus 100 according to the present embodiment has a higher effect of preventing the generation of auditory noise and is more efficient than the second comparative example that can indirectly relieve a steep level change in the time direction. It is considered to be appropriate. Moreover, since the signal processing apparatus 100 adds the signal which processed the differential signal by which the specific sound was suppressed, it is possible to implement | achieve a high suppression performance, without paying the suppression performance of a specific sound.

  The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, but the technical scope of the present disclosure is not limited to such examples. It is obvious that a person having ordinary knowledge in the technical field of the present disclosure can come up with various changes or modifications within the scope of the technical idea described in the claims. Of course, it is understood that it belongs to the technical scope of the present disclosure.

  For example, the signal processing apparatus 100 according to the present embodiment can be mounted on various devices. For example, when the signal processing apparatus 100 according to the present embodiment is installed in an apparatus that reproduces a sound source such as a stereo component system, the user can easily enjoy karaoke by playing while suppressing vocals of music. it can. In addition, when the signal processing apparatus 100 according to the present embodiment is mounted on an apparatus that reproduces a voice guide such as a car navigation system, the signal processing apparatus 100 uses the vocal of the music being played when the voice guide is played It may be suppressed. In this case, since the voice guide is prevented from being erased by the vocal of the music, the user can hear the voice guide clearly while enjoying the reproduction of the music.

  Each device described in this specification may be realized as a single device, or a part or all of the devices may be realized as separate devices. For example, some or all of the components of the signal processing device 100 may be provided in a device such as a server connected by a network or the like. Processing may be performed.

  Further, the processing described with reference to the flowcharts and sequence diagrams in this specification may not necessarily be executed in the order shown. Some processing steps may be performed in parallel. Further, additional processing steps may be employed, and some processing steps may be omitted.

  Further, the effects described in the present specification are merely illustrative or exemplary and are not limited. That is, the technology according to the present disclosure can exhibit other effects that are apparent to those skilled in the art from the description of the present specification in addition to or instead of the above effects.

The following configurations also belong to the technical scope of the present disclosure.
(1)
A differential signal calculation unit that calculates a differential signal between the acoustic signal of the first channel and the acoustic signal of the second channel forming the input acoustic signal;
A processing unit for adding a signal obtained by processing the differential signal to the differential signal calculated by the differential signal calculating unit;
A signal processing apparatus comprising:
(2)
The signal processing apparatus according to (1), wherein the processing unit generates a delayed signal obtained by delaying the differential signal as a signal obtained by processing the differential signal.
(3)
The signal processing device according to (2), further including a delay amount setting unit that sets a delay amount of the delay signal.
(4)
The signal processing device according to (3), wherein the delay amount setting unit sets the delay amount using compression encoding information of the input acoustic signal.
(5)
The signal processing device according to (4), wherein the delay amount setting unit sets the delay amount to be equal to or less than a frame width indicated by the compression encoding information.
(6)
The signal processing apparatus according to any one of (2) to (5), wherein the processing unit generates the delayed signal using an IIR (Infinite impulse response) filter.
(7)
The signal processing apparatus according to any one of (2) to (5), wherein the processing unit generates the delayed signal using a FIR (Finite impulse response) filter.
(8)
The signal processing device according to any one of (2) to (6), further including a coefficient setting unit that sets a weighting coefficient related to the addition performed by the processing unit.
(9)
The signal processing apparatus according to (8), wherein the coefficient setting unit sets the weighting coefficient based on compression coding information of the input acoustic signal.
(10)
The signal processing device according to (8) or (9), wherein the coefficient setting unit sets the weighting coefficient based on a degree to which the input acoustic signal is close to monaural.
(11)
The signal processing device includes:
A band dividing unit for dividing the difference signal into a plurality of bands;
A combining unit that combines the plurality of difference signals divided by the band dividing unit;
Further comprising
The processing unit adds a signal obtained by processing the difference signal to the difference signal in at least one band among a plurality of bands divided by the band dividing unit,
The synthesizing unit synthesizes the difference signal of the band processed by the processing unit and the difference signal of the band not processed by the processing unit among the plurality of bands divided by the band dividing unit. The signal processing device according to any one of (1) to (10).
(12)
The signal processing device includes:
A gain level setting unit for setting a gain level of the differential signal;
A gain control unit for controlling the gain of the differential signal using the gain level set by the gain level setting unit;
The signal processing apparatus according to any one of (1) to (11), further including:
(13)
The signal processing apparatus according to (12), wherein the gain level setting unit sets the gain level based on a degree to which the input acoustic signal is close to monaural.
(14)
The signal processing device includes:
A band dividing unit for dividing the difference signal into a plurality of bands;
A combining unit that combines the plurality of difference signals divided by the band dividing unit;
Further comprising
The gain control unit controls the gain of the differential signal using the gain level set by the gain level setting unit in at least one of the plurality of bands divided by the band dividing unit,
The synthesizing unit synthesizes the differential signal in the band controlled by the gain control unit and the differential signal in a band not controlled by the gain control unit among a plurality of bands divided by the band dividing unit. The signal processing device according to (12) or (13).
(15)
The processing unit adds a signal obtained by processing the difference signal to the difference signal in at least one band among a plurality of bands divided by the band dividing unit,
The gain control unit controls the gain of the signal processed by the processing unit using the gain level set by the gain level setting unit,
The synthesizing unit synthesizes the signal controlled by the gain control unit and the differential signal in a band not controlled by the gain control unit among a plurality of bands divided by the band dividing unit; 14) The signal processing apparatus according to 14).
(16)
The signal processing device according to any one of (1) to (15), wherein the difference signal calculation unit calculates the difference signal in a time domain.
(17)
The signal processing device according to any one of (1) to (15), wherein the difference signal calculation unit calculates the difference signal in a frequency domain.
(18)
Calculating a differential signal between the acoustic signal of the first channel and the acoustic signal of the second channel forming the input acoustic signal;
Adding a signal obtained by processing the difference signal to the calculated difference signal by a processor;
A signal processing method including:
(19)
Computer
A differential signal calculation unit that calculates a differential signal between the acoustic signal of the first channel and the acoustic signal of the second channel forming the input acoustic signal;
A processing unit for adding a signal obtained by processing the differential signal to the differential signal calculated by the differential signal calculating unit;
Program to function as.

DESCRIPTION OF SYMBOLS 100 Signal processing apparatus 110 Differential signal calculation part 120 Blur processing part 121 Delay buffer DB
122 delay unit 123 delay amount setting unit 124 coefficient setting unit 125 blurring level calculation unit 130 band division unit 131 synthesis unit 140 gain level setting unit 141 gain control unit 150 FFT unit 151 IFFT unit

Claims (14)

A differential signal calculation unit that calculates a differential signal between the acoustic signal of the first channel and the acoustic signal of the second channel forming the input acoustic signal;
A processing unit for adding a delayed signal obtained by delaying the differential signal to the differential signal calculated by the differential signal calculating unit;
A delay amount setting unit for setting a delay amount of the delay signal to be equal to or less than a frame width indicated by the compression encoding information of the input acoustic signal;
A signal processing apparatus comprising: The signal processing apparatus according to claim 1 , wherein the processing unit generates the delayed signal using an IIR (Infinite impulse response) filter. The signal processing apparatus according to claim 1 , wherein the processing unit generates the delayed signal using a FIR (Finite impulse response) filter. The signal processing unit, the processing unit further comprising a coefficient setting unit for setting a weighting coefficient according to the addition by the signal processing apparatus according to any one of claims 1 to 3. The signal processing device according to claim 4 , wherein the coefficient setting unit sets the weighting coefficient based on a bit rate indicated by the compression encoding information of the input acoustic signal and / or a use situation of joint stereo encoding. . The signal processing device according to claim 4 , wherein the coefficient setting unit sets the weighting coefficient based on a degree to which the input acoustic signal is close to monaural. The signal processing device includes:
A band dividing unit for dividing the difference signal into a plurality of bands;
A combining unit that combines the plurality of difference signals divided by the band dividing unit;
Further comprising
The processing unit adds the delay signal to the differential signal in at least one band among a plurality of bands divided by the band dividing unit,
The synthesizing unit synthesizes the difference signal of the band processed by the processing unit and the difference signal of the band not processed by the processing unit among the plurality of bands divided by the band dividing unit. The signal processing device according to any one of claims 1 to 6 . The signal processing device includes:
A gain level setting unit for setting a gain level of the differential signal;
A gain control unit that controls the gain of the signal obtained by adding the delay signal to the differential signal using the gain level set by the gain level setting unit;
The signal processing apparatus according to any one of claims 1 to 6 , further comprising: The signal processing apparatus according to claim 8 , wherein the gain level setting unit sets the gain level based on a degree to which the input acoustic signal is close to monaural. The signal processing device includes:
A band dividing unit for dividing the difference signal into a plurality of bands;
A combining unit that combines the plurality of difference signals divided by the band dividing unit;
Further comprising
The processing unit adds the delay signal to the differential signal in at least one band among a plurality of bands divided by the band dividing unit,
The gain control unit controls the gain of the signal processed by the processing unit using the gain level set by the gain level setting unit,
The synthesizer synthesizes the signal controlled by the gain controller and the differential signal in a band not controlled by the gain controller among a plurality of bands divided by the band divider. The signal processing device according to 8 or 9 . The difference signal calculating unit calculates the difference signal in the time domain, the signal processing apparatus according to any one of claims 1-10. The difference signal calculating unit calculates the difference signal in the frequency domain, the signal processing apparatus according to any one of claims 1-10. Calculating a differential signal between the acoustic signal of the first channel and the acoustic signal of the second channel forming the input acoustic signal;
Adding a delayed signal obtained by delaying the differential signal to the calculated differential signal by a processor;
Setting the delay amount of the delay signal to be equal to or less than the frame width indicated by the compression encoding information of the input acoustic signal;
A signal processing method including: Computer
A differential signal calculation unit that calculates a differential signal between the acoustic signal of the first channel and the acoustic signal of the second channel forming the input acoustic signal;
A processing unit for adding a delayed signal obtained by delaying the differential signal to the differential signal calculated by the differential signal calculating unit;
A delay amount setting unit for setting a delay amount of the delay signal to be equal to or less than a frame width indicated by the compression encoding information of the input acoustic signal;
Program to function as.

Images