JP2010171585A - Method, device and program for separating sound source - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To separate a sound source satisfactorily even when sparseness is not established in a received sound signal outputted from main/sub microphones and there is no amplitude difference between time frequency components. <P>SOLUTION: Time frequency analyzing portions 11 and 12 analyze received sound signals x1(t) and x2(t) outputted from the main/sub microphones, and convert them into time frequency components X1(f, t) and X2(f, t), respectively. A gain giving portion 13 gives a gain Gf to the X2(f, t). A level difference comparing portion 14 compares amplitudes of the time frequency components X1(f, t) and GfX2(f, t) by time frequency component, and generates a mask pattern m1(f, t). A masking processing portion 15 uses the mask pattern m1(f, t) to mask the time frequency component X1(f, t). A time frequency synthesizing portion 16 synthesizes time frequency components Y1(f, t) outputted from the masking processing portion 15. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a sound source separation method, apparatus, and program, and more particularly, to a sound source separation method, apparatus, and program for separating and extracting a target sound and an interference sound based on sound reception signals output from two microphones.

  In noisy environments such as streets, cars, and station platforms, even if you use a microphone placed close to your mouth, such as a handset or headset, you can use the target sound as the target sound and other sounds that are interference sounds. And ambient noise may be mixed. In order to solve this problem, various interference sound suppression methods and sound source separation methods have been proposed so far. These methods can be broadly classified into those using a single microphone and those using a plurality of microphones. In the case of using a plurality of microphones, it is possible to obtain a higher disturbing sound suppression performance than in the case of using a single microphone.

  In the method using multiple microphones, multiple microphones are arranged spatially, and the received signal output from each microphone reflects the time difference or amplitude difference depending on the spatial positional relationship between each microphone and the sound source. . According to this, only the target sound is selectively collected or the target sound and the interfering sound are separated using the statistical information of the time difference and the amplitude difference of the received sound signals output from each microphone. be able to.

  As a technique using a plurality of microphones, a technique called time-frequency masking using the sparsity of an audio signal has been proposed. The sparseness of an audio signal refers to the property that the energy of the audio signal is concentrated in a part of the time frequency domain and is almost zero in the other time frequency domain.

  In the method based on time frequency masking, the direction of the target sound and the disturbing sound may be unknown, and in order to extract the target sound, one or both of the amplitude difference and the time difference of each time frequency component of the received signal output from the plurality of microphones is used. Calculate both. And each time frequency component is classified based on those differences, and the target sound and the interference sound are separated. In calculating the amplitude difference and the time difference of each time frequency component of the received sound signal output from the plurality of microphones, frequency analysis is performed for each predetermined time length.

  Among the methods based on time-frequency masking, especially those that use the amplitude difference of each time-frequency component of the received signal output by multiple microphones, the computer calculates the auditory masking phenomenon in which stronger signals mask weaker signals. Simulated above. When two microphones are used, a mask pattern that masks the interference sound superimposed on the target sound is generated by comparing the amplitude difference of each time frequency component of the received sound signal output by the two microphones, and is stored in the main microphone. It is used for selectively extracting the time frequency component of the high amplitude received sound signal of the adjacent sound source. This processing is performed in the time frequency domain, and the frequency component in which the received sound signal output from the main microphone of the two microphones is dominant is output as it is, and the received sound signal output from the other sub microphone is dominant. Masking is performed on the frequency components that are not correct. Mask processing for the received sound signal of the sound source close to the main microphone is defined by the following equation (1).

In this masking process, it is assumed that the received sound signals output from the main and sub microphones are sparse and there is an amplitude difference between their time frequency components. This is described in Non-Patent Documents 1-4.
RFLyon: "A computational model of binaural localization and separation," In Proc. ICASSP, 1983. M. Bodden: "Modeling human sound-source localization and the cocktail-party-effect," Acta Acoustica, vol.1, pp.43--55, 1993. O. Yilmaz and S. Rickard: "Blind Separation of Speech Mixtures via Time-Frequency Masking," IEEE Transaction on Signal Processing, Vol. 52, No. 7, pp. 1830-1847, 2004. S. Rickard and O. Yilmaz: "On the Approximate W-disjoint Orthogonality of Speech," Proc. ICASSP, Vol. I, pp. 529-532, 2002.

  However, in general, a sparseness is established in a received signal using a human sound source, but a sparseness is not established in a received signal of a disturbing sound (ambient noise), for example. Further, in the sound reception signals output from the two microphones, even if there is an amplitude difference between the target sound reception signals, there is often no amplitude difference between the interference sound reception signals. As a result, there is a problem that the conventional interference sound suppression method and sound source separation method cannot obtain sufficient interference sound suppression and sound source separation performance.

  The object of the present invention is to solve the above-mentioned problems, and even if the received sound signals output from the two microphones are not sparse and there is no difference in amplitude between their time-frequency components, interference sound suppression and sound source separation are achieved. It is an object of the present invention to provide a sound source separation device, method and program in which performance does not deteriorate.

  In order to achieve the above object, a sound source separation apparatus of the present invention is a sound source separation apparatus that separates and outputs at least one of a target sound component and an interference sound component from a received sound signal output from a main / sub microphone, Conversion means for converting the received sound signals output from the main and sub microphones into time frequency components, respectively, and the time and frequency components provided in at least one of the signal paths of the main and sub microphones. A gain applying means for applying a gain to the received sound signal before being converted into the time frequency component, or a time frequency component after being converted into the time frequency component, and a gain is applied by the gain applying means, and converted by the converting means Level difference comparison means for comparing the amplitudes of subsequent time frequency components for each time frequency component and generating a mask pattern, and the gain applying means A masking processing unit that masks at least one of the time frequency components after the gain is given and converted by the conversion unit using a mask pattern generated by the level difference comparison unit, and the masking processing unit outputs the masking processing unit And a time-frequency synthesizing unit for synthesizing the time-frequency components.

  Further, in the sound source separation device of the present invention, the gain applied by the gain applying unit causes an amplitude difference between the time frequency components of the received sound signal output from the main and sub microphones with respect to the disturbing sound, and It is characterized in that it is set to a constant value or a frequency-dependent value so that the magnitude relationship of the amplitudes of the time frequency components of the received sound signals output from the main and sub microphones with respect to the target sound is not reversed.

  Further, in the sound source separation device of the present invention, the gain applying unit applies a gain of a frequency dependent value to a time frequency component in the signal path of the main microphone, and is further input to the masking processing unit through the signal path of the main microphone. Gain removing means for removing the gain by performing a process reverse to the gain applying by the gain applying means on the time frequency component of the main microphone signal path output from the masking processing means. It is characterized by that.

  The present invention is characterized not only as a sound source separation device but also as a sound source separation method specified by a processing procedure of received sound signals, and a program for causing a computer to realize sound source separation and interference sound suppression functions There are also features.

  In the present invention, at least one of the signal paths of the main / sub microphones is provided with a gain applying means, and a sound receiving signal passing through the signal path or a sound receiving signal output from the main / sub microphone by applying gain to a time frequency component. A difference in amplitude is generated between the time frequency components of the two, and then a mask pattern is generated, so that the received signal output from the main and sub microphones is not sparse, and there is no difference in amplitude between these time frequency components. Even in such a case, the sound source can be satisfactorily separated without deteriorating the performance of sound source separation and interference sound suppression.

  The present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a sound source separation device according to the present invention. The present invention can be realized not only as a sound source separation device but also as a sound source separation method specified by a received signal processing procedure, and as a program for causing a computer to realize sound source separation and interference sound suppression functions. Can also be realized. Each unit in the sound source separation apparatus can be realized by hardware or software.

  The sound source separation device of FIG. 1 includes time frequency analysis units 11 and 12, a gain addition unit 13, a level difference comparison unit 14, a masking processing unit 15, and a time frequency synthesis unit 16. In this embodiment, the time path analysis unit 11, the masking processing unit 15 and the time frequency synthesis unit 16 form a signal path of the main microphone, and the time frequency analysis unit 12 and the gain applying unit 13 form a signal path of the sub microphone. ing.

  Each of the time frequency analysis units 11 and 12 analyzes the received sound signal output from the main and sub microphones in the time frequency domain, and outputs each time frequency component. The gain assigning unit 13 assigns a gain to each time frequency component of the input sound reception signal.

  The level difference comparison unit 14 compares the amplitude (level (absolute value)) of each time frequency component output from the time frequency analysis unit 11 and the gain applying unit 13 for each component, and masks based on the comparison result Generate a pattern.

  The masking processing unit 15 masks the time frequency component output from the time frequency analysis unit 11 according to the mask pattern generated by the level difference comparison unit. The time frequency synthesis unit 16 synthesizes the time frequency component output from the masking processing unit 15.

  Next, the operation of the sound source separation device of FIG. 1 will be described.

  The time frequency analysis unit 11 receives the sound reception signal x1 (t) output from the main microphone. In the case of a mobile terminal (for example, a mobile phone), the target sound is a voice in a call. The main microphone is arranged, for example, on the front surface of the mobile terminal in order to receive a high-level target sound. Although the main microphone is at a lower level than the target sound, it also receives interference sounds such as ambient noise. Therefore, the received sound signal x1 (t) is obtained by converting a high level target sound and a low level interference sound. The time frequency analysis unit 11 converts the received sound signal x1 (t) into a time frequency component X1 (f, t).

  On the other hand, the sound reception signal x2 (t) output from the sub microphone is input to the time frequency analysis unit 12. The sub microphone is disposed on the back surface of the mobile terminal, for example, in order to receive the interference sound. The sub microphone receives the target sound and the interference sound, although the level is lower than the target sound received by the main microphone. The target sound received by the sub microphone is at a considerably lower level than the target sound received by the main microphone, and the disturbing sound received by the sub microphone is at the same level as the disturbing sound received by the main microphone. The received sound signal x2 (t) is obtained by converting the interference sound and the low-level target sound. The time frequency analysis unit 12 converts the received sound signal x2 (t) into a time frequency component X2 (f, t).

  The gain assigning unit 13 assigns the gain Gf calculated in advance from the spatial positional relationship between the main and sub microphones, the nature of the interference sound, and the like to the time frequency component X2 (f, t), and the gain Gf is given. The time frequency component Gf · X2 (f, t) is transmitted. The gain Gf takes into account the amplitude difference between the received signals output by the main and sub microphones for the target sound, and the received signal of the interference sound is at a high level in the low frequency range and in the high frequency range. Considering the general property of low level, for example, it is set for each frequency component. The gain Gf is a frequency dependent value greater than 1.

  The amplitude difference between the received sound signals output from the main and sub microphones with respect to the target sound can be obtained by measuring the impulse responses of the main and sub microphones in advance. This amplitude difference depends on the distance between the target sound source and the main / sub microphone, the installation position of the main / sub microphone in the casing of the portable terminal, and the like. In addition, the nature of the received signal of the interfering sound can be obtained by measuring the received signal of various ambient sound sources and calculating the amplitude for each frequency of the received signal of the average ambient sound source from their frequency characteristics. Can do.

  The gain Gf provided by the gain applying unit 13 causes an amplitude difference between the time frequency components of the received signal output from the main and sub microphones for the disturbing sound, and the main and sub microphones for the target sound. What is necessary is just to make it the magnitude relationship of the amplitude of the time frequency component of each received sound signal not to be reversed.

  The level difference comparison unit 14 includes the level | X1 (f, t) | of the time frequency component X1 (f, t) output from the time frequency analysis unit 11 and the time frequency component Gf · X2 output from the gain applying unit 13. The level | Gf · X2 (f, t) | of (f, t) is compared, and a mask pattern m1 (f, t) is generated using the following equation (2). From the following equation (2), out of the time frequency component X1 (f, t) of the received sound signal x1 (t) output by the main microphone, the received sound signal x1 (t) output by the main microphone is not the dominant component A mask pattern m1 (f, t) for masking is generated. The mask pattern m1 (f, t) generated by the level difference comparison unit 14 is output to the masking processing unit 15.

  The masking processing unit 15 masks the time frequency component X1 (f, t) input from the time frequency analysis unit 11 with the mask pattern m1 (f, t). Therefore, from the masking processing unit 15, the sound reception signal x1 (t) output by the main microphone is dominant among the time frequency components X1 (f, t) of the sound reception signal x1 (t) output by the main microphone. Only the component Y1 (f, t) is output.

  The time frequency synthesizer 16 is a component Y1 in which the sound reception signal x1 (t) output from the main microphone is dominant among the time frequency components X1 (f, t) of the sound reception signal x1 (t) output from the main microphone. Only (f, t) is synthesized and the output signal y1 (t) is sent out.

  FIG. 2 is an explanatory diagram of the gain applying operation in the gain applying unit 13. FIGS. 9A and 9B show the amplitude (level) for each frequency component of the received sound signal output from the main / sub microphone at a certain time. Here, the white portion is the frequency component of the target sound reception signal, and the black portion is the frequency component of the interference sound reception signal.

For example, the frequency components near f ₁ is only received sound signal of the target sound, there is a rather large amplitude difference between the received sound signal mainly Vice microphone outputs, separate the target sound by utilizing this amplitude difference can do. However, the frequency components near f ₂ is only received sound signals of interference sound, the amplitude of the received sound signal mainly Vice microphone outputs are substantially the same. Since the magnitude relation between the amplitude varies depending on the situation, the frequency components near f ₂ is or separated as target sound, or isolated as a disturbing sound.

  Therefore, as shown in FIG. 2 (c), a gain Gf is given to each frequency component X2 (f, t) of the sound reception signal output from the sub microphone, and the main and sub microphones receive only the interference sound reception signal. Even in the case of output, an amplitude difference is generated between the frequency components of the received sound signals output from the main and sub microphones so that they are not separated as target sounds. Here, the target sound in the region is easily separated by lowering the gain Gf in the high frequency region.

  The gain Gf causes an amplitude difference between the frequency components of the received signal output by the main and auxiliary microphones for the disturbing sound, and the frequency component of the received signal output by the main and auxiliary microphones for the target sound. The amplitude difference between the two may not be canceled, that is, the magnitude relationship between the two may not be reversed.

However, the gain Gf can also be given so that the target sound with the specific frequency component or the target sound on which the interference sound is superimposed is not separated. For example, in FIG. 2 (b), the if very large gain Gf for the frequency components in the vicinity of f _3, will not be separated, including the target sound in the frequency component. The value of the gain Gf may be adjusted or selected.

  FIG. 3 is a block diagram showing another embodiment of the sound source separation device according to the present invention, and the same or equivalent parts as in FIG. In the present embodiment, the time-frequency analysis unit 12 and the gain applying unit 13 are arranged in the opposite direction to those in FIG. The signal path of the microphone is configured, and the signal path of the sub microphone is configured by the time frequency analysis unit 12 and the gain applying unit 13.

  The time frequency analysis unit 11 receives the sound reception signal x1 (t) output from the main microphone and converts the sound reception signal x1 (t) into a time frequency component X1 (f, t).

  The gain assigning unit 13 assigns a gain G calculated in advance from the spatial positional relationship between the main and sub microphones, the nature of the interference sound, and the like to the received sound signal x2 (t) output from the sub microphone. The received sound signal G · x2 (t) is output. The gain G is a constant value greater than 1.

  A sound reception signal x2 (t) output from the sub microphone is input to the time frequency analysis unit 12 via the gain applying unit 13. Therefore, the time-frequency analysis unit 12 converts the received sound signal G · x2 (t) to which the gain is given into the time-frequency component G · X2 (f, t).

  The level difference comparison unit 14 compares the level | X1 (f, t) | of the time frequency component X1 (f, t) output from the time frequency analysis unit 11 and the time frequency component G · The level | G · X2 (f, t) | of X2 (f, t) is compared, and a mask pattern m1 (f, t) is generated using the following equation (3). According to the following formula (3), among the time frequency components X1 (f, t) of the sound reception signal x1 (t) output from the main microphone, the sound reception signal x1 (t) output from the main microphone is not the dominant component. A mask pattern m1 (f, t) for masking is generated. The mask pattern m1 (f, t) generated by the level difference comparison unit 14 is output to the masking processing unit 15.

  The masking processing unit 15 masks the time frequency component X1 (f, t) input from the time frequency analysis unit 11 with the mask pattern m1 (f, t). Therefore, from the masking processing unit 15, the sound reception signal x1 (t) output by the main microphone is dominant among the time frequency components X1 (f, t) of the sound reception signal x1 (t) output by the main microphone. Only the component Y1 (f, t) is output.

  The time frequency synthesizer 16 is a component Y1 in which the sound reception signal x1 (t) output from the main microphone is dominant among the time frequency components X1 (f, t) of the sound reception signal x1 (t) output from the main microphone. Only (f, t) is synthesized and the output signal y1 (t) is sent out.

  FIG. 4 is a block diagram showing still another embodiment of the sound source separation device according to the present invention. The same or equivalent parts as those in FIG.

  In the present embodiment, the time-frequency analysis unit 11, the gain applying unit 13, the gain removal unit 17, the masking processing unit 15, and the time-frequency synthesis unit 16 form a signal path of the main microphone, and the time-frequency analysis unit 12 sets the sub microphone. A signal path is configured.

  The time frequency analysis unit 11 receives the sound reception signal x1 (t) output from the main microphone. The time frequency analysis unit 11 converts the received sound signal x1 (t) into a time frequency component X1 (f, t).

  The gain assigning unit 13 assigns the gain Gf calculated in advance from the spatial positional relationship between the main and sub microphones, the nature of the interference sound, and the like to the time frequency component X1 (f, t), and the gain Gf is given. Sends the time-frequency component Gf · X1 (f, t). The gain Gf is a frequency dependent value smaller than 1.

  On the other hand, the sound reception signal x2 (t) output from the sub microphone is input to the time frequency analysis unit 12. The time frequency analysis unit 12 converts the received sound signal x2 (t) into a time frequency component X2 (f, t).

  The level difference comparison unit 14 includes the level | Gf · X1 (f, t) | of the time frequency component Gf · X1 (f, t) output from the gain applying unit 13 and the time frequency output from the time frequency analysis unit 12. The level | X2 (f, t) | of the component X2 (f, t) is compared, and a mask pattern m1 (f, t) is generated using the following equation (4). According to the following formula (4), among the time frequency components X1 (f, t) of the sound reception signal x1 (t) output from the main microphone, the sound reception signal x1 (t) output from the main microphone is not the dominant component. A mask pattern m1 (f, t) for masking is generated. The mask pattern m1 (f, t) generated by the level difference comparison unit 14 is output to the masking processing unit 15.

  The gain removing unit 17 performs a process reverse to that of the gain applying unit 13 on the time frequency component Gf · X1 (f, t), and outputs the time frequency component X1 (f, t) to the masking processing unit 15. The gain removing unit 17 is provided to eliminate distortion of the output signal y1 (t) due to gain application by the gain applying unit 13, but can be omitted when distortion is allowable. The gain removing unit 17 may be provided on the output side of the masking processing unit 15.

  The masking processing unit 15 masks the time frequency component X1 (f, t) input from the gain removing unit 17 with the mask pattern m1 (f, t). Therefore, from the masking processing unit 15, the sound reception signal x1 (t) output by the main microphone is dominant among the time frequency components X1 (f, t) of the sound reception signal x1 (t) output by the main microphone. Only the component Y1 (f, t) is output.

  The time frequency synthesizer 16 is a component Y1 in which the sound reception signal x1 (t) output from the main microphone is dominant among the time frequency components X1 (f, t) of the sound reception signal x1 (t) output from the main microphone. Only (f, t) is synthesized and the output signal y1 (t) is sent out.

  Although the embodiment has been described above, the present invention is not limited to the above embodiment and can be variously modified. For example, the gain applying unit is provided in the signal path of the sub microphone in the embodiment of FIGS. 1 and 3, and is provided in the signal path of the main microphone in the embodiment of FIG. It is also possible to provide gain imparting sections on both and adjust their gains. However, when gain Gf (frequency-dependent value) is given in the signal path of the main microphone, the received sound signal output from the main microphone is transformed by the gain Gf, so that gain removal is performed as in the embodiment of FIG. It is preferable to provide a part.

  Moreover, although the said embodiment isolate | separates and outputs the received signal of a target sound, in addition to this, the received signal of a disturbing sound is isolate | separated and output, or only the received signal of a disturbing sound is output. It can also be output separately. The received signal of the disturbing sound can be used, for example, for measurement of ambient noise, extraction of sound coming from the back side of the mobile terminal, and the like.

  FIG. 5 is a block diagram illustrating a modified example in which the received sound signals of the target sound and the interference sound are separated and output. In the figure, the time-frequency analysis units 11 and 12, the gain applying unit 13, the masking processing unit 15 and the time-frequency synthesis unit 16 are the same as those in FIG. , t), a mask pattern m2 (f, t) obtained by inverting this is generated by the following equation (5).

  The masking processing unit 18 masks the time frequency component Gf · X2 (f, t) input from the gain applying unit 13 with the mask pattern m2 (f, t). Therefore, the masking processing unit 18 removes the time-frequency component Gf · X2 (f, t) from the component excluding the component in which the received signal x1 (t) output from the main microphone is dominant, that is, the interference sound. Only frequency components are output.

  The time frequency synthesizer 19 removes the component GfY2 (f, t) from the time frequency component GfX2 (f, t) excluding the component in which the received sound signal x1 (t) output from the main microphone is dominant. That is, only the time frequency component Gf · Y2 (f, t) is synthesized with the disturbing sound, and the output signal Gfy2 (t) is output. Even in this case, the output signal y2 (t) can be output by providing a gain removal circuit.

It is a block diagram showing an embodiment of a sound source separation device according to the present invention. It is explanatory drawing of the gain provision operation | movement in a gain provision part. It is a block diagram which shows other embodiment of the sound source separation apparatus which concerns on this invention. It is a block diagram which shows other embodiment of the sound source separation apparatus which concerns on this invention. It is a block diagram which shows the modification of the sound source separation apparatus which concerns on this invention.

11,12 ... Time frequency analysis unit, 13 ... Gain addition unit, 14 ... Level difference comparison unit, 15,19 ... Masking processing unit, 16,19 ... Time frequency synthesis unit, 17 ... Gain removal unit

Claims (5)

In the sound source separation method for separating and outputting at least one of the target sound component and the interference sound component from the received sound signal output from the main / sub microphone,
A first step of converting sound reception signals output from the main and sub microphones into time frequency components in the signal path of the main and sub microphones,
A second step of applying gain to the received sound signal before being converted to the time frequency component or the time frequency component after being converted to the time frequency component in at least one of the signal paths of the main and sub microphones;
A third step of generating a mask pattern by adding a gain in the second step and comparing the amplitude of the time frequency component after being converted in the first step for each time frequency component;
A fourth step of gaining at the second step and masking at least one of the time-frequency components after the conversion by the first step using the mask pattern generated by the third step. When,
A sound source separation method comprising a fifth step of synthesizing the time-frequency components output in the fourth step. In the sound source separation device that separates and outputs at least one of the target sound component and the interference sound component from the received sound signal output by the main / sub microphone
Conversion means provided in the signal path of the main and sub microphones, and converting the received sound signals output from the main and sub microphones into time frequency components, respectively;
A gain applying means which is provided in at least one of the signal paths of the main and sub microphones and applies a gain to the received sound signal before being converted to the time frequency component, or to the time frequency component after being converted to the time frequency component;
A level difference comparing means for generating a mask pattern by comparing the amplitude of the time frequency component after gain is given by the gain giving means and converted by the converting means for each time frequency component;
A masking processing means for masking at least one of the time frequency components after the gain is applied by the gain applying means and converted by the converting means, using the mask pattern generated by the level difference comparing means;
A sound source separation apparatus comprising: a time frequency synthesis unit that synthesizes a time frequency component output from the masking processing unit. The gain applied by the gain applying means causes an amplitude difference between the time frequency components of the received signal output from the main and sub microphones for the disturbing sound, and the main and sub microphones for the target sound. 3. The sound source separation device according to claim 2, wherein the sound source separation device is set to a constant value or a frequency-dependent value so that the magnitude relationship between the amplitudes of the time frequency components of the received sound signals to be output does not reverse. The gain applying means applies a gain of a frequency dependent value to a time frequency component in the signal path of the main microphone, and further from a time frequency component input to the masking processing means through the signal path of the main microphone or from the masking processing means. 3. A gain removing unit that removes the gain by applying a process reverse to the gain applying by the gain applying unit to the time frequency component of the signal path of the main microphone that is output. 3. The sound source separation device according to 3. A program for realizing a function of separating and outputting at least one of a target sound component and an interference sound component from a received sound signal output from a main / sub microphone,
A first function for converting received sound signals output from the main and sub microphones into time frequency components in the signal path of the main and sub microphones;
A second function for giving a gain to the received sound signal before being converted into the time frequency component or the time frequency component after being converted into the time frequency component in at least one of the signal paths of the main and sub microphones;
A third function for generating a mask pattern by adding a gain by the second function and comparing the amplitude of the time frequency component after being converted by the first function for each time frequency component;
A fourth function for masking at least one of the time frequency components after gain is given by the second function and converted by the first function, using the mask pattern generated by the third function When,
A program for executing a fifth function for synthesizing time-frequency components output by the fourth function.

Images