WO2014083999A1 - 信号処理装置、信号処理方法、および信号処理プログラム - Google Patents
信号処理装置、信号処理方法、および信号処理プログラム Download PDFInfo
- Publication number
- WO2014083999A1 WO2014083999A1 PCT/JP2013/079701 JP2013079701W WO2014083999A1 WO 2014083999 A1 WO2014083999 A1 WO 2014083999A1 JP 2013079701 W JP2013079701 W JP 2013079701W WO 2014083999 A1 WO2014083999 A1 WO 2014083999A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- change amount
- signal
- phase
- phase component
- unit
- Prior art date
Links
- 238000012545 processing Methods 0.000 title claims abstract description 61
- 238000003672 processing method Methods 0.000 title claims 2
- 238000006243 chemical reaction Methods 0.000 claims abstract description 73
- 230000008859 change Effects 0.000 claims description 114
- 238000000819 phase cycle Methods 0.000 claims 1
- 230000009466 transformation Effects 0.000 abstract description 19
- 238000001228 spectrum Methods 0.000 description 91
- 238000012937 correction Methods 0.000 description 56
- 238000004364 calculation method Methods 0.000 description 38
- 230000001629 suppression Effects 0.000 description 38
- 238000010586 diagram Methods 0.000 description 31
- 238000000034 method Methods 0.000 description 25
- 230000014509 gene expression Effects 0.000 description 24
- 230000006866 deterioration Effects 0.000 description 20
- 230000006870 function Effects 0.000 description 20
- 230000000694 effects Effects 0.000 description 17
- 230000015556 catabolic process Effects 0.000 description 12
- 238000006731 degradation reaction Methods 0.000 description 12
- 238000009826 distribution Methods 0.000 description 12
- 230000008569 process Effects 0.000 description 11
- 230000002829 reductive effect Effects 0.000 description 11
- 230000015572 biosynthetic process Effects 0.000 description 10
- 230000007423 decrease Effects 0.000 description 10
- 238000003786 synthesis reaction Methods 0.000 description 10
- 238000012935 Averaging Methods 0.000 description 9
- 230000001276 controlling effect Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 6
- 230000000670 limiting effect Effects 0.000 description 5
- 230000002411 adverse Effects 0.000 description 2
- 101150068393 argx gene Proteins 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 230000000875 corresponding effect Effects 0.000 description 2
- 230000001934 delay Effects 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/18—Phase-modulated carrier systems, i.e. using phase-shift keying
- H04L27/22—Demodulator circuits; Receiver circuits
- H04L27/227—Demodulator circuits; Receiver circuits using coherent demodulation
- H04L27/2275—Demodulator circuits; Receiver circuits using coherent demodulation wherein the carrier recovery circuit uses the received modulated signals
- H04L27/2278—Demodulator circuits; Receiver circuits using coherent demodulation wherein the carrier recovery circuit uses the received modulated signals using correlation techniques, e.g. for spread spectrum signals
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
- G10L21/0232—Processing in the frequency domain
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0264—Noise filtering characterised by the type of parameter measurement, e.g. correlation techniques, zero crossing techniques or predictive techniques
Definitions
- the present invention relates to a signal processing technique for controlling a phase component of a signal.
- Patent Document 1 and Non-Patent Document 1 disclose a noise suppression technique that focuses on a phase spectrum.
- the techniques described in Patent Document 1 and Non-Patent Document 1 suppress the amplitude spectrum related to noise and simultaneously shift the phase spectrum by a random value up to ⁇ / 4.
- the techniques described in Patent Document 1 and Non-Patent Document 1 realize noise suppression that cannot be suppressed only by attenuation of the noise spectrum by randomly shifting the phase spectrum.
- Patent Document 1 in order to randomly shift the phase spectrum, it is necessary to generate a random number. As a result, the amount of computation for generating random numbers increases.
- An object of the present invention is to provide a signal processing technique that solves the above-described problems.
- an apparatus provides: Conversion means for converting a mixed signal in which the first signal and the second signal are mixed into a phase component for each frequency and an amplitude component or a power component for each frequency; A change amount generating means for generating a change amount of the phase component of a predetermined frequency by using a data series having a weaker cross-correlation than the phase component and lower randomness than a random number; Phase control means for controlling the phase component using the change amount provided from the change amount generation means; Inverse conversion means for generating an enhancement signal using the phase component subjected to control processing by the phase control means; It is provided with.
- the method according to the present invention comprises: A conversion step of converting a mixed signal in which the first signal and the second signal are mixed into a phase component for each frequency and an amplitude component or a power component for each frequency; A change amount generation step for generating a change amount of the phase component of a predetermined frequency using a data series having a weaker cross-correlation than the phase component and lower randomness than a random number; A phase control step for controlling the phase component using the change amount generated in the change amount generation step; An inverse conversion step of generating an enhancement signal using the phase component to which control processing is applied in the phase control step; It is characterized by including.
- a program provides: A conversion step of converting a mixed signal in which the first signal and the second signal are mixed into a phase component for each frequency and an amplitude component or a power component for each frequency; A change amount generation step for generating a change amount of the phase component of a predetermined frequency using a data series having a weaker cross-correlation than the phase component and lower randomness than a random number; A phase control step for controlling the phase component using the change amount generated in the change amount generation step; An inverse conversion step of generating an enhancement signal using the phase component to which control processing is applied in the phase control step; Is executed by a computer.
- FIG. 1 is a block diagram showing a schematic configuration of a signal processing apparatus as a first embodiment of the present invention. It is a block diagram which shows schematic structure of the noise suppression apparatus as 2nd Embodiment of this invention. It is a block diagram which shows the structure of the conversion part contained in 2nd Embodiment of this invention. It is a block diagram which shows the structure of the inverse transformation part contained in 2nd Embodiment of this invention. It is a block diagram which shows schematic structure of the noise suppression apparatus as 3rd Embodiment of this invention. It is a block diagram which shows schematic structure of the variation
- FIG. 1 is a diagram showing a schematic configuration of a signal processing apparatus 100 according to the first embodiment of the present invention.
- the signal processing apparatus 100 includes a conversion unit 101, a phase control unit 102, a change amount generation unit 103, and an inverse conversion unit 104.
- the conversion unit 101 converts the mixed signal 110 in which the first signal and the second signal are mixed into a phase component 120 for each frequency and an amplitude component or power component 130 for each frequency.
- the change amount generation unit 103 generates a change amount of a phase component of a predetermined frequency using a data series having a cross-correlation weaker than that of the phase component 120 and lower randomness than a random number.
- the phase control unit 102 controls the phase component 120 using the change amount provided from the change amount generation unit 103.
- the inverse transform unit 104 generates the enhancement signal 170 using the phase component 140 that has been subjected to control processing by the phase control unit 102.
- the cross-correlation is weaker than the phase component 120 and the phase component 120 is controlled using a data sequence that is less random than the random number, and noise suppression that cannot be suppressed only by attenuation of the amplitude spectrum is efficiently performed. Can be realized.
- FIG. 2 is a block diagram showing the overall configuration of the noise suppression apparatus 200.
- the noise suppression apparatus 200 of this embodiment functions also as a part of apparatuses, such as a digital camera, a notebook personal computer, a mobile telephone, etc., this invention is not limited to this.
- the noise suppression apparatus 200 can be applied to any information processing apparatus that is required to remove noise from an input signal.
- the noise suppression apparatus according to the present embodiment appropriately removes the impact sound generated by the button operation, for example, in a form in which an operation such as button pressing is performed near the microphone. Briefly, by converting a signal including an impact sound into a frequency domain signal and controlling a phase component in the frequency space using a data series having a weak cross-correlation, the impact sound is appropriately removed.
- a deterioration signal (a signal in which a desired signal and noise are mixed) is supplied to the input terminal 206 as a sample value series.
- the conversion unit 201 performs transformation such as Fourier transformation on the supplied degradation signal and divides the degradation signal into a plurality of frequency components.
- the conversion unit 201 processes a plurality of frequency components independently at each frequency. Here, the description will be continued focusing on a specific frequency component.
- the conversion unit 201 supplies the degradation signal amplitude spectrum (amplitude component) 230 among the plurality of frequency components to the inverse conversion unit 204.
- the conversion unit 201 supplies the phase spectrum (phase component) 220 among the plurality of frequency components to the phase control unit 202 and the change amount generation unit 203.
- the conversion unit 201 supplies the deteriorated signal amplitude spectrum 230 to the inverse conversion unit 204, but the present invention is not limited to this.
- the conversion unit 201 may supply a power spectrum corresponding to the square of the degraded signal amplitude spectrum 230 to the inverse conversion unit 204.
- the change amount generation unit 203 generates a change amount using the deteriorated signal phase spectrum 220 received from the conversion unit 201 and supplies the change amount to the phase control unit 202.
- the “change amount” of the phase is a concept including the “rotation amount” and the “replacement amount” of the phase, and means a control amount of the phase.
- the phase control unit 202 reduces the correlation of the phase by changing the deterioration signal phase spectrum 220 supplied from the conversion unit 201 using the change amount supplied from the change amount generation unit 203, and obtains an enhanced signal phase spectrum 240. This is supplied to the inverse conversion unit 204.
- the inverse conversion unit 204 synthesizes the enhancement signal phase spectrum 240 supplied from the phase control unit 202 and the deteriorated signal amplitude spectrum 230 supplied from the conversion unit 201 to perform inverse conversion, and outputs an enhancement signal 270 as an output terminal. 207.
- FIG. 3 is a block diagram illustrating a configuration of the conversion unit 201.
- the converting unit 201 includes a frame dividing unit 301, a windowing processing unit (windowing unit) 302, and a Fourier transform unit 303.
- the deteriorated signal samples are supplied to the frame dividing unit 301 and divided into frames for every K / 2 samples.
- K is an even number.
- the deteriorated signal sample divided into frames is supplied to the windowing processing unit 302, and is multiplied by w (t) which is a window function.
- the windowing processing unit 302 may use a symmetric window function for a real signal.
- the windowing processing unit 302 may use, for example, a Hanning window represented by the following equation (3) as w (t).
- various window functions such as a Hamming window and a triangular window are known.
- the windowed output is supplied to the Fourier transform unit 303 and converted into a degraded signal spectrum Y n (k).
- the deteriorated signal spectrum Y n (k) is separated into a phase and an amplitude, and the deteriorated signal phase spectrum argY n (k) is transmitted to the phase control unit 202 and the change amount generating unit 203 by the deteriorated signal amplitude spectrum
- a power spectrum may be used instead of the amplitude spectrum.
- FIG. 4 is a block diagram showing the configuration of the inverse transform unit 204.
- the inverse transform unit 204 includes an inverse Fourier transform unit 401, a windowing processing unit 402, and a frame synthesis unit 403.
- the inverse Fourier transform unit 401 includes the degraded signal amplitude spectrum 230 (
- the inverse Fourier transform unit 401 performs inverse Fourier transform on the obtained enhancement signal.
- the output signal at -1 (the left side of equation (7)) is obtained.
- the obtained output signal is transmitted from the frame synthesis unit 403 to the output terminal 207.
- the transformation in the transformation unit 201 and the inverse transformation unit 204 has been described as a Fourier transformation.
- the transformation unit 201 and the inverse transformation unit 204 are replaced with a Hadamard transformation, a Haar transformation, and a wavelet transformation in place of the Fourier transformation.
- Other conversions may be used.
- the conversion unit 201 and the inverse conversion unit 204 use Haar transform, multiplication is not necessary, and the area of the LSI can be reduced.
- the transform unit 201 and the inverse transform unit 204 use wavelet transform, the time resolution can be changed depending on the frequency, so that an improvement in noise suppression effect can be expected.
- the change amount generation unit 203 may generate the change amount, and the phase control unit 202 may control the phase.
- high frequency quality is achieved by integrating more frequency components so that the band after integration is widened from the low frequency region where the discrimination characteristics of auditory characteristics are high to the high frequency region where the ability is low. Can do.
- the phase control is executed after integrating a plurality of frequency components, the number of frequency components to which the phase control is applied is reduced, and the entire calculation amount can be reduced.
- the change amount generation unit 203 receives the deterioration signal phase spectrum 220 from the conversion unit 201 and generates a change amount for reducing the phase correlation. Since the degradation signal phase spectrum 220 supplied from the conversion unit 201 is argY n (k) (0 ⁇ k ⁇ K), the change amount generation unit 203 uses the enhanced signal phase spectrum argX n (k) with reduced correlation. For example, it can be obtained by the following processing. This corresponds to inversion of the sign of every other phase. Of course, you may invert every arbitrary integer smaller than K instead of every other.
- the change amount generation unit 203 obtains the rotation amount ⁇ argY n (k) as the change amount necessary for the phase control of Expression (8) as the following expression. That is, the change amount generation unit 203 generates the rotation amount ⁇ argY n (k) shown in Equation (9) as the change amount. Also, It can also be. mod [k, K] represents the remainder when k is divided by K.
- the rotation amount ⁇ argY n (k) at this time corresponds to a value obtained by shifting the original phase by K / 2 samples. Obviously, the amount of deviation is not limited to K / 2, and may be any integer.
- the phase at a position symmetrical to the original phase with K / 2 as the center may be used as the rotation amount ⁇ argY n (k). In that case, the following equation is used.
- the amount of change can be generated by combining these two types of processing, that is, sign inversion and addition of the shifted phase. That is, Or It is.
- the shift amount K / 2 can be changed. For example, if the shift amount is the frame number n at that time, the shift amount automatically changes with time.
- the formula (10) may be combined instead of the formula (11).
- a constant multiple may be combined with the phase selective sign inversion and shift addition processing.
- a constant multiple when a constant multiple is combined with Expression (10), it can be expressed by the following Expression (14). This is an example of multiplying a term to be shift-added by a constant by the value of k corresponding to the position of the term.
- phase samples can be exchanged.
- any integer smaller than K may be used.
- ⁇ L (k) and ⁇ R (k) are non-correlated components (components having no correlation).
- the effect of removing the correlation is reduced, the reduction of the effect can be minimized by selectively using a j having a large value.
- the best example is to perform phase correlation removal based on the following equation using only the largest a j .
- jmax is a value of j at which the correlation coefficient a takes the maximum value.
- the amount of calculation required for the correlation removal can be reduced.
- LPC linear prediction coefficient
- the coefficient a j in the linear correlation equation is known as a linear prediction coefficient (LPC) in speech coding.
- LPC can be obtained at high speed using the Levinson-Durbin recursion method. Further, using the difference (error) between the original phase sample value and the prediction result, an LPC can be obtained by using a coefficient update algorithm of an adaptive filter represented by a normalized LMS algorithm or the like.
- correlation removal may be performed assuming a linear combination of K j -1 samples (K j ⁇ K) instead of a linear combination of adjacent K-1 samples.
- the nonlinear function can be generally approximated by a polynomial.
- the non-linear function f NL [ ⁇ ] When a polynomial approximation of argY n (j), limiting the type of argY n (j), further it is also possible to limit the degree. For example, if we use only argY n (k), argY n (k + 1) and their squares, f NL [•] is argY n (k), argY n (k + 1) and their It is approximated by only four kinds of terms including the square. Such approximation of the nonlinear function can reduce the amount of calculation required for correlation removal.
- phase control unit 202 adds the change amount ⁇ argY n (k) supplied from the change amount generation unit 203 to the deteriorated signal phase spectrum 220 supplied from the conversion unit 201, so that the enhanced signal phase spectrum 240argX n (k ) Is obtained and supplied to the inverse transform unit 204. That is, the following equation is executed.
- the phase control unit 202 replaces the change amount ⁇ argY n (k) supplied from the change amount generation unit 203 with the deterioration signal phase spectrum 220 without adding it to the deterioration signal phase spectrum 220 supplied from the conversion unit 201.
- the enhanced signal phase spectrum 240argX n (k) can be obtained and supplied to the inverse transform unit 204. That is, by executing the following equation, the amount of phase rotation and the amount of phase replacement are equivalent.
- the replacement is realized by subtracting the emphasized signal phase spectrum itself and adding the rotation amount, but the replacement may be realized by simply replacing the phase data with the replacement amount. .
- the phase control unit 202 changes the value of ⁇ argY n (k), the shape of the degraded signal phase spectrum 220 Change. Due to this shape change, the correlation of the degraded signal phase spectrum 220 becomes weak, and the characteristics of the input signal can be weakened.
- phase expansion can also be applied prior to the phase processing described so far. This is because the degradation signal phase spectrum 220 has a range of ⁇ ⁇ . That is, phase expansion is performed so that the value range is not limited to ⁇ ⁇ .
- the correlation can be obtained with high accuracy when obtaining the correlation represented by the equations (15), (16), (20) and the like.
- B. Rad and T. Virtanen "Phase spectrum prediction of audio signals," Proc. ISCCSP2012, CD-ROM, May 2012.
- FIG. 5 is a block diagram showing the overall configuration of the noise suppression apparatus 500.
- the noise suppression device 500 of the present embodiment has the same configuration except for the change amount generation unit 503. Therefore, only the change amount generation unit 503 will be described, and detailed description regarding other components will be omitted.
- FIG. 6 is a block diagram illustrating configurations of the phase control unit 202 and the change amount generation unit 503.
- the change amount generation unit 503 includes an amplitude holding unit 601 and an amplitude analysis unit 602.
- the amplitude holding unit 601 holds the degradation signal amplitude spectrum 230 and supplies it to the amplitude analysis unit 602.
- the phase control unit 202 is supplied with the deterioration signal phase spectrum 220 from the conversion unit 201, and is supplied with the phase rotation amount from the change amount generation unit 503.
- the phase control unit 202 rotates (shifts) the deterioration signal phase spectrum 220 by the rotation amount supplied from the change amount generation unit 503, and supplies it to the inverse conversion unit 204 as the enhanced signal phase spectrum 240.
- the amplitude analysis unit 602 sets the product obtained by multiplying the degraded signal amplitude spectrum held by the amplitude holding unit 601 by ⁇ as the rotation amount.
- the same effect can be obtained by collecting the deteriorated signal amplitude spectrum held by the amplitude holding unit 601 in the frequency direction or the time axis direction and directly using it as the rotation amount.
- the phase control unit 202 changes (rotates or replaces) the deterioration signal phase spectrum at each frequency using the change amount generated by the change amount generation unit 503 based on the deterioration signal amplitude spectrum.
- the shape of the deteriorated signal phase spectrum 220 is changed by the control performed by the phase control unit 202. This shape change can weaken the characteristics of noise.
- the amplitude analysis unit 602 may also supply the phase control unit 202 with a rotation amount obtained by normalizing the deteriorated signal amplitude spectrum 230 held by the amplitude holding unit 601. In this case, the amplitude analysis unit 602 first obtains an average of the deterioration signal amplitude spectrum 230 (all K positive values). The product obtained by multiplying the quotient obtained by dividing the deteriorated signal amplitude spectrum 230 by the obtained average value is taken as the rotation amount. In this case, a similar effect can be obtained even if the quotient is directly used as the rotation amount without multiplying by ⁇ .
- the dispersion can be relatively increased compared to the case without normalization, the effect of removing the correlation with respect to the rotated phase can be enhanced. Further, when obtaining the average, it is also possible to obtain the average after first excluding values (outliers) that are extremely different from others. The adverse effect of outliers can be eliminated, and a more effective amount of rotation can be obtained.
- the amplitude analysis unit 602 can also normalize the distribution of the deteriorated signal amplitude spectrum 230 to obtain the rotation amount.
- the amplitude analyzer 602 obtains the maximum value
- the minimum value is subtracted from the deteriorated signal amplitude spectrum and divided by the difference between the maximum value and the minimum value.
- the product obtained by multiplying this quotient by ⁇ is the rotation amount. That is, the rotation amount ⁇ argY n (k) is obtained by the following equation.
- the rotation amount is distributed between 0 and ⁇ , so that the effect of removing the correlation with respect to the rotated phase can be enhanced. In this case, a similar effect can be obtained even if the quotient is directly used as the rotation amount without multiplying by ⁇ .
- the rotation amount generation unit 502 can also normalize the distribution of the deteriorated signal amplitude spectrum with its own envelope to obtain the rotation amount. For example, a regression curve of a degraded signal amplitude spectrum is obtained from N samples, and each sample is divided by the value of the regression curve. A regression curve may be obtained by using a part of the N samples, or a regression curve may be obtained after outliers are excluded. By excluding outliers, adverse effects of outliers can be excluded, and a more effective rotation amount can be obtained. The quotient thus obtained has a distribution centered at 1.
- the rotation amount ⁇ argY n (k) can be obtained by the following equation (27).
- is a degraded signal amplitude spectrum normalized by an envelope.
- the noise suppression apparatus 700 according to the present embodiment is different from the second embodiment in that an amplitude control unit 708 compensates for a decrease in output level due to phase control by the phase control unit 202. Since other configurations and operations are the same as those of the second embodiment, description thereof is omitted here.
- the amplitude control unit 708 includes a correction amount calculation unit 881 and an amplitude correction unit 882.
- the correction amount calculation unit 881 calculates the amplitude correction coefficient using the phase rotation amount transmitted from the change amount generation unit 203.
- the amplitude correction unit 882 multiplies the calculated amplitude correction coefficient by the deteriorated signal amplitude spectrum supplied from the conversion unit 201 and supplies the result to the inverse conversion unit 204. By multiplying by the amplitude correction coefficient, it is possible to eliminate the output level decrease when the deteriorated signal phase spectrum 220 is controlled to obtain the enhanced signal phase spectrum 240.
- FIG. 9 and 10 show signals when the deteriorated signal is processed by the block diagram shown in FIG. The difference between FIG. 9 and FIG. 10 is the presence or absence of phase rotation.
- FIG. 9 shows a signal when the phase rotation is not performed
- FIG. 10 shows a signal when the phase rotation is performed from the frame 3.
- FIG. 9 Depicted at the top of FIG. 9 is a degraded signal.
- the deteriorated signal is divided into frames by the frame dividing unit 301.
- the second signal from the top divided by the dotted line is a signal after frame division.
- signals for four consecutive frames are shown.
- the overlap rate of frames is 50%.
- the windowing processing unit 302 performs windowing on the signal divided into frames.
- the third signal from the top divided by the dotted line is the signal after the windowing process.
- weighting by a rectangular window is performed in order to clearly show the influence of the phase rotation.
- the signal is converted into a frequency domain signal by the Fourier transform unit 303, but the signal in the frequency domain is omitted in FIG.
- the signal converted into the time domain by the inverse Fourier transform unit 401 of the inverse transform unit 204 is illustrated.
- the fourth signal from the top divided by the dotted line is the signal after phase rotation. However, since phase rotation is not performed in FIG. 9, there is no change from the signal after the windowing process.
- the enhancement signal output from the inverse Fourier transform unit 401 of the inverse transform unit 204 is subjected to windowing processing again by the windowing processing unit 402.
- FIG. 9 shows a case where weighting by a rectangular window is performed.
- the signal subjected to the windowing process is synthesized by the frame synthesis unit 403. At this time, it is necessary to align the time between frames. Since the overlap rate of frames is 50%, the frames overlap by exactly half. When phase rotation is not performed, the input signal and the output signal match as shown in FIG.
- FIG. 10 shows a signal when the phase rotation is performed from the frame 3. Depicted at the top is the same degraded signal as in FIG. The signals after the frame division and the windowing process are the same as in FIG.
- FIG. 10 illustrates a case where a constant phase rotation is performed from the frame 3. Pay attention to the right-triangulated section shown below the dotted line in the phase rotation process. Due to the phase rotation process, the signals of frames 3 and 4 are shifted in the time direction. The signal subjected to the phase rotation is subjected to windowing processing again, and frame synthesis is performed. At this time, a difference occurs in the signals of the frames 2 and 3 in the section ii where the frames 2 and 3 overlap. As a result, the output signal level after frame synthesis decreases in the interval ii. That is, when phase rotation is performed, the output signal level decreases in the section ii in FIG.
- the decrease in the output signal level due to this phase rotation can be explained by frequency domain vector synthesis by replacing the time domain addition with the frequency domain addition.
- FIG. 11 shows the deteriorated signals of two consecutive frames after frame division and windowing processing as x 1 [n] and x 2 [m].
- the overlap rate is 50%.
- n represents a discrete time x 1.
- m represents a discrete time x 2.
- the overlap rate is 50%, the following equation (28) is established. Further, the relationship between x 1 and x 2 is expressed by the following equation (29).
- the frequency domain signal X [k] is expressed as the following Expression (30) by Fourier transform of the time domain signal x [n].
- k represents a discrete frequency
- L is a frame length.
- interval m L / 2 to L-1.
- Equation (36) Substituting Equation (34) and Equation (35) for addition in the time domain, it is expressed as Equation (36) below. Further, when Expressions (32) and (33) are substituted into the frequency domain signals X 1 [k] and X 2 [k] in Expression (36), the following Expression (37) is obtained. Further, when the expression (37) is expanded, it is expressed as the following expression (38).
- the correction amount calculation unit 881 determines the amplitude correction amount of the emphasized signal amplitude spectrum so as to correct the decrease amount of the output signal level.
- each frequency component is normalized to a unit vector by paying attention to the magnitude variation due to phase rotation.
- the occurrence probability of ⁇ is determined by a normal distribution. Therefore, in order to obtain the expected power value when phase rotation is performed using normal random numbers, weighting based on the occurrence probability of ⁇ needs to be performed.
- a weight function f ( ⁇ ) based on the occurrence probability of ⁇ is introduced. Cos ( ⁇ ) is weighted by the weight function f ( ⁇ ). Further, the expected power value can be obtained by normalizing with the integral value of the weighting function f ( ⁇ ).
- Equation (53) The output power expectation value E (S ′ 2 ) when phase rotation is performed with a normal random number is obtained by substituting the weighting function f ( ⁇ ) and its integral value into Equation (53), which is the output power expectation value of a uniform normal random number. By introducing, it is expressed as the following formula (54).
- the correction amount calculation unit 881 transmits the correction coefficient as sqrt (1 / 0.805) to the amplitude correction unit 882 when the phase is rotated with a normal random number of the standard normal distribution.
- the phase control unit 202 may perform phase rotation on all frequencies or a part of frequencies.
- the amplitude control unit 708 performs amplitude correction only on the frequency on which phase rotation has been performed. Therefore, the correction coefficient for the frequency at which phase rotation is not performed is 1.0. Only the correction coefficient for the frequency at which the phase is rotated is derived.
- the noise suppression apparatus 700 can remove the influence on the output signal level by controlling the phase spectrum by the amplitude control unit 708. Therefore, the noise suppression apparatus 700 can obtain a high-quality enhanced signal.
- a noise suppression apparatus 1500 according to the fifth embodiment of the present invention will be described with reference to FIG.
- This embodiment is different from the configuration of the third embodiment in that an amplitude control unit 708 is provided. Since the configuration other than the amplitude control unit 708 is the same as that of the third embodiment, and the amplitude control unit 708 is the same as that of the fourth embodiment, the same components are denoted by the same reference numerals and detailed description thereof is omitted. To do.
- the noise derived from the phase can be efficiently suppressed by rotating or replacing the deteriorated signal phase spectrum by using the deteriorated signal amplitude spectrum or a value obtained therefrom, and the phase is controlled by the amplitude control. It is possible to suppress a decrease in output level accompanying control.
- a noise suppression device 1600 according to a sixth embodiment of the present invention will be described with reference to FIG. This embodiment differs from the second embodiment in that the upper limit of the phase rotation amount is limited. Since other configurations and operations are the same as those in the second embodiment, a detailed description thereof is omitted here.
- FIG. 16 is a block diagram showing a configuration of a noise suppression apparatus 1600 according to the present embodiment.
- the noise suppression device 1600 according to the present embodiment includes a change amount limiting unit 1601 in addition to the change amount generation unit 203 and the phase control unit 202 described in the second embodiment.
- the change amount generation unit 203 generates a change amount of the deteriorated signal phase spectrum while being restricted by the change amount restriction unit 1601, and supplies the change amount to the phase control unit 202.
- the change amount restriction unit 1601 restricts the rotation amount generated by the change amount generation unit 203 within a certain range. That is, the change amount limiting unit 1601 limits the ⁇ distribution to an arbitrary range from 0 to 2 ⁇ . For example, the change amount limiting unit 1601 limits the distribution of ⁇ to 0 to ⁇ / 2. As a result, the characteristic of the deteriorated signal phase spectrum remains in the enhanced signal phase spectrum to some extent. Compared with the case where the phase is completely randomly rotated, the characteristics of the deteriorated signal are retained to some extent, and therefore the influence on the target sound is reduced. Therefore, distortion of the target sound is reduced.
- the present invention according to this embodiment can reduce the deterioration of the target sound by limiting the amount of phase rotation.
- a seventh embodiment of the present invention will be described with reference to FIG.
- the present invention according to the present embodiment differs from the fourth embodiment in that the phase component is delayed, the difference between the phase components between frames is obtained, and the correction amount is calculated therefrom. That is, the internal configuration of the amplitude control unit 1708 is different from that of the second embodiment. Since other configurations and operations are the same as those of the second embodiment, description thereof is omitted here.
- FIG. 17 is a block diagram illustrating a configuration of the amplitude control unit 1708 in the present embodiment.
- the phase control unit 202 in this embodiment supplies the phase after rotation to the amplitude control unit 1708.
- the amplitude control unit 1708 includes a phase component delay unit 1781, a correction amount calculation unit 1782, and an amplitude correction unit 882.
- the phase component delay unit 1781 holds the emphasized signal phase spectrum supplied from the phase control unit 202 for one frame, and supplies it to the correction amount calculation unit 1782.
- the correction amount calculation unit 1782 calculates the amplitude correction amount from the enhancement signal phase spectrum one frame before from the phase component delay unit 1781 and the current enhancement signal phase spectrum from the phase control unit 202, and transmits the amplitude correction amount to the amplitude correction unit 882.
- the output level can be corrected even when the expected value of the output level cannot be derived mathematically from the phase change amount.
- the correction amount calculation unit 1782 obtains the magnitude of the combined vector at each frequency from the emphasized signal phase spectra of the previous frame and the current frame, and determines the correction coefficient from the magnitude. Assuming that the phase of the previous frame is ⁇ and the phase of the current frame is ⁇ , the magnitude of the combined vector
- this value is supplied to the amplitude control unit 1708, and the enhancement signal amplitude spectrum is corrected, so that the decrease in the output level can be eliminated.
- the configuration and operation other than the phase rotation unit are the same as those in the second embodiment, the description thereof is omitted here.
- FIG. 18 is a block diagram showing the configuration of the phase control unit 202 and the amplitude control unit 1808 according to this embodiment.
- the present invention according to this embodiment is different from FIG. 8 (fourth embodiment) in that an input / output ratio calculation unit 1881 is included.
- the input / output ratio calculation unit 1881 receives the deterioration signal from the input terminal 206 and the enhancement signal from the inverse conversion unit 204, and calculates the input / output level ratio.
- the input / output ratio calculation unit 1881 supplies the input / output level ratio to the correction amount calculation unit 1882.
- the correction amount calculation unit 1882 calculates the correction amount so that the level of the enhancement signal is equal to the deterioration signal.
- the amplitude correction unit 882 corrects the emphasized signal amplitude spectrum with the calculated correction amount.
- the input / output ratio calculation unit 1881 obtains the level ratio from the time domain signal of the deterioration signal and the enhancement signal.
- the level ratio R between the degradation signal y n (t) of the nth frame and the enhancement signal x n (t) of the nth frame is expressed by the following equation (62).
- t indicates the sample time.
- L indicates the frame length of the Fourier transform.
- the correction amount calculation unit 1882 obtains the amplitude correction amount G based on the ratio value R and the number of frequency components subjected to phase rotation.
- the amplitude correction amount G is obtained as in the following equation (63).
- the amplitude correction unit 882 performs amplitude correction only with the frequency at which the phase rotation is performed based on the information on the presence / absence of the phase rotation transmitted from the change amount generation unit 203.
- the configuration and operation other than the input / output ratio calculation unit 1881 and the correction amount calculation unit 1882 are the same as those in the fourth embodiment, description thereof is omitted here.
- the present invention obtains a correction coefficient from a signal in the time domain, the output level can be corrected regardless of how the phase rotation amount is determined.
- FIG. 19 is a block diagram showing the configuration of the amplitude control unit 1908 according to this embodiment.
- the amplitude control unit 1908 in the present embodiment includes an averaging processing unit 1981 in addition to the input / output ratio calculation unit 1881 included in the eighth embodiment. Since the configuration and operation other than the averaging processing unit 1981 are the same as those in the eighth embodiment, description thereof is omitted here.
- the averaging processing unit 1981 receives the deterioration signal from the input terminal 206, performs the averaging process, and supplies the average value to the input / output ratio calculation unit 1881.
- the averaging processing unit 1981 receives the enhancement signal from the inverse transformation unit 204, performs an averaging process, and supplies the average value to the input / output ratio calculation unit 1881.
- the input / output ratio calculation unit 1881 receives the average value of the deterioration signal and the enhancement signal from the averaging processing unit 1981, and calculates the level ratio thereof.
- the averaging processing unit 1981 averages the levels of the deterioration signal and the enhancement signal with an arbitrary time length. Specifically, the averaging processing unit 1981 averages the levels of the deterioration signal and the enhancement signal using a moving average, leakage integration, or the like.
- the present invention according to the present embodiment uses an averaged level in addition to the effects of the eighth embodiment, so that fluctuations in the correction amount are suppressed and the quality of the output signal can be improved.
- FIG. 20 is a diagram illustrating a configuration of the noise suppression device 2000 according to the present embodiment.
- the noise suppression apparatus 2000 according to the present embodiment includes an amplitude component delay unit 2011, a phase component delay unit 2012, and an inverse conversion unit 2013 in addition to the configuration of FIG. 2 of the second embodiment.
- operations other than the amplitude component delay unit 2011, the phase component delay unit 2012, and the amplitude control unit 2008 are the same as those in the second embodiment, and thus the description thereof is omitted here.
- the deterioration signal 210 supplied to the input terminal 206 is supplied to the conversion unit 201 and the amplitude control unit 2008.
- the conversion unit 201 supplies the degraded signal amplitude spectrum 230 to the amplitude component delay unit 2011 and the inverse conversion unit 2013. Further, the conversion unit 201 supplies the degradation signal phase spectrum 220 to the phase control unit 202 and the change amount generation unit 203.
- the phase control unit 202 controls the degradation signal phase spectrum 220 supplied from the conversion unit 201 using the change amount generated by the change amount generation unit 203, and uses the inverse conversion unit 2013 and the phase component delay as an enhanced signal phase spectrum. To the unit 2012. Further, the change amount generation unit 203 transmits the presence / absence of phase rotation at each frequency to the amplitude control unit 2008.
- the inverse conversion unit 2013 uses the deteriorated signal amplitude spectrum 230 supplied from the conversion unit 201 and the deteriorated signal phase spectrum supplied from the phase control unit 202, to the amplitude control unit 2008 to output a signal whose level has been reduced due to phase rotation. introduce.
- the amplitude component delay unit 2011 delays the degradation signal amplitude spectrum 230 from the conversion unit 201 and supplies the delayed signal amplitude spectrum 230 to the amplitude control unit 2008.
- the phase component delay unit 2012 delays the emphasized signal phase spectrum from the phase control unit 202 and supplies it to the inverse transform unit 204.
- the amplitude control unit 2008 generates a corrected amplitude spectrum 250 from the degraded signal amplitude spectrum supplied from the amplitude component delay unit 2011 using the output of the inverse transform unit 2013 and the degraded signal 210.
- the inverse conversion unit 204 synthesizes the enhancement signal phase spectrum 240 supplied from the phase control unit 202 via the phase component delay unit 2012 and the corrected amplitude spectrum 250 supplied from the amplitude control unit 2008 and performs inverse conversion. And supplied to the output terminal 207 as an emphasis signal.
- the deteriorated signal phase spectrum 220 is controlled by the phase control unit 202 and converted into a time domain signal by the inverse conversion unit 2013.
- the amplitude control unit 2008 uses the signal and the degradation signal 210 to obtain a level fluctuation amount due to phase rotation.
- This amount of variation is a variation of only the rotation processing by the phase control unit 202. Therefore, the amplitude control unit 2008 can accurately grasp the level fluctuation due to the rotation of the phase.
- the amplitude control unit 2008 performs amplitude correction using this level ratio, and the obtained level ratio is one frame before. Therefore, an amplitude component delay unit 2011 and a phase component delay unit 2012 are introduced, and the amplitude control unit 2008 performs amplitude correction on the frequency component one frame before.
- FIG. 21 is a block diagram for explaining the internal configuration of the phase control unit 202 and the amplitude control unit 2008 according to the present embodiment.
- the input / output ratio calculation unit 2181 calculates a level ratio from a deterioration signal supplied from the input terminal 206 and a signal supplied from the inverse conversion unit 2013 and includes a level decrease due to phase rotation, and a correction amount calculation unit 2182. To supply.
- the correction amount calculation unit 2182 receives information on the presence / absence of phase rotation at each frequency from the change amount generation unit 203, and calculates an amplitude correction amount. Based on the amplitude correction amount, the amplitude correction unit 882 corrects the enhanced signal amplitude spectrum at each frequency and supplies the corrected signal amplitude spectrum to the inverse conversion unit 204.
- the noise suppression device 2000 can avoid the delay of the input / output ratio that was unavoidable in the eighth and ninth embodiments, and can correct the output level more accurately. realizable.
- the noise suppression device 2200 includes a frame overlap control unit 2208 in addition to the configuration of the fourth embodiment.
- Frame overlap control section 2208 controls the overlap rate when frames are divided and combined in conversion section 201 and inverse conversion section 204.
- the frame overlap control unit 2208 supplies the overlap ratio to the amplitude control unit 708.
- the level drop due to phase rotation is caused by overlap. This level reduction amount varies depending on the overlap rate, and the decrease amount increases as the overlap rate increases. Therefore, when the overlap rate changes, it is necessary to control the amplitude correction amount. Specifically, the correction amount is obtained based on the amplitude correction amount G when the overlap ratio is 50%.
- the amplitude correction amount is G when the overlap ratio is 50%. Therefore, using the ratio of the frame length L and the overlap length Q, the following equation (64) is obtained.
- the amplitude control unit 708 corrects the enhancement signal amplitude spectrum by correcting the correction coefficient transmitted from the phase control unit 202 based on the equation (64).
- the description thereof is omitted here.
- the noise suppression device 2200 according to the present embodiment can freely set the frame overlap rate.
- the present invention may be applied to a system composed of a plurality of devices, or may be applied to a single device. Furthermore, the present invention is also applicable to a case where a software signal processing program that implements the functions of the embodiments is supplied directly or remotely to a system or apparatus. Therefore, in order to realize the functions of the present invention on a computer, a program installed in the computer, a medium storing the program, and a WWW server that downloads the program are also included in the scope of the present invention.
- FIG. 23 is a configuration diagram of a computer 2300 that executes a signal processing program when the first embodiment is configured by a signal processing program.
- the computer 2300 includes an input unit 2301, a CPU 2302, an output unit 2303, and a memory 2304.
- the CPU 2302 controls the operation of the computer 2300 by reading the signal processing program. That is, the CPU 2302 executes the signal processing program stored in the memory 2304, and converts the mixed signal in which the first signal and the second signal are mixed into a phase component and an amplitude component or a power component for each frequency (S2311). . Next, the CPU 2302 generates a change amount of the phase component of the predetermined frequency using a data series having a cross-correlation weaker than the phase component and lower randomness than the random number (S2312). In accordance with the generated change amount, the CPU 2302 controls the phase component (S2313). The CPU 2302 generates an enhancement signal using the phase component that has been subjected to the control processing in step S2313 (S2314).
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Computer Networks & Wireless Communication (AREA)
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
Abstract
Description
第1信号と第2信号とが混在した混在信号を、周波数ごとの位相成分および周波数ごとの振幅成分またはパワー成分に変換する変換手段と、
前記位相成分よりも相互相関が弱く、乱数よりもランダム性の低いデータ系列を用いて、所定周波数の前記位相成分の変化量を生成する変化量生成手段と、
前記変化量生成手段から提供された変化量を用いて、前記位相成分を制御する位相制御手段と、
前記位相制御手段によって制御処理を加えられた位相成分を用いて強調信号を生成する逆変換手段と、
を備えたことを特徴とする。
第1信号と第2信号とが混在した混在信号を、周波数ごとの位相成分および周波数ごとの振幅成分またはパワー成分に変換する変換ステップと、
前記位相成分よりも相互相関が弱く、乱数よりもランダム性の低いデータ系列を用いて、所定周波数の前記位相成分の変化量を生成する変化量生成ステップと、
前記変化量生成ステップで生成された変化量を用いて、前記位相成分を制御する位相制御ステップと、
前記位相制御ステップにおいて制御処理を加えられた位相成分を用いて強調信号を生成する逆変換ステップと、
を含むことを特徴とする。
第1信号と第2信号とが混在した混在信号を、周波数ごとの位相成分および周波数ごとの振幅成分またはパワー成分に変換する変換ステップと、
前記位相成分よりも相互相関が弱く、乱数よりもランダム性の低いデータ系列を用いて、所定周波数の前記位相成分の変化量を生成する変化量生成ステップと、
前記変化量生成ステップで生成された変化量を用いて、前記位相成分を制御する位相制御ステップと、
前記位相制御ステップにおいて制御処理を加えられた位相成分を用いて強調信号を生成する逆変換ステップと、
をコンピュータに実行させることを特徴とする。
図1は、本発明の第1実施形態に係る信号処理装置100の概略構成を示す図である。図1において、信号処理装置100は、変換部101と位相制御部102と変化量生成部103と逆変換部104とを備える。
《全体構成》
本発明の第2実施形態としての雑音抑圧装置200について図2乃至図4を用いて説明する。図2は、雑音抑圧装置200の全体構成を示すブロック図である。本実施形態の雑音抑圧装置200は、例えばデジタルカメラ、ノートパソコン、携帯電話などといった装置の一部としても機能するが、本発明はこれに限定されるものではない。雑音抑圧装置200は、入力信号からのノイズ除去を要求されるあらゆる情報処理装置に適用可能である。本実施形態としての雑音抑圧装置は、例えば、マイクの近くでボタン押下などの操作がなされるような形態において、かかるボタン操作により発生する衝撃音を適切に除去する。簡単に説明すると、衝撃音を含む信号を周波数領域信号に変換し、周波数空間における位相成分を、相互相関が弱いデータ系列を用いて制御することにより、衝撃音を適切に除去する。
図3は、変換部201の構成を示すブロック図である。図3に示すように、変換部201はフレーム分割部301、窓掛け処理部(windowing unit)302、およびフーリエ変換部303を含む。劣化信号サンプルは、フレーム分割部301に供給され、K/2サンプル毎のフレームに分割される。ここで、Kは偶数とする。フレームに分割された劣化信号サンプルは、窓掛け処理部302に供給され、窓関数(window function)であるw(t)との乗算が行なわれる。第nフレームの入力信号yn(t)(t=0,1,...,K/2-1)に対するw(t)で窓掛け(windowing)された信号は、次式(1)で与えられる。
窓掛け処理部302は、実数信号に対しては、左右対称窓関数を用いてもよい。また、窓関数は、位相制御部202が何も制御をしないときに、変換部201の入力信号と逆変換部204の出力信号が計算誤差を除いて一致するように設計される。これは、w(t)+w(t+K/2)=1となることを意味する。
このほかにも、ハミング窓、三角窓など、様々な窓関数が知られている。窓掛けされた出力はフーリエ変換部303に供給され、劣化信号スペクトルYn(k)に変換される。劣化信号スペクトルYn(k)は位相と振幅に分離され、劣化信号位相スペクトルargYn(k)は、位相制御部202と変化量生成部203に、劣化信号振幅スペクトル|Yn(k)|は、逆変換部204に供給される。既に説明したように、振幅スペクトルの代わりにパワースペクトルを利用してもよい。
図4は、逆変換部204の構成を示すブロック図である。図4に示すように、逆変換部204は逆フーリエ変換部401、窓掛け処理部402およびフレーム合成部403を含む。逆フーリエ変換部401は、変換部201から供給された劣化信号振幅スペクトル230(|Xn(k)|)と位相制御部202から供給された強調信号位相スペクトル240(argXn(k))とを乗算して、強調信号(以下の式(4)の左辺)を求める。
変化量生成部203は、変換部201から劣化信号位相スペクトル220の供給を受け、位相の相関を減じるための変化量を生成する。変換部201から供給される劣化信号位相スペクトル220はargYn(k)(0≦k<K)であるので、変化量生成部203は、相関低減された強調信号位相スペクトルargXn(k)を、例えば、次のような処理で求めることができる。
これは、位相を1つおきに符号反転することに相当する。当然、1つおきではなく、Kより小さい任意の整数ごとに反転してもよい。
すなわち、変化量生成部203は、変化量として数式(9)に示す回転量ΔargYn(k)を生成する。また、
とすることもできる。mod[k,K]はkをKで割ったときの余りを表す。このときの回転量ΔargYn(k)は、もとの位相をK/2サンプルずらせたものに相当する。ずれ量はK/2に限らず、任意の整数でよいことは明らかである。
これは、シフト加算する項をその項の位置に対応したkの値で定数倍する例である。
となる。ここに、sgn(・)は符号を取り出す演算子である。右辺の分数は、位相が正の値をとるときだけ1、それ以外はゼロとなるので、argYn(k)の値に応じて選択的に前記処理を適用することができる。これらの変化量を用いた相関除去処理は、相関除去の程度と必要な演算量が異なる。実際に適用する際には、相関除去の程度と必要演算量を考慮して、適切なものを選択し、あるいは組み合わせて使用する。
あるいは、逆方向の相関に着目して、
とすることもできる。ここにδL(k)、δR(k)は、非相関成分(相関のない成分)である。
ここに、jmaxは相関係数aが最大値をとるjの値である。Nサンプルを用いた相関除去と比較して、相関除去に必要な演算量を削減することができる。
ここに、fNL[・]は非線形関数、δ(k)は非相関成分である。このとき、相関除去に用いる変化量は、
で求めることができる。非線形関数を用いた相関除去により、非線形な相関を有する場合に十分に相関を除去することができる。
位相制御部202は、変化量生成部203から供給された変化量ΔargYn(k)を、変換部201から供給された劣化信号位相スペクトル220に加算することにより、強調信号位相スペクトル240argXn(k)を求めて、逆変換部204に供給する。すなわち、次式を実行する。
《全体構成》
本発明の第3実施形態としての雑音抑圧装置500について図5を用いて説明する。図5は、雑音抑圧装置500の全体構成を示すブロック図である。本実施形態の雑音抑圧装置500は、第2実施形態の雑音抑圧装置200と比較して、変化量生成部503以外の構成は、同一である。そこで、変化量生成部503に関してだけ説明し、その他の構成要素に関する詳細な説明は省略する。
図6は位相制御部202および変化量生成部503の構成を示すブロック図である。図6に示すように、変化量生成部503は振幅保持部601と振幅分析部602とを含む。振幅保持部601は、劣化信号振幅スペクトル230を保持し、振幅分析部602に供給する。
振幅分析部602は、例えば、振幅保持部601が保持した劣化信号振幅スペクトルにπを乗じて得られた積を回転量とする。あるいは、振幅保持部601が保持した劣化信号振幅スペクトルを周波数方向または時間軸方向に集めてそのまま回転量としても、同様の効果が得られる。位相制御部202は、変化量生成部503が劣化信号振幅スペクトルにより生成した変化量を用いて、劣化信号位相スペクトルを各周波数で変化(回転または置換)させる。位相制御部202が行なう制御により、劣化信号位相スペクトル220の形状が変化する。この形状の変化により、雑音の特徴を弱めることができる。
振幅分析部602は、また、振幅保持部601が保持した劣化信号振幅スペクトル230を正規化したものを回転量として位相制御部202に供給してもよい。この場合、振幅分析部602は、まず、劣化信号振幅スペクトル230(全て正の値K個)の平均を求める。求めた平均値で劣化信号振幅スペクトル230を除して得られた商にπを乗じて得られた積を回転量とする。なお、このときにπを乗じずに商をそのまま回転量としても、類似の効果が得られる。正規化なしの場合と比べて分散を相対的に大きくすることができるので、回転させられる位相に対する相関除去の効果を強化することができる。また、平均を求める際に、最初に極端に他と異なる値(外れ値)を除外してから、平均を求めることもできる。外れ値の悪影響を排除することができ、より効果的な回転量を求めることができる。
振幅分析部602は、また、劣化信号振幅スペクトル230の分布を正規化してから回転量とすることもできる。まず、振幅分析部602は劣化信号振幅スペクトル230(全て正の値K個)の最大値|Xn(K)|maxと最小値|Xn(K)|minとを求める。次に、劣化信号振幅スペクトルから最小値を減算して、最大値と最小値の差で除する。この商にπを乗じて得られた積を回転量とする。すなわち、回転量ΔargYn(k)を次式で求める。
このように回転量を求めることによって、回転量は0とπの間に分布するので、回転させられる位相に対する相関除去の効果を強化することができる。なお、このときにπを乗じずに商をそのまま回転量としても、類似の効果が得られる。
回転量生成部502は、また、劣化信号振幅スペクトルの分布を自身の包絡線で正規化して、回転量とすることもできる。包絡線は、例えば、劣化信号振幅スペクトルの回帰曲線をN個のサンプルから求めて、回帰曲線の値で各サンプルを除する。N個のサンプルのうち、一部を用いて回帰曲線を求めてもよいし、外れ値を除外してから回帰曲線を求めることもできる。外れ値の除外によって、外れ値の悪影響を除外することができ、より効果的な回転量を求めることができる。このようにして得られた商は、1を中心とした分布となる。
式(27)において、|Xチルダn(k)|は、包絡線で正規化した劣化信号振幅スペクトルである。このようにして回転量を求めることによって、回転量が均一にπと-πの間に分布し、相関除去の効果を強化することができる。なお、このときにπを乗じずに商をそのまま回転量としても、類似の効果が得られる。
本発明の第4実施形態について、図7を用いて説明する。本実施形態に係る雑音抑圧装置700は、位相制御部202での位相の制御による出力レベルの低下を振幅制御部708を用いて補填する点で第2実施形態と異なる。他の構成および動作は第2実施形態と同様であるためここではその説明を省略する。
また、x1とx2の関係は、以下の式(29)のようになる。
逆フーリエ変換部401は、逆フーリエ変換により周波数領域信号を時間領域信号に変換する。その後、フレーム合成部403は、前フレームと現フレームの強調音声をオーバーラップ加算する。
さらに、式(36)中の周波数領域信号X1[k],X2[k]に式(32),式(33)を代入すると、以下の式(37)のように表現される。
さらに、式(37)を展開すると、以下の式(38)のように表現される。
式(40)より、式(39)は、以下の式(42)に変形できる。
式(42)の関係から、式(38)は、以下の式(43)で表わされる。
よって、式(38)は、以下の式(44)となる。
さて、ここで周波数領域信号X2[k]に対し、位相回転を行った場合を考える。このときの時間領域信号は、図12のようになる。
これを、式(36)に代入すると、以下の式(46)が成立する。
これを展開すると、以下の式(47)が成立する。
ここで、それぞれの項にある括弧内の項(式(49)に示す)は、ベクトル合成であるから、特定の周波数kに注目すると、図13のように描ける。
もし、位相回転が行なわれていないとき、つまりφ[k]=0の場合は、図14のようになる。
よって、式(49)の絶対値が最大になる条件は、φ[k]=0の場合であり、その値は2である。つまり、位相回転が行なわれると、出力信号の大きさが小さくなることがわかる。
補正量算出部881は、位相を標準正規分布の正規乱数で回転させる場合、振幅補正部882に補正係数をsqrt(1/0.805)として、伝達する。位相制御部202は、位相回転を全ての周波数に対して行ってもよいし、一部の周波数に対して行ってもよい。振幅制御部708は、振幅補正を、位相回転が行なわれた周波数に対してのみ行なう。よって、位相回転を行なわない周波数の補正係数は、1.0とする。位相回転を行った周波数の補正係数のみ導出した値とする。
本発明の第5実施形態にかかる雑音抑圧装置1500について、図15を用いて説明する。本実施形態では、振幅制御部708を備えている点で、第3実施形態の構成と異なる。振幅制御部708以外の構成については、第3実施形態と同様であり、振幅制御部708は、第4実施形態と同様であるため、同じ構成については同じ符号を付してその詳しい説明を省略する。
本発明の第6実施形態に係る雑音抑圧装置1600について、図16を用いて説明する。本実施形態では、位相の回転量の上限を制限する点で第2実施形態と異なる。それ以外の構成および動作については第2実施形態と同様であるためここでは詳しい説明を省略する。
本発明の第7実施形態について、図17を用いて説明する。本実施形態に係る本発明は、位相成分を遅延させ、フレーム間での位相成分の差分を求めた上でそこから補正量を算出する点で第4実施形態と異なる。つまり、振幅制御部1708の内部構成において第2実施形態と異なる。他の構成および動作は第2実施形態と同様であるためここではその説明を省略する。
本発明の第8実施形態について、図18を用いて説明する。図18は、本実施形態に係る位相制御部202および振幅制御部1808の構成を示すブロック図である。
本発明の第9実施形態について、図19を用いて説明する。図19は、本実施形態に係る振幅制御部1908の構成を示すブロック図である。図19に示すように、本実施形態における振幅制御部1908は、第8実施形態に含まれる入出力比算出部1881に加え、平均化処理部1981を含む。平均化処理部1981以外の構成および動作については、第8実施形態と同様であるためここではその説明を省略する。
本発明の第10実施形態について、図20および図21を用いて説明する。図20は、本実施形態に係る雑音抑圧装置2000の構成を示す図である。本実施形態に係る雑音抑圧装置2000は、第2実施形態の図2の構成に加え、振幅成分遅延部2011、位相成分遅延部2012および逆変換部2013を含む。また、振幅制御部2008の内部構成にも差異がある。本実施形態において、振幅成分遅延部2011、位相成分遅延部2012および振幅制御部2008以外の動作については、第2実施形態と同様であるためここではその説明を省略する。
本発明の第11実施形態について、図22を用いて説明する。図22に示すように、本実施形態に係る雑音抑圧装置2200は、第4実施形態の構成に加え、フレームオーバーラップ制御部2208を含む。フレームオーバーラップ制御部2208は、変換部201および逆変換部204において、フレーム分割、合成されるときのオーバーラップ率の制御を行なう。フレームオーバーラップ制御部2208は、そのオーバーラップ率を振幅制御部708に供給する。既に説明したとおり、位相回転によるレベル低下はオーバーラップによって生じる。このレベル低下量はオーバーラップ率により変化し、オーバーラップ率が大きくなるほど、低下量も大きくなる。よって、オーバーラップ率が変化した場合には、振幅補正量を制御する必要がある。具体的には、オーバーラップ率50%の場合の振幅補正量Gを基準に補正量を求める。
例えば、オーバーラップ率50%の場合は、Q=L/2であるので、以下の式(65)
が成立する。オーバーラップ率25%の場合には、Q=L/4より以下の式(66)が成立する。
以上説明してきた第1乃至第11実施形態では、それぞれ別々の特徴を持つ雑音抑圧装置について説明したが、それらの特徴を如何様に組み合わせた雑音抑圧装置も、本発明の範疇に含まれる。
この出願は、2012年11月27日に出願された日本出願特願2012-259218を基礎とする優先権を主張し、その開示の全てをここに取り込む。
Claims (15)
- 第1信号と第2信号とが混在した混在信号を、周波数ごとの位相成分および周波数ごとの振幅成分またはパワー成分に変換する変換手段と、
前記位相成分よりも相互相関が弱く、乱数よりもランダム性の低いデータ系列を用いて、所定周波数の前記位相成分の変化量を生成する変化量生成手段と、
前記変化量生成手段から提供された変化量を用いて、前記位相成分を制御する位相制御手段と、
前記位相制御手段によって制御処理を加えられた位相成分を用いて強調信号を生成する逆変換手段と、
を備えたことを特徴とする信号処理装置。 - 前記変化量生成手段は、前記変換手段によって導き出された前記位相成分に基づくデータ系列を用いて変化量を生成することを特徴とする請求項1に記載の信号処理装置。
- 前記変化量生成手段は、前記変換手段によって導き出された前記位相成分を、少なくとも1サンプル値おきに符号反転したデータ系列を用いて変化量を生成することを特徴とする請求項2に記載の信号処理装置。
- 前記変化量生成手段は、前記変換手段によって導き出された前記位相成分を少なくとも1サンプルずらせたデータ系列を用いて、変化量を生成することを特徴とする請求項2または3に記載の信号処理装置。
- 前記変化量生成手段は、1フレーム中の全サンプル数の半分の位置を中心として元の位相成分と対称の位置にある位相成分を変化量とすることを特徴とする請求項2、3または4に記載の信号処理装置。
- 前記変化量生成手段は、前記変換手段によって導き出された前記位相成分を、1フレームのサンプル中で、交換したデータ系列を変化量とすることを特徴とする請求項2乃至5のいずれか1項に記載の信号処理装置。
- 前記変化量生成手段は、前記位相成分の隣接するサンプル間での相関を求め、求めた相関を除去すべく前記変化量を定めることを特徴とする請求項2乃至6のいずれか1項に記載の信号処理装置。
- 前記変化量生成手段は、前記データ系列として、所定の周波数よりも高い周波数の振幅成分の複数の値を用いることを特徴とする請求項1乃至7のいずれか1項に記載の信号処理装置。
- 前記変化量生成手段は、前記データ系列として、所定の周波数よりも高い周波数の振幅成分を、周波数方向に集めた複数の値を用いることを特徴とする請求項1乃至8のいずれか1項に記載の信号処理装置。
- 前記変化量生成手段は、前記データ系列として、特定の周波数の時間軸に沿った振幅成分の複数の値を用いることを特徴とする請求項1乃至9のいずれか1項に記載の信号処理装置。
- 前記変化量生成手段は、特定の周波数の前記振幅成分を保持する振幅保持手段を含み、保持した前記振幅成分のうち、所定の周波数の振幅成分を前記位相成分の変化量として用いることを特徴とする請求項1乃至10のいずれか1項に記載の信号処理装置。
- 前記変化量生成手段は、前記変換手段が変換した前記位相成分を入力し、前記位相成分との相関の弱い変化量を生成することを特徴とする請求項1乃至11のいずれか1項に記載の信号処理装置。
- 前記位相制御手段は、前記変化量生成手段から提供された変化量を用いて、前記位相成分を置換または回転することを特徴とする請求項1乃至12のいずれか1項に記載の信号処理装置。
- 第1信号と第2信号とが混在した混在信号を、周波数ごとの位相成分および周波数ごとの振幅成分またはパワー成分に変換する変換ステップと、
前記位相成分よりも相互相関が弱く、乱数よりもランダム性の低いデータ系列を用いて、所定周波数の前記位相成分の変化量を生成する変化量生成ステップと、
前記変化量生成ステップで生成された変化量を用いて、前記位相成分を制御する位相制御ステップと、
前記位相制御ステップにおいて制御処理を加えられた位相成分を用いて強調信号を生成する逆変換ステップと、
を含むことを特徴とする信号処理方法。 - 第1信号と第2信号とが混在した混在信号を、周波数ごとの位相成分および周波数ごとの振幅成分またはパワー成分に変換する変換ステップと、
前記位相成分よりも相互相関が弱く、乱数よりもランダム性の低いデータ系列を用いて、所定周波数の前記位相成分の変化量を生成する変化量生成ステップと、
前記変化量生成ステップで生成された変化量を用いて、前記位相成分を制御する位相制御ステップと、
前記位相制御ステップにおいて制御処理を加えられた位相成分を用いて強調信号を生成する逆変換ステップと、
をコンピュータに実行させることを特徴とする信号処理プログラム。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014550098A JP6350871B2 (ja) | 2012-11-27 | 2013-11-01 | 信号処理装置、信号処理方法、および信号処理プログラム |
US14/647,175 US10447516B2 (en) | 2012-11-27 | 2013-11-01 | Signal processing apparatus, signal processing method, and signal processing program |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2012-259218 | 2012-11-27 | ||
JP2012259218 | 2012-11-27 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2014083999A1 true WO2014083999A1 (ja) | 2014-06-05 |
Family
ID=50827645
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/079701 WO2014083999A1 (ja) | 2012-11-27 | 2013-11-01 | 信号処理装置、信号処理方法、および信号処理プログラム |
Country Status (3)
Country | Link |
---|---|
US (1) | US10447516B2 (ja) |
JP (2) | JP6350871B2 (ja) |
WO (1) | WO2014083999A1 (ja) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1998020483A1 (fr) * | 1996-11-07 | 1998-05-14 | Matsushita Electric Industrial Co., Ltd. | Generateur de vecteur de source sonore, codeur et decodeur vocal |
JPH10149198A (ja) * | 1996-11-21 | 1998-06-02 | Matsushita Electric Ind Co Ltd | ノイズ削減装置 |
WO1999030315A1 (fr) * | 1997-12-08 | 1999-06-17 | Mitsubishi Denki Kabushiki Kaisha | Procede et dispositif de traitement du signal sonore |
WO2000011646A1 (fr) * | 1998-08-21 | 2000-03-02 | Matsushita Electric Industrial Co., Ltd. | Codeur et decodeur de la parole multimodes |
JP2004272292A (ja) * | 1997-12-08 | 2004-09-30 | Mitsubishi Electric Corp | 音信号加工方法 |
WO2007029536A1 (ja) * | 2005-09-02 | 2007-03-15 | Nec Corporation | 雑音抑圧の方法及び装置並びにコンピュータプログラム |
WO2012070671A1 (ja) * | 2010-11-24 | 2012-05-31 | 日本電気株式会社 | 信号処理装置、信号処理方法、及び信号処理プログラム |
WO2012114628A1 (ja) * | 2011-02-26 | 2012-08-30 | 日本電気株式会社 | 信号処理装置、信号処理方法、及び記憶媒体 |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6952460B1 (en) * | 2001-09-26 | 2005-10-04 | L-3 Communications Corporation | Efficient space-time adaptive processing (STAP) filter for global positioning system (GPS) receivers |
US7248656B2 (en) * | 2002-12-02 | 2007-07-24 | Nortel Networks Limited | Digital convertible radio SNR optimization |
US20050129139A1 (en) * | 2003-12-03 | 2005-06-16 | Jones Aled W. | Tag tracking |
US20060256978A1 (en) * | 2005-05-11 | 2006-11-16 | Balan Radu V | Sparse signal mixing model and application to noisy blind source separation |
US8050446B2 (en) * | 2005-07-12 | 2011-11-01 | The Board Of Trustees Of The University Of Arkansas | Method and system for digital watermarking of multimedia signals |
JP5018193B2 (ja) | 2007-04-06 | 2012-09-05 | ヤマハ株式会社 | 雑音抑圧装置およびプログラム |
US8046214B2 (en) * | 2007-06-22 | 2011-10-25 | Microsoft Corporation | Low complexity decoder for complex transform coding of multi-channel sound |
CA2767074C (en) * | 2009-08-24 | 2014-08-19 | Huawei Technologies Co., Ltd. | Clock recovery apparatus |
US8891491B2 (en) * | 2012-06-15 | 2014-11-18 | Intel Mobile Communications GmbH | Method of processing signals and a signal processor |
-
2013
- 2013-11-01 US US14/647,175 patent/US10447516B2/en active Active
- 2013-11-01 JP JP2014550098A patent/JP6350871B2/ja active Active
- 2013-11-01 WO PCT/JP2013/079701 patent/WO2014083999A1/ja active Application Filing
-
2018
- 2018-06-08 JP JP2018110354A patent/JP6662413B2/ja active Active
Patent Citations (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6330534B1 (en) * | 1996-11-07 | 2001-12-11 | Matsushita Electric Industrial Co., Ltd. | Excitation vector generator, speech coder and speech decoder |
EP0883107A1 (en) * | 1996-11-07 | 1998-12-09 | Matsushita Electric Industrial Co., Ltd | Sound source vector generator, voice encoder, and voice decoder |
CN1207195A (zh) * | 1996-11-07 | 1999-02-03 | 松下电器产业株式会社 | 声源矢量生成装置以及声音编码装置和声音解码装置 |
WO1998020483A1 (fr) * | 1996-11-07 | 1998-05-14 | Matsushita Electric Industrial Co., Ltd. | Generateur de vecteur de source sonore, codeur et decodeur vocal |
JPH10149198A (ja) * | 1996-11-21 | 1998-06-02 | Matsushita Electric Ind Co Ltd | ノイズ削減装置 |
WO1999030315A1 (fr) * | 1997-12-08 | 1999-06-17 | Mitsubishi Denki Kabushiki Kaisha | Procede et dispositif de traitement du signal sonore |
JP2004272292A (ja) * | 1997-12-08 | 2004-09-30 | Mitsubishi Electric Corp | 音信号加工方法 |
US6526378B1 (en) * | 1997-12-08 | 2003-02-25 | Mitsubishi Denki Kabushiki Kaisha | Method and apparatus for processing sound signal |
EP1041539A1 (en) * | 1997-12-08 | 2000-10-04 | Mitsubishi Denki Kabushiki Kaisha | Sound signal processing method and sound signal processing device |
CN1281576A (zh) * | 1997-12-08 | 2001-01-24 | 三菱电机株式会社 | 声音信号加工方法和声音信号加工装置 |
JP2002023800A (ja) * | 1998-08-21 | 2002-01-25 | Matsushita Electric Ind Co Ltd | マルチモード音声符号化装置及び復号化装置 |
US6334105B1 (en) * | 1998-08-21 | 2001-12-25 | Matsushita Electric Industrial Co., Ltd. | Multimode speech encoder and decoder apparatuses |
CN1275228A (zh) * | 1998-08-21 | 2000-11-29 | 松下电器产业株式会社 | 多模式语音编码装置及解码装置 |
EP1024477A1 (en) * | 1998-08-21 | 2000-08-02 | Matsushita Electric Industrial Co., Ltd. | Multimode speech encoder and decoder |
WO2000011646A1 (fr) * | 1998-08-21 | 2000-03-02 | Matsushita Electric Industrial Co., Ltd. | Codeur et decodeur de la parole multimodes |
EP1930880A1 (en) * | 2005-09-02 | 2008-06-11 | NEC Corporation | Method and device for noise suppression, and computer program |
KR20080042166A (ko) * | 2005-09-02 | 2008-05-14 | 닛본 덴끼 가부시끼가이샤 | 잡음 억압을 위한 방법과 장치, 및 컴퓨터 프로그램 |
WO2007029536A1 (ja) * | 2005-09-02 | 2007-03-15 | Nec Corporation | 雑音抑圧の方法及び装置並びにコンピュータプログラム |
CN101300623A (zh) * | 2005-09-02 | 2008-11-05 | 日本电气株式会社 | 用于抑制噪声的方法、设备和计算机程序 |
US20090196434A1 (en) * | 2005-09-02 | 2009-08-06 | Nec Corporation | Method, apparatus, and computer program for suppressing noise |
JP5092748B2 (ja) * | 2005-09-02 | 2012-12-05 | 日本電気株式会社 | 雑音抑圧の方法及び装置並びにコンピュータプログラム |
WO2012070671A1 (ja) * | 2010-11-24 | 2012-05-31 | 日本電気株式会社 | 信号処理装置、信号処理方法、及び信号処理プログラム |
CN103250208A (zh) * | 2010-11-24 | 2013-08-14 | 日本电气株式会社 | 信号处理装置、信号处理方法和信号处理程序 |
WO2012114628A1 (ja) * | 2011-02-26 | 2012-08-30 | 日本電気株式会社 | 信号処理装置、信号処理方法、及び記憶媒体 |
Non-Patent Citations (1)
Title |
---|
AKIHIKO SUGIYAMA: "Single-Channel Impact-Noise Suppression with No Auxiliary Information for its Detection", 2007 IEEE WORKSHOP ON APPLICATIONS OF SIGNAL PROCESSING TO AUDIO AND ACOUSTICS, October 2007 (2007-10-01), pages 127 - 130 * |
Also Published As
Publication number | Publication date |
---|---|
JPWO2014083999A1 (ja) | 2017-01-05 |
US20150295741A1 (en) | 2015-10-15 |
JP6662413B2 (ja) | 2020-03-11 |
US10447516B2 (en) | 2019-10-15 |
JP6350871B2 (ja) | 2018-07-04 |
JP2018156108A (ja) | 2018-10-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6079236B2 (ja) | 信号処理装置、信号処理方法、及び信号処理プログラム | |
CN105788607B (zh) | 应用于双麦克风阵列的语音增强方法 | |
US9318119B2 (en) | Noise suppression using integrated frequency-domain signals | |
US10811026B2 (en) | Noise suppression method, device, and program | |
JP6300031B2 (ja) | 信号処理装置、信号処理方法、および信号処理プログラム | |
JP6070953B2 (ja) | 信号処理装置、信号処理方法、及び記憶媒体 | |
WO2014136628A1 (ja) | 信号処理装置、信号処理方法および信号処理プログラム | |
JP5788873B2 (ja) | 信号処理方法、情報処理装置、及び信号処理プログラム | |
WO2012070670A1 (ja) | 信号処理装置、信号処理方法、及び信号処理プログラム | |
JP2008216721A (ja) | 雑音抑圧の方法、装置、及びプログラム | |
US20080215330A1 (en) | Audio Signal Modification | |
JP6662413B2 (ja) | 信号処理装置、信号処理方法、および信号処理プログラム | |
JP5413575B2 (ja) | 雑音抑圧の方法、装置、及びプログラム | |
JP6182862B2 (ja) | 信号処理装置、信号処理方法、及び信号処理プログラム | |
JP2011100030A (ja) | 信号処理方法、情報処理装置、及び信号処理プログラム | |
US11488574B2 (en) | Method and system for implementing a modal processor | |
WO2013032025A1 (ja) | 信号処理装置、信号処理方法、およびコンピュータ・プログラム | |
JP6119604B2 (ja) | 信号処理装置、信号処理方法、および信号処理プログラム | |
KR20210102695A (ko) | 신호 필터링 장치 및 방법 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 13858895 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2014550098 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14647175 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 13858895 Country of ref document: EP Kind code of ref document: A1 |