WO2015129760A1 - 信号処理装置、方法及びプログラム - Google Patents
信号処理装置、方法及びプログラム Download PDFInfo
- Publication number
- WO2015129760A1 WO2015129760A1 PCT/JP2015/055442 JP2015055442W WO2015129760A1 WO 2015129760 A1 WO2015129760 A1 WO 2015129760A1 JP 2015055442 W JP2015055442 W JP 2015055442W WO 2015129760 A1 WO2015129760 A1 WO 2015129760A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- noise
- component
- derived
- target area
- spectral density
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000012545 processing Methods 0.000 title claims abstract description 45
- 230000003595 spectral effect Effects 0.000 claims abstract description 47
- 238000000605 extraction Methods 0.000 claims abstract description 28
- 230000008569 process Effects 0.000 claims abstract description 24
- 238000012935 Averaging Methods 0.000 claims abstract description 12
- 239000000284 extract Substances 0.000 claims abstract description 6
- 238000009499 grossing Methods 0.000 claims description 17
- 238000007493 shaping process Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000001228 spectrum Methods 0.000 claims description 8
- 238000003672 processing method Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 19
- 238000012546 transfer Methods 0.000 description 7
- 238000009408 flooring Methods 0.000 description 6
- 230000001629 suppression Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 4
- 230000001427 coherent effect Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 230000017105 transposition Effects 0.000 description 2
- 238000004378 air conditioning Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- 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
-
- 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
- G10L21/0316—Speech enhancement, e.g. noise reduction or echo cancellation by changing the amplitude
- G10L21/0324—Details of processing therefor
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R1/00—Details of transducers, loudspeakers or microphones
- H04R1/20—Arrangements for obtaining desired frequency or directional characteristics
- H04R1/32—Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only
- H04R1/40—Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only by combining a number of identical transducers
- H04R1/406—Arrangements for obtaining desired frequency or directional characteristics for obtaining desired directional characteristic only by combining a number of identical transducers microphones
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R3/00—Circuits for transducers, loudspeakers or microphones
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R3/00—Circuits for transducers, loudspeakers or microphones
- H04R3/005—Circuits for transducers, loudspeakers or microphones for combining the signals of two or more microphones
-
- 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
- G10L2021/02161—Number of inputs available containing the signal or the noise to be suppressed
- G10L2021/02166—Microphone arrays; Beamforming
Definitions
- the present invention relates to a technology for clearly collecting sound source signals arriving from a target direction using several microphones.
- M is an integer of 2 or more. For example, M is about 2 to 4. M may be about 100.
- K interference noises S k ( ⁇ , ⁇ ) (k 1, 2,..., K) and incoherent stationary noise N m ( ⁇ , ⁇ ).
- K be a predetermined positive integer.
- m is the number of each microphone
- the observation signal X m ( ⁇ , ⁇ ) is a signal obtained by converting a time domain signal collected by the microphone m into a frequency domain.
- the target sound is sound coming from a predetermined target area.
- the target area is an area including a sound source to be collected.
- the number of sound sources to be collected and the position of the sound source to be collected in the target area may be unknown. For example, as illustrated in FIG. 6, it is assumed that an area where six speakers and three microphones are arranged is divided into three areas (area 1, area 2, and area 3). When the sound source to be collected is included in area 1, area 1 is the target area.
- the target sound may include a reflected sound from a sound source outside the target area.
- the target sound may include sound arriving at the microphone from the direction of area 1 due to reflection among sounds generated from sound sources included in area 2 and area 3. .
- the target area may be an area within a predetermined distance from the microphone. In other words, it may be an area having a finite area. Furthermore, there may be a plurality of target areas.
- FIG. 7 is a diagram illustrating an example when there are two target areas.
- an area including a sound source that generates noise is also referred to as a noise area.
- each of area 2 and area 3 is a noise area.
- each of area 2 and area 3 is a noise area, but an area that is a combination of area 2 and area 3 may be a noise area.
- a noise area including a sound source that emits interference noise is particularly called an interference noise area. The noise area is set to be different from the target area.
- the transfer characteristic from the mth microphone to the target sound S 0 ( ⁇ , ⁇ ) is described as A m, 0 ( ⁇ ), and the transfer characteristic from the mth microphone to the kth interference noise is denoted by A m, k ( When ⁇ ) is described, the observed signal X m ( ⁇ , ⁇ ) is modeled as follows.
- FIG. 1 shows a processing flow of the post-filter type array.
- the filter coefficient w 0 ( ⁇ ) [W 0,1 ( ⁇ ),..., W 0, M ( ⁇ )] T designed to enhance the target sound is calculated as follows.
- x is an arbitrary vector or matrix
- xT means transposition of x
- xH means conjugate transposition of x
- h 0 ( ⁇ ) [H 0,1 ( ⁇ ),..., H 0, M ( ⁇ )]
- T is an array manifold vector in the target sound direction.
- the array manifold vector, and the transfer characteristic H 0 from the sound source to the microphone, m the (omega) which was the vector h 0 (omega), the transfer characteristic H 0 from the sound source to the microphone, m (omega) is the sound source Transfer characteristics assuming only direct sound that can be theoretically calculated from the microphone position, measured transfer characteristics, and transfer characteristics estimated by computer simulation such as mirror image method and finite element method. Assuming that the source signals are uncorrelated with each other, the spatial correlation matrix R ( ⁇ ) can be modeled as follows:
- h k ( ⁇ ) is an array manifold vector of the k-th interference noise.
- the beamforming output signal Y 0 ( ⁇ , ⁇ ) is obtained by the following equation.
- x ( ⁇ , ⁇ ) [X 1 ( ⁇ , ⁇ ),..., X M ( ⁇ , ⁇ )] T.
- the post filter G ( ⁇ , ⁇ ) is multiplied.
- the output signal is obtained by performing inverse fast Fourier transform (IFFT) on Z ( ⁇ , ⁇ ).
- IFFT inverse fast Fourier transform
- Non-Patent Document 2 proposes a method of designing a post filter based on the power spectrum density (PSD) of each area estimated using a plurality of beam forming (for example, see Non-Patent Document 2).
- this method is referred to as an LPSD method (Local PSD-based post-filter design). The processing flow of the LPSD method will be described with reference to FIG.
- G ( ⁇ , ⁇ ) is calculated as follows.
- ⁇ S ( ⁇ , ⁇ ) represents the power spectral density of the target area
- ⁇ N ( ⁇ , ⁇ ) represents the power spectral density of the noise area
- the power spectrum density of a certain area means the power spectrum density of sound coming from that area. That is, for example, the power spectral density of the target area is the power spectral density of sound coming from the target area
- the power spectral density of the noise area is the power spectral density of sound coming from the noise area.
- 2 is, for example,
- 2
- 2 an actual measurement value may be used.
- ⁇ Y ( ⁇ , ⁇ ) [
- ⁇ S ( ⁇ , ⁇ ) [
- the power spectrum density of each area is calculated by solving the inverse problem of equation (7).
- b + represents a pseudo inverse matrix operation on b, where b is an arbitrary matrix.
- the local PSD estimation unit estimates and outputs the power spectral density ⁇ ⁇ S ( ⁇ , ⁇ ) of each area.
- the target area / noise area PSD estimation unit 12 is defined by the following equation with the local power spectral density ⁇ ⁇ S ( ⁇ , ⁇ ) estimated based on the equation (8) for each frequency ⁇ and frame ⁇ as an input. ⁇ S ( ⁇ , ⁇ ) and ⁇ ⁇ N ( ⁇ , ⁇ ) are calculated.
- the Wiener gain calculation unit 13 calculates the post filter G ( ⁇ , ⁇ ) defined by the equation (6) with ⁇ ⁇ S ( ⁇ , ⁇ ) and ⁇ ⁇ N ( ⁇ , ⁇ ) as inputs. Output. Specifically, the Wiener gain calculation unit 13 obtains ⁇ ⁇ S ( ⁇ , ⁇ ) and ⁇ ⁇ N ( ⁇ , ⁇ ) as ⁇ S ( ⁇ , ⁇ ) and ⁇ N ( ⁇ , ⁇ ) in Equation (6), respectively. ), G ( ⁇ , ⁇ ) is calculated and output.
- the main advantages of the LPSD method are the following two.
- (i) The relationship between the beamforming output and each sound source is formulated in the power spectrum region, and control freedom exceeding the number of microphones can be obtained, so that noise can be effectively suppressed, and
- (ii) L If the beam forming filter w u ( ⁇ ) (u 0, 1,..., L) and D ( ⁇ ) in Equation (7) are calculated in advance, the advantage of (i) can be implemented with low computation. .
- the LPSD method has formulated the problem on the assumption that the target sound and interference noise are mixed.
- practical problems often include not only coherent interference noise but also stationary noise with high incoherence (air conditioning noise, microphone internal noise, etc.).
- estimation errors of ⁇ S ( ⁇ , ⁇ ) and ⁇ N ( ⁇ , ⁇ ) become large, and noise suppression performance may be deteriorated.
- An object of the present invention is to provide a signal processing apparatus, method, and program in which noise suppression performance is improved as compared with the prior art.
- a signal processing apparatus includes a target area and at least one different from the target area based on a frequency domain observation signal obtained from signals collected by M microphones constituting a microphone array.
- a local PSD estimator for estimating the local power spectral density of each noise area, and ⁇ as a frequency and ⁇ as a frame index, based on the estimated local power spectral density ⁇ ⁇ From the target area / noise area PSD estimation unit for estimating the power spectral density ⁇ ⁇ N ( ⁇ , ⁇ ) of S ( ⁇ , ⁇ ) and the noise area, and the power spectral density ⁇ ⁇ S ( ⁇ , ⁇ ) of the target area, Due to non-stationary components ⁇ ⁇ S (A) ( ⁇ , ⁇ ) and incoherent noise derived from the sound coming from the target area From the first component extractor that extracts the incoming steady component ⁇ ⁇ S (B) ( ⁇ , ⁇ ) and the noise power spectrum density ⁇ ⁇ N ( ⁇ , ⁇ ), the unsteady component
- Noise suppression performance can be improved compared to the conventional one.
- mold array The block diagram of the conventional post filter estimation part.
- the LPSD method is extended to estimate the post filter robustly against various noise environments. Specifically, the estimation error of the ratio between the power of the target sound and the power of other noise is reduced by estimating the power spectral density by dividing each noise type.
- FIG. 3 shows a block diagram of an example of the post filter estimation unit 1 which is a signal processing device according to an embodiment of the present invention.
- the signal processing apparatus includes a local PSD estimation unit 11, a target area / noise area PSD estimation unit 12, a first component extraction unit 14, a second component extraction unit 15, and multi-noise compatible gain calculation.
- a unit 16 a time frequency averaging unit 17, and a gain shaping unit 18 are provided.
- FIG. 4 shows each step of signal processing realized by this signal processing device, for example.
- the local PSD estimation unit 11 is the same as the conventional local PSD estimation unit 11.
- the estimated local power spectral density ⁇ ⁇ S ( ⁇ , ⁇ ) is output to the target area / noise area PSD estimation unit 12.
- the local PSD estimation unit 11 may prepare a plurality of filter sets and select a filter that takes the maximum power.
- the target area / noise area PSD estimation unit 12 is the same as the conventional target area / noise area PSD estimation unit 12.
- the target area / noise area PSD estimation unit 12 determines the power spectral density ⁇ ⁇ S ( ⁇ , ⁇ ) of the target area and the power spectral density ⁇ ⁇ N ( ⁇ of the noise area based on the estimated local power spectral density. , ⁇ ) is estimated (step S2).
- the estimated power spectral density ⁇ ⁇ S ( ⁇ , ⁇ ) of the target area is output to the first component extraction unit 14.
- the estimated power spectral density of the noise area ⁇ ⁇ N ( ⁇ , ⁇ ) is output to the second component extraction unit 15.
- ⁇ ⁇ S ( ⁇ , ⁇ ) defined by equation (9) is derived from the non-stationary component ⁇ ⁇ S (A) ( ⁇ , ⁇ ) derived from the sound coming from the target area and incoherent noise.
- the stationary component is a component with little temporal change
- the unsteady component is a component with much temporal change.
- Interference noise is noise generated from a noise source arranged in a noise area.
- the incoherent noise is noise that is not emitted from the target area and the noise area, but is emitted from a place other than these areas and exists constantly.
- the first component extraction unit 14 determines the unsteady component ⁇ ⁇ S (A) ( ⁇ , ⁇ ) derived from the sound arriving from the target area from the power spectral density ⁇ ⁇ S ( ⁇ , ⁇ ) of the target area.
- a stationary component ⁇ ⁇ S (B) ( ⁇ , ⁇ ) derived from incoherent noise is extracted by a smoothing process (step S3).
- the smoothing process is realized by an exponential moving average process, a time average process, or a weighted average process as in Expression (11) and Expression (12).
- the first component extraction unit 14 performs exponential moving average processing as in Expression (11) and Expression (12), so that ⁇ ⁇ S ( ⁇ , ⁇ ) to ⁇ ⁇ S (B) ( ⁇ , ⁇ ).
- ⁇ S is a set of frames index for a specific section. For example, the specific section is set to be about 3 to 4 seconds. min is a function that outputs the minimum value.
- ⁇ ⁇ S (B) ( ⁇ , ⁇ ) is a component obtained by smoothing ⁇ ⁇ S ( ⁇ , ⁇ ) by, for example, Equation (11) and Equation (12). More specifically, ⁇ ⁇ S (B) ( ⁇ , ⁇ ) is a minimum value in a predetermined time interval of a value obtained by smoothing ⁇ ⁇ S ( ⁇ , ⁇ ) by, for example, the equation (11).
- the first component extraction unit 14 as in Equation (13), ⁇ ⁇ S ( ⁇ , ⁇ ) from ⁇ ⁇ S (B) ( ⁇ , ⁇ ) by subtracting ⁇ ⁇ S (A) ( ⁇ , ⁇ ) is calculated.
- ⁇ S ( ⁇ ) is a weighting coefficient, which is a predetermined positive real number.
- ⁇ S ( ⁇ ) is set to a real number of about 1 to 3, for example.
- ⁇ S (A) ( ⁇ , ⁇ ) is a ⁇ ⁇ S ( ⁇ , ⁇ ) from ⁇ ⁇ S (B) ( ⁇ , ⁇ ) components except.
- ⁇ ⁇ S (A) ( ⁇ , ⁇ ) may be floored so as to satisfy the condition of ⁇ ⁇ S (A) ( ⁇ , ⁇ ) ⁇ 0.
- This flooring process is performed by, for example, the first component extraction unit 14.
- ⁇ Second component extraction unit 15 For example, ⁇ ⁇ N ( ⁇ , ⁇ ) defined by equation (10) includes non-stationary components derived from interference noise ⁇ ⁇ N (A) ( ⁇ , ⁇ ) and stationary components derived from incoherent noise ⁇ ⁇ N (B) ( ⁇ , ⁇ ) is included.
- the second component extraction unit 15 determines the unsteady component ⁇ ⁇ N (A) ( ⁇ , ⁇ ) derived from the interference noise and the incoherent noise from the power spectral density ⁇ ⁇ N ( ⁇ , ⁇ ) in the noise area.
- the steady component ⁇ ⁇ N (B) ( ⁇ , ⁇ ) derived from is extracted by a smoothing process (step S4).
- the smoothing process is realized by an exponential moving average process, a time average process, or a weighted average process like Expression (14) and Expression (15).
- the non-stationary component ⁇ ⁇ N (A) ( ⁇ , ⁇ ) derived from the extracted interference noise and the stationary component ⁇ ⁇ N (B) ( ⁇ , ⁇ ) derived from incoherent noise are It is output to the calculation unit 16.
- the second component extraction unit 15 performs an exponential moving average process as in Expression (14) and Expression (15), so that ⁇ ⁇ N ( ⁇ , ⁇ ) to ⁇ ⁇ N (B) ( ⁇ , ⁇ ).
- ⁇ N is a set of frames index for a specific section. For example, the specific section is set to be about 3 to 4 seconds.
- ⁇ ⁇ N (B) ( ⁇ , ⁇ ) is a component obtained by smoothing ⁇ ⁇ N ( ⁇ , ⁇ ) by, for example, Expression (14) and Expression (15). More specifically, ⁇ ⁇ N (B) ( ⁇ , ⁇ ) is a minimum value in a predetermined time interval of a value obtained by smoothing ⁇ ⁇ N ( ⁇ , ⁇ ) by, for example, the equation (14).
- the second component extractor 15 as in Equation (16), ⁇ ⁇ N ( ⁇ , ⁇ ) from ⁇ ⁇ N (B) ( ⁇ , ⁇ ) by subtracting ⁇ ⁇ N (A) ( ⁇ , ⁇ ) is calculated.
- ⁇ N ( ⁇ ) is a weighting coefficient, which is a predetermined positive real number.
- ⁇ N ( ⁇ ) is set to a real number of about 1 to 3, for example.
- ⁇ N (A) ( ⁇ , ⁇ ) is a ⁇ ⁇ N ( ⁇ , ⁇ ) from ⁇ ⁇ N (B) ( ⁇ , ⁇ ) components except.
- ⁇ ⁇ N (A) ( ⁇ , ⁇ ) may be floored so as to satisfy the condition of ⁇ ⁇ N (A) ( ⁇ , ⁇ ) ⁇ 0.
- This flooring process is performed by, for example, the second component extraction unit 15.
- ⁇ N may be the same as or different from ⁇ S.
- ⁇ N may be the same as or different from ⁇ S.
- ⁇ N ( ⁇ ) may be the same as or different from ⁇ S ( ⁇ ).
- the second component extraction unit 15 obtains ⁇ ⁇ N (B) ( ⁇ , ⁇ ). It does not have to be. In other words, in this case, the second component extraction unit 15 may obtain only ⁇ ⁇ N (A) ( ⁇ , ⁇ ) from ⁇ ⁇ N ( ⁇ , ⁇ ).
- the multi-noise compatible calculation unit 16 uses the non-stationary component ⁇ ⁇ S (A) ( ⁇ , ⁇ ) derived from the sound arriving from the target area and the stationary component ⁇ ⁇ S (B) ( Post-filter that emphasizes the unsteady component of sound coming from the target area using at least ⁇ , ⁇ ) and unsteady component ⁇ ⁇ N (A) ( ⁇ , ⁇ ) derived from interference noise ⁇ G ( ⁇ , ⁇ ) is calculated (step S5).
- the calculated post filter ⁇ G ( ⁇ , ⁇ ) is output to the time frequency averaging unit 17.
- the various noise corresponding gain calculation unit 16 uses, for example, the following equation (17). Calculate the defined post filter ⁇ G ( ⁇ , ⁇ ).
- the various noise corresponding gain calculation unit 16 may calculate a post filter ⁇ G ( ⁇ , ⁇ ) defined by the following equation (18).
- the time frequency averaging unit 17 performs a smoothing process on at least one of the time direction and the frequency direction for the post-filters ⁇ G ( ⁇ , ⁇ ) (step S6).
- the smoothed post filter ⁇ G ( ⁇ , ⁇ ) is output to the gain shaping unit 18.
- ⁇ 0 and ⁇ 1 are set to integers equal to or larger than 0, and the time frequency averaging unit 17 performs, for example, a post filter in the time direction of post filters to G ( ⁇ , ⁇ ). What is necessary is just to perform an addition average for ⁇ G ( ⁇ , ⁇ - ⁇ 0 ), ... ⁇ G ( ⁇ , ⁇ + ⁇ 1 ).
- the time frequency averaging unit 17 may perform weighted addition for ⁇ G ( ⁇ , ⁇ - ⁇ 0 ), ... ⁇ G ( ⁇ , ⁇ + ⁇ 1 ).
- ⁇ 0 and ⁇ 1 are set to real numbers of 0 or more, and the time frequency averaging unit 17 is, for example, in the frequency direction of the post filter to G ( ⁇ , ⁇ ). What is necessary is just to perform an addition average about ⁇ G ( ⁇ - ⁇ 0 , ⁇ ),... ⁇ G ( ⁇ + ⁇ 1 , ⁇ ) as post filters.
- the time frequency averaging unit 17 may perform weighted addition for ⁇ G ( ⁇ - ⁇ 0 , ⁇ ), ... ⁇ G ( ⁇ + ⁇ 1 , ⁇ ).
- the gain shaping unit 18 generates the post filter G ( ⁇ , ⁇ ) by performing gain shaping on the post filter to G ( ⁇ , ⁇ ) subjected to the smoothing process (step S7).
- the gain shaping unit 18 generates, for example, a post filter G ( ⁇ , ⁇ ) defined by the following equation (19).
- ⁇ is a weighting factor and is a positive real number.
- ⁇ may be set to about 1 to 1.3.
- the gain shaping unit 18 may perform a flooring process on the post filter G ( ⁇ , ⁇ ) so as to satisfy A ⁇ G ( ⁇ , ⁇ ) ⁇ 1.
- A is a real number from 0 to 0.3, usually about 0.1. If G ( ⁇ , ⁇ ) is larger than 1, there is a possibility of overemphasis, and if G ( ⁇ , ⁇ ) is too small, musical noise may be generated. By performing an appropriate flooring process, it is possible to prevent this enhancement and the generation of musical noise.
- a function f whose domain and range are real numbers.
- the function f is a non-decreasing function.
- Gain shaping means an operation for obtaining an output value when ⁇ G ( ⁇ , ⁇ ) before gain shaping is input to the function f.
- the output value when ⁇ G ( ⁇ , ⁇ ) is input to the function f is G ( ⁇ , ⁇ ).
- An example of the function f is Expression (19).
- FIG. 8 Another example of another function f will be described with reference to FIG. In FIG. 8, the index is omitted. That is, G in FIG. 8 means G ( ⁇ , ⁇ ), and ⁇ G means ⁇ G ( ⁇ , ⁇ ).
- G in FIG. 8 means G ( ⁇ , ⁇ )
- ⁇ G means ⁇ G ( ⁇ , ⁇ ).
- FIGS. 8B to 8C flooring processing is performed so as to satisfy 0 ⁇ G ( ⁇ , ⁇ ) ⁇ 1.
- the function specified by the graph indicated by the bold line in FIG. 8C is another example of the function f.
- the graph of the function f is not limited to that shown in FIG.
- the graph of the function f is composed of a straight line, but the graph of the function f may be composed of a curve.
- the function f may be a function obtained by performing a flooring process on a hyperbolic tangent function.
- this signal processing apparatus and method it is possible to design a post filter for suppressing noise robustly in an environment where noise having various properties exists.
- a post filter can be designed by processing with real-time characteristics.
- represent the index set and the total number of frames, respectively.
- represent the frequency bin index and the total number, respectively.
- the processing of the time frequency averaging unit 17 and the gain shaping unit 18 is performed to suppress so-called musical noise.
- the processing of the time frequency averaging unit 17 and the gain shaping unit 18 may not be performed.
- the first component extraction unit 14 may extract ⁇ ⁇ S (B) ( ⁇ , ⁇ ) and ⁇ ⁇ S (A) ( ⁇ , ⁇ ) by other processing.
- the second component extraction unit 15 may extract ⁇ ⁇ N (B) ( ⁇ , ⁇ ) and ⁇ ⁇ N (A) ( ⁇ , ⁇ ) by other processing.
- each unit in the signal processing device is realized by a computer
- processing contents of functions that each unit of the signal processing device should have are described by a program.
- each part is implement
- the program describing the processing contents can be recorded on a computer-readable recording medium.
- a computer-readable recording medium any recording medium such as a magnetic recording device, an optical disk, a magneto-optical recording medium, and a semiconductor memory may be used.
- each processing means may be configured by executing a predetermined program on a computer, or at least a part of these processing contents may be realized by hardware.
- Speech recognition is generally used as a command input for smartphones. Under noisy conditions such as in cars and factories, there is a high demand for hands-free operation of devices and calls with remote locations.
- This invention can be used, for example, in such a case.
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Otolaryngology (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Quality & Reliability (AREA)
- Multimedia (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
Description
局所PSD推定部11は、従来の局所PSD推定部11と同様である。
ターゲットエリア/雑音エリアPSD推定部12は、従来のターゲットエリア/雑音エリアPSD推定部12と同様である。
例えば式(9)により定義される^φS(ω,τ)には、ターゲットエリアから到来する音に由来する非定常成分^φS (A)(ω,τ)及びインコヒーレントな雑音に由来する定常成分^φS (B)(ω,τ)が含まれる。ここで、定常成分とは時間的に変化の少ない成分のことであり、非定常成分とは時間的に変化の多い成分のことである。
例えば式(10)により定義される^φN(ω,τ)には、干渉雑音に由来する非定常成分^φN (A)(ω,τ)及びインコヒーレントな雑音に由来する定常成分^φN (B)(ω,τ)が含まれる。
多様雑音対応型計算部16は、ターゲットエリアから到来する音に由来する非定常成分^φS (A)(ω,τ)と、インコヒーレントな雑音に由来する定常成分^φS (B)(ω,τ)と、干渉雑音に由来する非定常成分^φN (A)(ω,τ)とを少なくとも用いて、ターゲットエリアから到来する音の非定常成分を強調するポストフィルタ~G(ω,τ)を計算する(ステップS5)。
時間周波数平均化部17は、ポストフィルタ~G(ω,τ)について時間方向と周波数方向との少なくとも一方の方向への平滑化処理を行う(ステップS6)。
ゲインシェーピング部18は、平滑化処理が行われたポストフィルタ~G(ω,τ)についてゲインシェーピングを行うことにより、ポストフィルタG(ω,τ)を生成する(ステップS7)。ゲインシェーピング部18は、例えば、以下の式(19)により定義されるポストフィルタG(ω,τ)を生成する。
LPSD法を従来方式として、提案方式の効果を検証するための実験を行なった。図5のように、残響時間110ms(1.0kHz)の室に音源やアレイを配置した。ターゲット音(男女発話)、K=3個の干渉雑音(#1:男女発話、#2,3:音楽)、室の四隅のスピーカから白色雑音を放射して再現した背景雑音がある中で、M=4本の無指向性マイクロホンを用いて収録した。観測時のSN比は、平均-1dBであった。また、サンプリング周波数を16.0kHzとし、FFT解析長を512ptとし、FFTシフト長を256ptとした。
時間周波数平均化部17及びゲインシェーピング部18の処理は、いわゆるミュージカルノイズを抑えるために行われる。時間周波数平均化部17及びゲインシェーピング部18の処理は、行われなくてもよい。
Claims (6)
- マイクロホンアレーを構成するM個のマイクロホンで収音された信号から得られた周波数領域の観測信号に基づいて、所定のターゲットエリア及び上記ターゲットエリアと異なる少なくとも1個の雑音エリアのそれぞれの局所パワースペクトル密度を推定する局所PSD推定部と、
ωを周波数とし、τをフレームのインデックスとして、上記推定された局所パワースペクトル密度に基づいて、ターゲットエリアのパワースペクトル密度^φS(ω,τ)及び雑音エリアのパワースペクトル密度^φN(ω,τ)を推定するターゲットエリア/雑音エリアPSD推定部と、
上記ターゲットエリアのパワースペクトル密度^φS(ω,τ)から、ターゲットエリアから到来する音に由来する非定常成分^φS (A)(ω,τ)及びインコヒーレントな雑音に由来する定常成分^φS (B)(ω,τ)を抽出する第一成分抽出部と、
上記雑音エリアのパワースペクトル密度^φN(ω,τ)から、干渉雑音に由来する非定常成分^φN (A)(ω,τ)を抽出する第二成分抽出部と、
上記ターゲットエリアから到来する音に由来する非定常成分^φS (A)(ω,τ)と、上記インコヒーレントな雑音に由来する定常成分^φS (B)(ω,τ)と、上記干渉雑音に由来する非定常成分^φN (A)(ω,τ)とを少なくとも用いて、上記ターゲットエリアから到来する音の非定常成分を強調するポストフィルタ~G(ω,τ)を計算する多様雑音対応型ゲイン計算部と、
を含む信号処理装置。 - 請求項1の信号処理装置であって、
上記インコヒーレントな雑音に由来する定常成分^φS (B)(ω,τ)は、上記ターゲットエリアのパワースペクトル密度^φS(ω,τ)を平滑化した成分であり、
上記ターゲットエリアから到来する音に由来する非定常成分^φS (A)(ω,τ)は、上記ターゲットエリアのパワースペクトル密度^φS(ω,τ)から上記インコヒーレントな雑音に由来する定常成分^φS (B)(ω,τ)を除いた成分であり、
上記干渉雑音に由来する非定常成分^φN (A)(ω,τ)は、上記雑音エリアのパワースペクトル密度^φN(ω,τ)から上記雑音エリアのパワースペクトル密度^φN(ω,τ)を平滑化した成分を除いた成分である、
信号処理装置。 - 請求項1の信号処理装置であって、
上記第二成分抽出部は、上記雑音エリアのパワースペクトル密度^φN(ω,τ)から、干渉雑音に由来する非定常成分^φN (A)(ω,τ)を更に抽出し、
上記第一成分抽出部は、αSを所定の実数とし、ΥSを特定区間のフレームのインデックスの集合とし、βS(ω)を所定の実数とし、以下の式により定義される^φS (A)(ω,τ)及び^φS (B)(ω,τ)を計算し、計算された^φS (A)(ω,τ)を上記ターゲットエリアから到来する音に由来する非定常成分^φS (A)(ω,τ)とし、計算された^φS (B)(ω,τ)を上記インコヒーレントな雑音に由来する定常成分^φS (B)(ω,τ)とし、
上記第二成分抽出部は、αNを所定の実数とし、ΥNを特定区間のフレームのインデックスの集合とし、βN(ω)を所定の実数とし、以下の式により定義される^φN (A)(ω,τ)及び^φN (B)(ω,τ)を計算し、計算された^φN (A)(ω,τ)を上記干渉雑音に由来する非定常成分^φN (A)(ω,τ)とし、^φN (B)(ω,τ)を上記インコヒーレントな雑音に由来する定常成分^φN (B)(ω,τ)とし、
上記多様雑音対応型ゲイン計算部は、上記インコヒーレントな雑音に由来する定常成分^φN (B)(ω,τ)を更に用いて、上記ターゲットエリアから到来する音の非定常成分を強調するポストフィルタ~G(ω,τ)を計算する、
信号処理装置。 - 請求項1から3の何れかの信号処理装置であって、
上記ポストフィルタ~G(ω,τ)について時間方向と周波数方向との少なくとも一方の方向への平滑化処理を行う時間周波数平均化部と、
上記平滑化処理が行われたポストフィルタ~G(ω,τ)についてゲインシェーピングを行うゲインシェーピング部と、
を更に含む信号処理装置。 - マイクロホンアレーを構成するM個のマイクロホンで収音された信号から得られた周波数領域の観測信号に基づいて、ターゲットエリア及び上記ターゲットエリアと異なる少なくとも1個の雑音エリアのそれぞれの局所パワースペクトル密度を推定する局所PSD推定ステップと、
ωを周波数とし、τをフレームのインデックスとして、上記推定された局所パワースペクトル密度に基づいて、ターゲットエリアのパワースペクトル密度^φS(ω,τ)及び雑音エリアのパワースペクトル密度^φN(ω,τ)を推定するターゲットエリア/雑音エリアPSD推定部と、
上記ターゲットエリアのパワースペクトル密度^φS(ω,τ)から、ターゲットエリアから到来する音に由来する非定常成分^φS (A)(ω,τ)及びインコヒーレントな雑音に由来する定常成分^φS (B)(ω,τ)を抽出する第一成分抽出ステップと、
上記雑音のパワースペクトル密度^φN(ω,τ)から、干渉雑音に由来する非定常成分^φN (A)(ω,τ)を抽出する第二成分抽出ステップと、
上記ターゲットエリアから到来する音に由来する非定常成分^φS (A)(ω,τ)と、上記インコヒーレントな雑音に由来する定常成分^φS (B)(ω,τ)と、上記干渉雑音に由来する非定常成分^φN (A)(ω,τ)とを少なくとも用いて、上記ターゲットエリアから到来する音の非定常成分を強調するポストフィルタ~G(ω,τ)を計算する多様雑音対応型ゲイン計算ステップと、
を含む信号処理方法。 - 請求項1から4の何れかの信号処理装置の各部としてコンピュータを機能させるためのプログラム。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201580009993.1A CN106031196B (zh) | 2014-02-28 | 2015-02-25 | 信号处理装置、方法以及程序 |
JP2016505268A JP6225245B2 (ja) | 2014-02-28 | 2015-02-25 | 信号処理装置、方法及びプログラム |
US15/120,678 US9747921B2 (en) | 2014-02-28 | 2015-02-25 | Signal processing apparatus, method, and program |
EP15754624.3A EP3113508B1 (en) | 2014-02-28 | 2015-02-25 | Signal-processing device, method, and program |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014037820 | 2014-02-28 | ||
JP2014-037820 | 2014-02-28 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2015129760A1 true WO2015129760A1 (ja) | 2015-09-03 |
Family
ID=54009075
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/055442 WO2015129760A1 (ja) | 2014-02-28 | 2015-02-25 | 信号処理装置、方法及びプログラム |
Country Status (5)
Country | Link |
---|---|
US (1) | US9747921B2 (ja) |
EP (1) | EP3113508B1 (ja) |
JP (1) | JP6225245B2 (ja) |
CN (1) | CN106031196B (ja) |
WO (1) | WO2015129760A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019208137A1 (ja) * | 2018-04-24 | 2019-10-31 | 日本電信電話株式会社 | 音源分離装置、その方法、およびプログラム |
WO2022038673A1 (ja) * | 2020-08-18 | 2022-02-24 | 日本電信電話株式会社 | 収音装置、収音方法、プログラム |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10181329B2 (en) * | 2014-09-05 | 2019-01-15 | Intel IP Corporation | Audio processing circuit and method for reducing noise in an audio signal |
WO2017094862A1 (ja) * | 2015-12-02 | 2017-06-08 | 日本電信電話株式会社 | 空間相関行列推定装置、空間相関行列推定方法および空間相関行列推定プログラム |
JP6915579B2 (ja) * | 2018-04-06 | 2021-08-04 | 日本電信電話株式会社 | 信号分析装置、信号分析方法および信号分析プログラム |
CN109490626B (zh) * | 2018-12-03 | 2021-02-02 | 中车青岛四方机车车辆股份有限公司 | 一种基于非平稳随机振动信号的标准psd获取方法及装置 |
CN113808608B (zh) * | 2021-09-17 | 2023-07-25 | 随锐科技集团股份有限公司 | 一种基于时频掩蔽平滑策略的单声道噪声抑制方法和装置 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010211190A (ja) * | 2009-03-06 | 2010-09-24 | Harman Becker Automotive Systems Gmbh | バックグラウンドノイズ推定 |
WO2012026126A1 (ja) * | 2010-08-25 | 2012-03-01 | 旭化成株式会社 | 音源分離装置、音源分離方法、及び、プログラム |
JP2012088390A (ja) * | 2010-10-15 | 2012-05-10 | Honda Motor Co Ltd | 音声認識装置及び音声認識方法 |
JP2012177828A (ja) * | 2011-02-28 | 2012-09-13 | Pioneer Electronic Corp | ノイズ検出装置、ノイズ低減装置及びノイズ検出方法 |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4950733B2 (ja) * | 2007-03-30 | 2012-06-13 | 株式会社メガチップス | 信号処理装置 |
US8660281B2 (en) * | 2009-02-03 | 2014-02-25 | University Of Ottawa | Method and system for a multi-microphone noise reduction |
CN201418142Y (zh) * | 2009-05-22 | 2010-03-03 | 杨辉隆 | 一种麦克风 |
EP2395506B1 (en) * | 2010-06-09 | 2012-08-22 | Siemens Medical Instruments Pte. Ltd. | Method and acoustic signal processing system for interference and noise suppression in binaural microphone configurations |
JP5836616B2 (ja) * | 2011-03-16 | 2015-12-24 | キヤノン株式会社 | 音声信号処理装置 |
US9002027B2 (en) * | 2011-06-27 | 2015-04-07 | Gentex Corporation | Space-time noise reduction system for use in a vehicle and method of forming same |
EP2884491A1 (en) * | 2013-12-11 | 2015-06-17 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Extraction of reverberant sound using microphone arrays |
-
2015
- 2015-02-25 CN CN201580009993.1A patent/CN106031196B/zh active Active
- 2015-02-25 US US15/120,678 patent/US9747921B2/en active Active
- 2015-02-25 JP JP2016505268A patent/JP6225245B2/ja active Active
- 2015-02-25 WO PCT/JP2015/055442 patent/WO2015129760A1/ja active Application Filing
- 2015-02-25 EP EP15754624.3A patent/EP3113508B1/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2010211190A (ja) * | 2009-03-06 | 2010-09-24 | Harman Becker Automotive Systems Gmbh | バックグラウンドノイズ推定 |
WO2012026126A1 (ja) * | 2010-08-25 | 2012-03-01 | 旭化成株式会社 | 音源分離装置、音源分離方法、及び、プログラム |
JP2012088390A (ja) * | 2010-10-15 | 2012-05-10 | Honda Motor Co Ltd | 音声認識装置及び音声認識方法 |
JP2012177828A (ja) * | 2011-02-28 | 2012-09-13 | Pioneer Electronic Corp | ノイズ検出装置、ノイズ低減装置及びノイズ検出方法 |
Non-Patent Citations (1)
Title |
---|
KENTA NIWA ET AL.: "Implementation of microphone array for improving speech recognition rate in noisy environment", REPORT OF THE 2014 SPRING MEETING, 12 March 2014 (2014-03-12), pages 717 - 718, XP008184645 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019208137A1 (ja) * | 2018-04-24 | 2019-10-31 | 日本電信電話株式会社 | 音源分離装置、その方法、およびプログラム |
WO2022038673A1 (ja) * | 2020-08-18 | 2022-02-24 | 日本電信電話株式会社 | 収音装置、収音方法、プログラム |
Also Published As
Publication number | Publication date |
---|---|
EP3113508A1 (en) | 2017-01-04 |
JPWO2015129760A1 (ja) | 2017-03-30 |
JP6225245B2 (ja) | 2017-11-01 |
US20160372131A1 (en) | 2016-12-22 |
CN106031196B (zh) | 2018-12-07 |
EP3113508B1 (en) | 2020-11-11 |
EP3113508A4 (en) | 2017-11-01 |
US9747921B2 (en) | 2017-08-29 |
CN106031196A (zh) | 2016-10-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6225245B2 (ja) | 信号処理装置、方法及びプログラム | |
US10123113B2 (en) | Selective audio source enhancement | |
KR101934636B1 (ko) | 심화신경망 기반의 잡음 및 에코의 통합 제거 방법 및 장치 | |
JP6584930B2 (ja) | 情報処理装置、情報処理方法およびプログラム | |
WO2015196729A1 (zh) | 一种麦克风阵列语音增强方法及装置 | |
JP6780644B2 (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 | |
Niwa et al. | Post-filter design for speech enhancement in various noisy environments | |
KR101720514B1 (ko) | Dcica를 이용한 dnn 기반 특징향상을 수행하는 음성인식장치 및 방법 | |
JP6348427B2 (ja) | 雑音除去装置及び雑音除去プログラム | |
WO2016119388A1 (zh) | 一种基于语音信号构造聚焦协方差矩阵的方法及装置 | |
JPWO2016167141A1 (ja) | 信号処理装置、信号処理方法、およびプログラム | |
Martín-Doñas et al. | Dual-channel DNN-based speech enhancement for smartphones | |
KR20220022286A (ko) | 잔향 제거 오토 인코더를 이용한 잔향 환경 임베딩 추출 방법 및 장치 | |
JP2006178333A (ja) | 近接音分離収音方法、近接音分離収音装置、近接音分離収音プログラム、記録媒体 | |
Miyazaki et al. | Theoretical analysis of parametric blind spatial subtraction array and its application to speech recognition performance prediction | |
Ji et al. | Coherence-Based Dual-Channel Noise Reduction Algorithm in a Complex Noisy Environment. | |
JP6631127B2 (ja) | 音声判定装置、方法及びプログラム、並びに、音声処理装置 | |
JP6263890B2 (ja) | 音声信号処理装置及びプログラム | |
CN113870884B (zh) | 单麦克风噪声抑制方法和装置 | |
CN111863017B (zh) | 一种基于双麦克风阵列的车内定向拾音方法及相关装置 | |
KR101537653B1 (ko) | 주파수 또는 시간적 상관관계를 반영한 잡음 제거 방법 및 시스템 | |
JP2005091560A (ja) | 信号分離方法および信号分離装置 | |
JP6221463B2 (ja) | 音声信号処理装置及びプログラム | |
WO2021124537A1 (ja) | 情報処理装置、算出方法、及び算出プログラム | |
Koldovský et al. | Approximate MVDR and MMSE beamformers exploiting scale-invariant reconstruction of signals on microphones |
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: 15754624 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2016505268 Country of ref document: JP Kind code of ref document: A |
|
REEP | Request for entry into the european phase |
Ref document number: 2015754624 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2015754624 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15120678 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |