WO2012161555A2 - 방향성 마이크 어레이를 이용한 신호 분리시스템 및 그 제공방법 - Google Patents
방향성 마이크 어레이를 이용한 신호 분리시스템 및 그 제공방법 Download PDFInfo
- Publication number
- WO2012161555A2 WO2012161555A2 PCT/KR2012/004213 KR2012004213W WO2012161555A2 WO 2012161555 A2 WO2012161555 A2 WO 2012161555A2 KR 2012004213 W KR2012004213 W KR 2012004213W WO 2012161555 A2 WO2012161555 A2 WO 2012161555A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- signal
- microphone
- voice
- microphone array
- mixed
- Prior art date
Links
- 238000000926 separation method Methods 0.000 title claims abstract description 94
- 238000000034 method Methods 0.000 title claims abstract description 24
- 230000005236 sound signal Effects 0.000 claims abstract description 47
- 239000000203 mixture Substances 0.000 abstract 2
- 238000010586 diagram Methods 0.000 description 6
- 238000004088 simulation Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000012880 independent component analysis Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Images
Classifications
-
- 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
- G10L15/00—Speech recognition
- G10L15/20—Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech
-
- 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
- G10L21/0272—Voice signal separating
- G10L21/028—Voice signal separating using properties of sound source
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
-
- 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
Definitions
- the present invention relates to a system and a method for providing the same, which can effectively separate only a desired signal by using a microphone array including microphones installed in different directions, preferably in opposite directions.
- Voice-related systems receive a mix of desired signals and unwanted signals such as noise and reverberation from a microphone or microphone array. Accordingly, the voice related system has difficulty in performing a desired operation on the signal level of the unwanted signal included in the mixed signal or causes inconvenience to the user.
- the voice related system is a system for recognizing a specific voice command, it may cause a problem of low voice recognition rate.
- the voice-related system wants to record mainly the sound received in a specific direction (for example, the recording direction), such as a video camera or a camcorder, the unwanted sound (for example, the photographer) received in the other direction (for example, the rear) Noise, etc.) may be sounds that the user does not want to record or receive.
- various methods for separating the unwanted sound included in the mixed signal from the desired sound for example, whether the unwanted sound is a sound already known by the voice-related system (eg, LMS (Least Mean Square) BSS ( Blind Source Separation, Independent Component Analysis, etc.
- LMS Least Mean Square
- BSS Blind Source Separation, Independent Component Analysis
- the technical problem to be achieved by the present invention is to use a microphone array including directional (directional) microphones installed in different directions (for example, in the opposite direction) to easily mix only the sound received in a specific direction of the voice-related system in the mixed signal. It is to provide a system and a method of providing the same that can be separated.
- a voice related system provides a system and a method of providing the same that can be easily applied to an existing voice related system without the need for a complicated structural change or a separate resource such as hardware.
- a microphone array comprising at least one first microphone installed to face the first direction for achieving the technical problem and at least one second microphone installed to face a direction different from the first direction
- a signal receiver for receiving a mixed signal of a mixed first voice signal and a second voice signal, and receiving a reference signal through the at least one second microphone, and the second voice signal from the mixed signal received by the signal receiver
- a voice signal separator to cancel the voice signal separator, and cancels the second voice signal using the reference signal input through the at least one second microphone.
- the mixed signal may be a signal obtained by combining the signal received through the at least one first microphone or the signal received through the at least one first microphone with the signal received through the at least one second microphone. Can be.
- the microphone array may be connected to a predetermined voice output system and output the mixed signal and the reference signal to a signal separation system using the directional microphone array included in the voice output system.
- the signal separation system using the directional microphone array is included in a predetermined voice output system, and the second voice signal is an echo signal inputted through the microphone array. can do.
- the second direction may be opposite to the first direction.
- the signal separation unit uses a first sound source signal corresponding to the first sound signal as a first BSS sound source signal, a second sound source signal corresponding to the second sound signal as a second BSS sound source signal, and the at least one first microphone.
- BSS Breast Source Separation
- BSS may be performed using the signal inputted through the first BSS input signal and the reference signal as the second BSS input signal.
- the signal separation system using the directional microphone array is included in a predetermined voice storage system, and the first voice signal is output from a first sound source located in a first direction with respect to the voice storage system. It is a target voice signal to be stored in, the second voice signal is a signal that is output from a second sound source located in the opposite direction to the first direction with respect to the voice storage system to be removed from the mixed signal have.
- the signal separation system using the directional microphone array for solving the technical problem is at least one first microphone installed to face the first direction and at least one second microphone installed to face the direction opposite to the first direction
- the microphone array includes a microphone array comprising a mixed signal of the first audio signal and the second audio signal mixed through the microphone array and the reference signal received through the at least one second microphone, respectively;
- the signal separation system may cancel the second audio signal from the mixed signal based on the reference signal.
- the signal separation system using the directional microphone array for solving the technical problem is at least one first microphone installed to face the first direction and at least one second microphone installed to face a direction different from the first direction
- a microphone array including a signal receiving unit for receiving a mixed signal mixed with a first voice signal and a second voice signal through the microphone array, and receiving a reference signal through the at least one second microphone, and the signal And a voice signal separator for canceling the second voice signal from the mixed signal received by the receiver, wherein the voice signal separator comprises the second voice using the reference signal input through the at least one second microphone. The signal may be canceled out.
- Method for providing a signal separation system using a directional microphone array for solving the technical problem is at least one first microphone and a direction different from the first direction is installed so that the signal separation system using the directional microphone array toward the first direction
- the mixed signal received by the signal separation system using an array Standing step of canceling the second audio signal can be characterized in that to compensate for the second audio signal with the reference signal.
- the mixed signal may be a signal obtained by combining the signal received through the at least one first microphone or the signal received through the at least one first microphone with the signal received through the at least one second microphone. Can be.
- the generating of the input signal by removing the second audio signal from the mixed signal received by the signal separation system using the directional microphone array may include: converting the first sound source signal corresponding to the first audio signal into a first BSS sound source signal;
- the second sound source signal corresponding to the second audio signal is a second BSS sound source signal, and the signal input through the at least one first microphone is a first BSS input signal and the reference signal is a second BSS input signal. It may include performing blind source separation.
- the method of providing a signal separation system using the directional microphone array may be stored in a computer readable recording medium recording a program.
- the signal separation system using the directional microphone array according to the present invention can easily provide a signal separation function using the microphone array according to an embodiment of the present invention even in a voice related system that does not have a function of separating a desired signal from a mixed signal. It has an effect.
- an output signal for canceling a signal already known by the voice related system eg, an echo signal of an output signal outputted by the voice related system
- a hardware configuration for receiving the output signal in which a configuration for performing signal separation for example, an echo canceller, etc.
- the unwanted signal is a newly input signal that the voice-related system does not know
- only the sound received in a specific direction can be separated from the mixed signal, so that it is easily applied when the desired sound is mainly received in a specific direction. It can work.
- FIG. 1 is a view for explaining the concept of a microphone array according to an embodiment of the present invention.
- FIG. 2 is a diagram illustrating a schematic configuration of a signal separation system using a directional microphone array according to an embodiment of the present invention.
- FIG. 3 is a diagram illustrating beam patterns for explaining a concept of a signal separation system using a directional microphone array according to an embodiment of the present invention.
- FIG. 4 is a diagram illustrating a voice related system to which a signal separation system using a directional microphone array according to an embodiment of the present invention is applied.
- FIG. 5 is a diagram illustrating a simulation result of a beam pattern formed by a signal separation system providing method using a directional microphone array according to an embodiment of the present invention.
- the component when one component 'transmits' data to another component, the component may directly transmit the data to the other component, or through at least one other component. Means that the data may be transmitted to the other component.
- FIG. 1 is a view for explaining the concept of a microphone array according to an embodiment of the present invention.
- the microphone array 200 includes at least one first microphone (eg, 210, 211, 212, and 213) installed to face the first direction.
- the microphone array 200 includes at least one second microphone (eg, 220) installed to face a direction different from the first direction (eg, an opposite direction).
- the microphone array 200 includes a case in which the second microphone (eg, 220) is included.
- a plurality of microphones are installed to face a direction different from the first direction (eg, an opposite direction). May be included.
- the second microphone (eg, 220) may be located between a plurality of first microphones (eg, 210, 211, 212, and 213), and the second microphones (eg, 220) are adjacent to each other. It may be located to.
- the microphone array 200 may include a predetermined housing for fixing the first microphone (eg, 210, 211, 212, 213) and the second microphone (eg, 220).
- the first microphones (eg, 210, 211, 212, 213) and the second microphones (eg, 220) may be installed adjacent to each other, or may be spaced apart at predetermined intervals. In the case of spaced apart installation, the signal delay due to the spaced distance may be considered in the signal separation process.
- each of the first microphones (eg, 210, 211, 212, and 213) and the second microphones (eg, 220) may be implemented as a directional (directional) microphone or a cardioid microphone.
- the directional microphone may be a microphone that forms a cardioid beam pattern. Therefore, in this specification, that a predetermined microphone is installed to face a specific direction, which may mean that the beam pattern formed by the microphone is installed to face the specific direction.
- the specific direction may mean a direction (for example, a front direction or a rear direction) set based on the longitudinal section of the microphone array 200.
- the fact that the plurality of microphones are installed so as to face the first or second direction does not mean that each of the plurality of microphones is installed to face a common predetermined point, but rather a common direction (eg, front or rear). It may mean that is installed to face.
- the microphone array 200 includes one first microphone 210 and one second microphone (eg, 220). The case may be described as follows.
- the first microphone 210 may be installed to face the front surface of the microphone array 200 based on the longitudinal cross section of the microphone array 200.
- the second microphone eg, 220
- the second microphone may have a second microphone (eg, 220) opposite to a direction in which the first microphone 210 is installed, that is, the rear surface of the second microphone (eg, 220) based on a longitudinal section of the microphone array 200.
- Signals received from each of the first microphone 210 and the second microphone (eg, 220) may be transmitted to another device or system through predetermined signal transmission means (eg, jack, signal line, etc. 230, 231). Can be.
- the beam pattern formed by the microphone array 200 may be the same as the beam pattern 40 illustrated in FIG. 3B.
- a mixed signal in which a desired signal, i.e., a first audio signal and an undesired signal second audio signal, are mixed from the front and rear surfaces of the microphone array 200, respectively. can be entered.
- an unwanted second voice signal may be mixed in the mixed signal received through the first microphone 210, and the signal received through the second microphone (eg, 220), that is, a reference signal. Also, unwanted second audio signals may be mixed. However, in the reference signal received through the second microphone (eg, 220), the signal level of the second voice signal included in the second microphone is included in the mixed signal received through the first microphone 210. It may be higher than the signal level of.
- a sound source signal corresponding to a desired signal that is, the first audio signal
- a sound source signal corresponding to an unwanted signal that is, the second audio signal
- S1 a sound source signal corresponding to an unwanted signal
- the signal x1 (t) input through the first microphone 210 and the signal x2 (t) input through the second microphone may be expressed as Equation 1.
- the second microphone 210 when the first sound source exists in the direction that the first microphone 210 faces or most of the first sound signal output from the first sound source is input through the first microphone 210, the second It may be assumed that the input through the microphone (eg, 220) is weak. That is, the signal level of the first audio signal included in the signal received through the first microphone 210 is the signal of the first audio signal included in the signal received through the second microphone (eg, 220). It can be higher than the level.
- the unwanted signal ie, the second voice signal
- the second microphone eg, 220
- the signal level of the second audio signal included in the signal received through the first microphone 210 is the signal level of the second audio signal included in the signal received through the second microphone (eg, 220). It may be lower than the signal level.
- the microphone array 200 may be used to simply cancel the second audio signal from the mixed signal to separate the desired first audio signal.
- the first audio signal may be mainly input in the first direction
- the unwanted second voice signal may be mainly input in the second direction. This case will be described with reference to FIG. 4.
- FIG. 4 is a diagram illustrating a voice related system to which a signal separation system using a directional microphone array according to an embodiment of the present invention is applied.
- a predetermined voice output system (eg, IPTV, set-top box, telephone, computer, etc.) 300 may exist.
- the voice output system 300 may output voice by itself.
- the voice output system 300 may include a predetermined voice output device (eg, a speaker 310).
- the voice output system 300 may receive a voice signal from the outside.
- the microphone array 200 according to the embodiment of the present invention may be connected to the voice output system 300.
- the microphone array 200 may be installed at a predetermined position (eg, the upper end) of the voice output system 300 as shown in FIG. 4A, but is not limited thereto.
- the voice output device 310 provided in the voice output system 300 may be installed to face a user direction using the voice output system 300, that is, a first direction (eg, the front direction of FIG. 4A). have.
- the voice output device 310 may be installed on the rear surface of the voice output system 300 according to the side or the implementation example.
- a signal (eg, a voice command, a call voice, etc.) desired by the voice output system 300 may be output from the user.
- the user may be located in the first direction.
- the desired signal that is, the first audio signal
- the first microphone 210 mainly installed to face the first direction.
- the echo signal generated by the second voice signal that is, the signal output by the voice output device 310 may be mainly received through the second microphone (eg, 220).
- FIG. 4B is a view showing the side of the voice output system 300 as shown in FIG. 4A.
- the voice output device 310 of the voice output system 300 is shown in FIG.
- the echo signal generated by the output signal is received through the second microphone (eg, 220) mainly installed in the opposite direction, rather than the first microphone 210 installed to face the first direction.
- the second microphone eg, 220
- an echo signal received through the second microphone eg, 220
- the signal level of the second voice signal received through the first microphone 210 may be higher.
- the mixed signal effectively prevents the The first audio signal can be separated.
- the output signal output through the voice output device 310 is used as a reference signal. It was. That is, the methods for canceling a previous application or various conventional echo signals store an output signal output through the voice output device 310 and cancel the echo signal by using the same. For example, the echo signal is estimated by using the stored output signal through channel estimation and gain factor calculation, and the echo signal is canceled from the mixed signal.
- the output signal output from the voice output system 300 is not used as a reference signal, but is actually received through the second microphone (eg, 220) by applying a channel and a gain factor. Since the used signal is used as a reference signal, the calculation for the signal separation process can be performed quickly and efficiently.
- the voice output system 300 may perform hardware and / or software separation of signals such as storage means for storing the output signal and means for transmitting the stored signal to a signal separation device (eco canceller). There is a problem that the resource is implemented in advance or the structure that is implemented must be changed.
- the microphone array 200 is connected to the voice input terminal of the voice output system 300, and only by installing a signal separation system for signal separation, that is, a predetermined software or application.
- the signal separation system may be provided integrally with the microphone array 200 without being installed in the voice output system 300.
- the microphone output 200 is connected to the voice output system 300 by simply connecting the signal separation system using the directional microphone array according to an embodiment of the present invention to the voice output system 300. In this case, only the desired signal from which the second voice signal is canceled may be received as an input.
- the signal separation system itself using the directional microphone array according to an embodiment of the present invention may have to have a processing device having a predetermined computing power.
- the signal separation system using the directional microphone array according to an embodiment of the present invention may be installed embedded in the production of the voice output system 300.
- the unwanted signal may not be an echo signal as described with reference to FIGS. 4A and 4B. This case will be described with reference to FIG. 4C.
- the voice receiving system 400 is used to include all systems capable of receiving and / or storing voice.
- the voice receiving system 400 may further receive a video signal like a camcorder according to an embodiment.
- the microphone array 200 may be connected to the voice receiving system 400 or may be installed in advance.
- the first microphone 210 included in the microphone array 200 may be installed or manipulated so as to face a direction corresponding to the first voice signal that the voice receiving system 400 desires to receive voice.
- the first microphone 210 included in the microphone array 200 may be installed in a lens direction, that is, in a first direction. Then, the first microphone 210 may face the direction of the object that the user (photographer) wants to photograph, that is, the first direction.
- the object may output a first sound source signal.
- the first audio signal based on the first sound source signal may be mainly received through the first microphone 210.
- a signal that the voice receiving system 400 does not want to receive and / or store that is, a second voice signal may be mainly input from the second direction.
- the second voice signal may be noise caused by a user, unnecessary sound, or the like. That is, since the voice receiving system 400 may be manipulated to face an object to which voice is to be received, various sounds mainly received from opposite directions may be unwanted sounds. Therefore, the technical idea of the present invention can be usefully applied to the voice receiving system 400 as well. That is, the first voice signal may be separated by canceling the second voice signal from the mixed signal received through the microphone array 200. The separated signal may be stored in the voice receiving system 400 or transmitted to another system.
- FIG. 3 A schematic configuration of a signal separation system using a directional microphone array according to an embodiment of the present invention for implementing the technical idea is shown in FIG.
- a beam pattern that can be generated by the signal separation system using the directional microphone array shown in FIG. 2 is schematically illustrated in FIG. 3.
- the signal separation system 1 using the directional microphone array may include a signal separation system 100.
- the microphone array 200 may be included.
- the signal separation system 100 may receive a mixed signal from the microphone array 200.
- the signal separation system 100 may receive a reference signal.
- the reference signal may be received through at least one second microphone (eg, 220).
- the mixed signal may be received through the at least one first microphone (eg, 210).
- the signal separation system 100 may perform a function of canceling the reference signal from the mixed signal.
- the signal separation system 100 may include a signal receiver 110 and a signal separator 120.
- the signal receiver 110 may receive a mixed signal and a reference signal from the microphone array 200 and output the mixed signal and the reference signal to the signal separator 120.
- the signal separator 120 may cancel the reference signal from the mixed signal.
- the signal splitter 120 may cancel the reference signal from the mixed signal in various ways. That is, all technical ideas (eg, independent component analysis, principal component analysis, signal suppresion, etc.) capable of canceling any other known signal (reference signal) from any one known signal (mixed signal) may be applied. .
- the signal separator 120 may cancel the reference signal from the mixed signal using a BSS algorithm.
- the signal separation system 100 may separate the reference signal from the mixed signal more efficiently than the conventional BSS algorithm. Because the conventional BSS algorithm separates n unknown signals when n unknown signals are received through n microphones, according to an embodiment of the present invention, one signal to be separated, that is, the second This is because the reference signal received through the microphone (eg, 220) is already known. The technical idea that can be applied at this time has been disclosed in detail in the previous application as described above.
- first sound source S1 corresponding to a desired signal
- second sound source S2 corresponding to an undesired second voice signal.
- the sound source signal of the first sound source (S1) The sound source signal of the second sound source (S2) It can be said.
- the signal received through the first microphone 210 ie, a mixed signal
- the signal received through the second microphone eg, 220
- the reference signal Can be set.
- the first sound source signal is a first BSS sound source signal
- the second sound source signal corresponding to the second sound signal is a second BSS sound source signal
- the signal input through the at least one first microphone is a first BSS input signal.
- the reference signal may be set as a second BSS input signal to perform a BSS algorithm in a manner similar to the previous application.
- the gain factor matrix of Equation 12 of the previous application In a21 is 0, and a11 and a22 are set to 1, according to the assumption according to the technical concept of the present invention, if the calculation amount is sufficient, only a11 and a22 may be set to 1 and the calculation may be performed. According to another embodiment, it may be possible to set a11 and a22 to 1 and set a12 to 0, or set a11 and a22 to 1 and a21 to 0.
- signal suppression may be performed using each matrix result value to increase the cancellation rate of the S2 signal component.
- the mixed signal received by the signal receiver 110 is different from the signal received through the first microphone 210 and the second microphone 220, unlike FIG. 2A.
- the received signal may be a mixed signal, that is, a signal received by the microphone array 200 as a whole.
- a predetermined mixing means 240 may be further provided.
- the mixing means 240 may be included in the microphone array 200, or may be included in the signal receiver 110.
- the mixing means 240 may be implemented in a simple hardware structure (eg, connection of signal lines). As such, in the signal in which the signal received through the first microphone 210 and the signal received through the second microphone (eg, 220) are mixed, the signal received through the second microphone (eg, 220) may be used. Even when offsetting, the technical idea according to the embodiment of the present invention may exhibit good performance. This will be described using a beam pattern as follows.
- FIG. 3A illustrates a case in which a signal received through the first microphone 210 is used as a mixed signal as shown in FIG. 2A.
- the beam pattern 10 formed by the first microphone 210 is illustrated in FIG. 3A.
- the beam pattern 20 formed by the second microphone eg, 220
- the beam pattern 30 may be formed in a desired direction (first direction).
- the beam patterns formed on the first microphone 210 and the second microphone (eg, 220) may be beam patterns 10 and beam patterns 20, respectively. Therefore, when the two beam patterns 10 and 20 are combined, the beam pattern 40 illustrated in FIG. 3B may be formed.
- the signal separation system 1 using the directional microphone array according to the embodiment of the present invention may use the microphone array 200 according to the embodiment of the present invention to form the beam pattern 30 in the desired direction, that is, the first direction. 50) can be easily formed. Therefore, according to the type and environment of the voice related system to which the signal separation system 1 using the directional microphone array according to the embodiment of the present invention is applied, the embodiment as shown in FIGS. 2A and 2B may be selectively applied.
- the formation of the beam pattern as shown in FIG. 3 merely illustrates the formation of a theoretical or conceptual beam pattern, and the shape of the beam pattern actually formed may vary somewhat depending on the environment.
- FIG. 5 is a view showing a simulation result of the beam pattern formed by the method for providing a signal separation system using a directional microphone array according to an embodiment of the present invention
- Figures 5a to 5d is a direction according to an embodiment of the present invention in the tuning fork anechoic chamber
- FIG. 5 is a diagram showing a polar pattern (frequency pattern) for each frequency of the simulation results while rotating at intervals of 15 degrees from 0 to 360 degrees.
- Each of the voice signals shows simulation results of beam patterns formed in the 500 Hz, 1 KHz, 2.5 KHz, and 4 KHz bands.
- the y axis represents the signal level db.
- a signal separated by the signal separation system 1 using the directional microphone array in various frequency bands is a signal in which the signal level of the signal received in the first direction is received in the second direction. It can be seen that it is much higher than the signal level of. That is, it can be seen that a high performance that cannot be seen in the prior art can be obtained by the sensitivity of 40 dB or more lower than the front side.
- the microphone array 200 and the signal separation system 100 of the signal separation system 1 using the directional microphone array may be integrally implemented in a predetermined housing. Then, the signal separation system 1 using the directional microphone array may output the signal separated by the signal separation system 100 to the predetermined voice output system 300 or the voice reception system 400.
- the signal separation system 1 using the directional microphone array may further include a predetermined data processing unit.
- the microphone array 200 is connected to a predetermined voice output system 300 or a voice receiving system 400 through a jack, etc.
- the signal separation system 100 is the predetermined voice output system It may be included in the 300 or the voice receiving system 400 is installed.
- the mixed signal and the reference signal output through the microphone array 200 may be transmitted to the signal separation system 100 directly or through a predetermined path.
- the signal separation system 100 may be implemented in a predetermined software to implement the technical idea of the present invention by combining the hardware organically provided in the predetermined voice output system 300 or the voice receiving system 400.
- the signal separation system providing method using the directional microphone array can be implemented as computer readable codes on a computer readable recording medium.
- Computer-readable recording media include all kinds of recording devices that store data that can be read by a computer system. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, hard disk, floppy disk, optical data storage, and the like, as well as carrier wave (e.g., transmission over the Internet). It also includes implementations.
- the computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. And functional programs, codes and code segments for implementing the present invention can be easily inferred by programmers in the art to which the present invention belongs.
- the present invention can be applied to various systems in which it is necessary to separate desired and unwanted voice signals.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- General Health & Medical Sciences (AREA)
- Otolaryngology (AREA)
- Circuit For Audible Band Transducer (AREA)
- Obtaining Desirable Characteristics In Audible-Bandwidth Transducers (AREA)
Abstract
Description
Claims (13)
- 제1방향을 향하도록 설치되는 적어도 하나의 제1마이크 및 상기 제1방향과는 다른 방향을 향하도록 설치되는 적어도 하나의 제2마이크를 포함하는 마이크 어레이를 이용하여 제1음성신호와 제2음성신호가 혼합된 혼합신호를 수신하고, 상기 적어도 하나의 제2마이크를 통해 레퍼런스 신호를 수신하기 위한 신호수신부;상기 신호수신부에 의해 수신된 혼합신호에서 상기 제2음성신호를 상쇄하기 위한 음성신호 분리부를 포함하며,상기 음성신호 분리부는 상기 적어도 하나의 제2마이크를 통해 입력된 상기 레퍼런스 신호를 이용하여 상기 제2음성신호를 상쇄하는 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템.
- 제1항에 있어서, 상기 혼합신호는,상기 적어도 하나의 제1마이크를 통해 수신된 신호 또는 상기 적어도 하나의 제1마이크를 통해 수신된 신호와 상기 적어도 하나의 제2마이크를 통해 수신된 신호가 합쳐진 신호인 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템.
- 제 1항에 있어서, 상기 마이크 어레이는,소정의 음성출력 시스템과 연결되어 상기 혼합신호 및 상기 레퍼런스 신호를 상기 음성출력 시스템에 포함된 상기 방향성 마이크 어레이를 이용한 신호 분리시스템으로 출력하는 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템.
- 제 1항에 있어서, 상기 방향성 마이크 어레이를 이용한 신호 분리시스템은,소정의 음성출력 시스템에 포함되어 설치되며,상기 제2음성신호는,상기 음성출력 시스템으로부터 출력된 음성신호가 상기 마이크 어레이를 통해 입력된 에코신호인 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템.
- 제 1항에 있어서, 상기 제2방향은,상기 제1방향과 반대방향인 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템.
- 제 1항에 있어서, 상기 신호 분리부는,상기 제1음성신호에 대응되는 제1음원신호를 제1BSS 음원신호, 상기 제2음성신호에 대응되는 제2음원신호를 제2BSS 음원신호로 하고,상기 적어도 하나의 제1마이크를 통해 입력된 신호를 제1BSS 입력신호, 상기 레퍼런스 신호를 제2BSS 입력신호로 하여 BSS(Blind Source Separation)을 수행하는 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템.
- 제 1항에 있어서, 상기 방향성 마이크 어레이를 이용한 신호 분리시스템은,소정의 음성저장 시스템에 포함되어 설치되며,상기 제1음성신호는 상기 음성저장 시스템을 기준으로 제1방향에 위치하는 제1음원으로부터 출력되어 상기 음성기록 시스템에 저장될 타겟 음성신호이며,상기 제2음성신호는 상기 음성저장 시스템을 기준으로 상기 제1방향과 반대방향에 위치하는 제2음원으로부터 출력되어 상기 혼합신호에서 제거될 신호인 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템.
- 제1방향을 향하도록 설치되는 적어도 하나의 제1마이크; 및상기 제1방향과는 반대 방향을 향하도록 설치되는 적어도 하나의 제2마이크를 포함하는 마이크 어레이를 포함하며,상기 마이크 어레이는,상기 마이크 어레이를 통해 제1음성신호와 제2음성신호가 혼합된 혼합신호 및 상기 적어도 하나의 제2마이크를 통해 수신되는 레퍼런스 신호를 각각 소정의 신호 분리시스템으로 출력하면, 상기 레퍼런스 신호에 기초하여 상기 신호 분리시스템에 의해 상기 혼합신호에서 상기 제2음성신호를 상쇄하는 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템.
- 제1방향을 향하도록 설치되는 적어도 하나의 제1마이크 및 상기 제1방향과는 다른 방향을 향하도록 설치되는 적어도 하나의 제2마이크를 포함하는 마이크 어레이;상기 마이크 어레이를 통해 제1음성신호와 제2음성신호가 혼합된 혼합신호를 수신하고, 상기 적어도 하나의 제2마이크를 통해 레퍼런스 신호를 수신하기 위한 신호수신부; 및상기 신호수신부에 의해 수신된 혼합신호에서 상기 제2음성신호를 상쇄하기 위한 음성신호 분리부를 포함하며,상기 음성신호 분리부는 상기 적어도 하나의 제2마이크를 통해 입력된 상기 레퍼런스 신호를 이용하여 상기 제2음성신호를 상쇄하는 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템.
- 방향성 마이크 어레이를 이용한 신호 분리시스템이 제1방향을 향하도록 설치되는 적어도 하나의 제1마이크 및 상기 제1방향과는 다른 방향을 향하도록 설치된 적어도 하나의 제2마이크를 포함하는 마이크 어레이를 통해 제1음성신호와 제2음성신호가 혼합된 혼합신호를 수신하는 단계;상기 방향성 마이크 어레이를 이용한 신호 분리시스템이 상기 마이크 어레이에 포함된 상기 적어도 하나의 제2마이크를 통해 레퍼런스 신호를 수신하는 단계;상기 방향성 마이크 어레이를 이용한 신호 분리시스템이 수신된 상기 혼합신호에서 상기 제2음성신호를 상쇄하는 단계를 포함하며,상기 방향성 마이크 어레이를 이용한 신호 분리시스템이 수신된 상기 혼합신호에서 상기 제2음성신호를 상쇄하는 단계는,상기 레퍼런스 신호를 이용하여 상기 제2음성신호를 상쇄하는 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템 제공방법.
- 제10항에 있어서, 상기 혼합신호는,상기 적어도 하나의 제1마이크를 통해 수신된 신호 또는 상기 적어도 하나의 제1마이크를 통해 수신된 신호와 상기 적어도 하나의 제2마이크를 통해 수신된 신호가 합쳐진 신호인 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템 제공방법.
- 제 10항에 있어서, 상기 방향성 마이크 어레이를 이용한 신호 분리시스템이 수신된 상기 혼합신호에서 상기 제2음성신호를 제거하여 입력신호를 생성하는 단계는,상기 제1음성신호에 대응되는 제1음원신호를 제1BSS 음원신호, 상기 제2음성신호에 대응되는 제2음원신호를 제2BSS 음원신호로 하고, 상기 적어도 하나의 제1마이크를 통해 입력된 신호를 제1BSS 입력신호, 상기 레퍼런스 신호를 제2BSS 입력신호로 하여 BSS(Blind Source Separation)을 수행하는 단계를 포함하는 것을 특징으로 하는 방향성 마이크 어레이를 이용한 신호 분리시스템 제공방법.
- 제 10항 내지 제12항 중 어느 한 항에 기재된 방법을 수행하기 위한 프로그램을 기록한 컴퓨터 판독가능한 기록매체.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014512772A JP2014518053A (ja) | 2011-05-26 | 2012-05-29 | 指向性マイクアレイを用いた信号分離システム及びその提供方法 |
US14/119,982 US9516411B2 (en) | 2011-05-26 | 2012-05-29 | Signal-separation system using a directional microphone array and method for providing same |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020110050266A KR101248971B1 (ko) | 2011-05-26 | 2011-05-26 | 방향성 마이크 어레이를 이용한 신호 분리시스템 및 그 제공방법 |
KR10-2011-0050266 | 2011-05-26 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2012161555A2 true WO2012161555A2 (ko) | 2012-11-29 |
WO2012161555A3 WO2012161555A3 (ko) | 2013-01-24 |
Family
ID=47217932
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2012/004213 WO2012161555A2 (ko) | 2011-05-26 | 2012-05-29 | 방향성 마이크 어레이를 이용한 신호 분리시스템 및 그 제공방법 |
Country Status (4)
Country | Link |
---|---|
US (1) | US9516411B2 (ko) |
JP (1) | JP2014518053A (ko) |
KR (1) | KR101248971B1 (ko) |
WO (1) | WO2012161555A2 (ko) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2840571A3 (en) * | 2013-08-23 | 2015-03-25 | Samsung Electronics Co., Ltd | Display apparatus and control method thereof |
US20230395095A1 (en) * | 2013-02-25 | 2023-12-07 | Amazon Technologies, Inc. | Direction based end-pointing for speech recognition |
Families Citing this family (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CA2581982C (en) | 2004-09-27 | 2013-06-18 | Nielsen Media Research, Inc. | Methods and apparatus for using location information to manage spillover in an audience monitoring system |
US8855101B2 (en) | 2010-03-09 | 2014-10-07 | The Nielsen Company (Us), Llc | Methods, systems, and apparatus to synchronize actions of audio source monitors |
US8885842B2 (en) | 2010-12-14 | 2014-11-11 | The Nielsen Company (Us), Llc | Methods and apparatus to determine locations of audience members |
US9021516B2 (en) | 2013-03-01 | 2015-04-28 | The Nielsen Company (Us), Llc | Methods and systems for reducing spillover by measuring a crest factor |
US9118960B2 (en) | 2013-03-08 | 2015-08-25 | The Nielsen Company (Us), Llc | Methods and systems for reducing spillover by detecting signal distortion |
US9219969B2 (en) | 2013-03-13 | 2015-12-22 | The Nielsen Company (Us), Llc | Methods and systems for reducing spillover by analyzing sound pressure levels |
US9191704B2 (en) | 2013-03-14 | 2015-11-17 | The Nielsen Company (Us), Llc | Methods and systems for reducing crediting errors due to spillover using audio codes and/or signatures |
US9197930B2 (en) | 2013-03-15 | 2015-11-24 | The Nielsen Company (Us), Llc | Methods and apparatus to detect spillover in an audience monitoring system |
US20140379421A1 (en) | 2013-06-25 | 2014-12-25 | The Nielsen Company (Us), Llc | Methods and apparatus to characterize households with media meter data |
US9924224B2 (en) | 2015-04-03 | 2018-03-20 | The Nielsen Company (Us), Llc | Methods and apparatus to determine a state of a media presentation device |
US9554207B2 (en) * | 2015-04-30 | 2017-01-24 | Shure Acquisition Holdings, Inc. | Offset cartridge microphones |
US9848222B2 (en) | 2015-07-15 | 2017-12-19 | The Nielsen Company (Us), Llc | Methods and apparatus to detect spillover |
WO2017056288A1 (ja) * | 2015-10-01 | 2017-04-06 | 三菱電機株式会社 | 音響信号処理装置、音響処理方法、監視装置および監視方法 |
US9747920B2 (en) * | 2015-12-17 | 2017-08-29 | Amazon Technologies, Inc. | Adaptive beamforming to create reference channels |
JP7020799B2 (ja) * | 2017-05-16 | 2022-02-16 | ソニーグループ株式会社 | 情報処理装置、及び情報処理方法 |
US10522167B1 (en) * | 2018-02-13 | 2019-12-31 | Amazon Techonlogies, Inc. | Multichannel noise cancellation using deep neural network masking |
CN112102825B (zh) * | 2020-08-11 | 2021-11-26 | 湖北亿咖通科技有限公司 | 基于车机语音识别的音频处理方法、装置和计算机设备 |
CN112017681B (zh) * | 2020-09-07 | 2022-05-13 | 思必驰科技股份有限公司 | 定向语音的增强方法及系统 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004053839A1 (en) * | 2002-12-11 | 2004-06-24 | Softmax, Inc. | System and method for speech processing using independent component analysis under stability constraints |
US20060222184A1 (en) * | 2004-09-23 | 2006-10-05 | Markus Buck | Multi-channel adaptive speech signal processing system with noise reduction |
KR20090037692A (ko) * | 2007-10-12 | 2009-04-16 | 삼성전자주식회사 | 혼합 사운드로부터 목표 음원 신호를 추출하는 방법 및장치 |
KR20100068188A (ko) * | 2008-12-12 | 2010-06-22 | 신호준 | 신호 분리 방법, 상기 신호 분리 방법을 이용한 통신 시스템 및 음성인식시스템 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06292293A (ja) * | 1993-03-31 | 1994-10-18 | Sony Corp | マイクロホン装置 |
JPH0936940A (ja) * | 1995-07-14 | 1997-02-07 | Hitachi Ltd | 音声入力装置 |
WO2003013185A1 (en) * | 2001-08-01 | 2003-02-13 | Dashen Fan | Cardioid beam with a desired null based acoustic devices, systems and methods |
JP4138449B2 (ja) * | 2002-09-24 | 2008-08-27 | 株式会社ディーアンドエムホールディングス | 音声入力システム及び通信システム |
EP1830348B1 (en) * | 2006-03-01 | 2016-09-28 | Nuance Communications, Inc. | Hands-free system for speech signal acquisition |
US8238569B2 (en) | 2007-10-12 | 2012-08-07 | Samsung Electronics Co., Ltd. | Method, medium, and apparatus for extracting target sound from mixed sound |
KR101516589B1 (ko) * | 2008-03-25 | 2015-05-06 | 에스케이텔레콤 주식회사 | 이동통신단말기 및 그의 음성신호 처리 방법 |
KR101340520B1 (ko) * | 2008-07-22 | 2013-12-11 | 삼성전자주식회사 | 잡음을 제거하는 장치 및 방법 |
-
2011
- 2011-05-26 KR KR1020110050266A patent/KR101248971B1/ko active IP Right Grant
-
2012
- 2012-05-29 WO PCT/KR2012/004213 patent/WO2012161555A2/ko active Application Filing
- 2012-05-29 US US14/119,982 patent/US9516411B2/en active Active
- 2012-05-29 JP JP2014512772A patent/JP2014518053A/ja active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2004053839A1 (en) * | 2002-12-11 | 2004-06-24 | Softmax, Inc. | System and method for speech processing using independent component analysis under stability constraints |
US20060222184A1 (en) * | 2004-09-23 | 2006-10-05 | Markus Buck | Multi-channel adaptive speech signal processing system with noise reduction |
KR20090037692A (ko) * | 2007-10-12 | 2009-04-16 | 삼성전자주식회사 | 혼합 사운드로부터 목표 음원 신호를 추출하는 방법 및장치 |
KR20100068188A (ko) * | 2008-12-12 | 2010-06-22 | 신호준 | 신호 분리 방법, 상기 신호 분리 방법을 이용한 통신 시스템 및 음성인식시스템 |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20230395095A1 (en) * | 2013-02-25 | 2023-12-07 | Amazon Technologies, Inc. | Direction based end-pointing for speech recognition |
US11978478B2 (en) * | 2013-02-25 | 2024-05-07 | Amazon Technologies, Inc. | Direction based end-pointing for speech recognition |
EP2840571A3 (en) * | 2013-08-23 | 2015-03-25 | Samsung Electronics Co., Ltd | Display apparatus and control method thereof |
US9402094B2 (en) | 2013-08-23 | 2016-07-26 | Samsung Electronics Co., Ltd. | Display apparatus and control method thereof, based on voice commands |
Also Published As
Publication number | Publication date |
---|---|
KR20120131826A (ko) | 2012-12-05 |
WO2012161555A3 (ko) | 2013-01-24 |
KR101248971B1 (ko) | 2013-04-09 |
US9516411B2 (en) | 2016-12-06 |
JP2014518053A (ja) | 2014-07-24 |
US20140126746A1 (en) | 2014-05-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2012161555A2 (ko) | 방향성 마이크 어레이를 이용한 신호 분리시스템 및 그 제공방법 | |
US6868045B1 (en) | Voice control system with a microphone array | |
WO2018008885A1 (ko) | 영상처리장치, 영상처리장치의 구동방법 및 컴퓨터 판독가능 기록매체 | |
CN100370830C (zh) | 用于音频/图像的说话者检测和定位的方法和装置 | |
WO2017052056A1 (en) | Electronic device and method of audio processing thereof | |
WO2014196769A1 (ko) | 음성 향상 방법 및 그 장치 | |
WO2010067976A2 (ko) | 신호 분리 방법, 상기 신호 분리 방법을 이용한 통신 시스템 및 음성인식시스템 | |
WO2017026568A1 (ko) | 음질 개선을 위한 방법 및 헤드셋 | |
JPH10282993A (ja) | 機器の音声作動式遠隔制御システム | |
DK159356B (da) | Hoereapparat | |
WO2016056683A1 (ko) | 전자 장치 및 이의 잔향 제거 방법 | |
US5982906A (en) | Noise suppressing transmitter and noise suppressing method | |
WO2019156339A1 (ko) | 오디오 신호의 주파수의 변화에 따른 위상 변화율에 기반하여 노이즈가 감쇠된 오디오 신호를 생성하는 장치 및 방법 | |
WO2019216579A1 (ko) | 스피커 모듈을 이용한 발수 구조를 가진 웨어러블 전자 장치 및 그의 수분 침투 감지 방법 | |
EP0778714A2 (en) | Software-based bridging system for full duplex audio telephone conferencing | |
WO2019074238A1 (ko) | 마이크로폰, 마이크로폰을 포함하는 전자 장치 및 전자 장치의 제어 방법 | |
CA2240592A1 (en) | Sound system | |
WO2023085858A1 (ko) | 히어 모드 및 뮤직 모드를 제공하는 보청 이어폰의 모드 제공 방법 및 그 시스템 | |
CN112243182A (zh) | 拾音电路、方法及装置 | |
WO2016167464A1 (ko) | 스피커 정보에 기초하여, 오디오 신호를 처리하는 방법 및 장치 | |
WO2021091063A1 (ko) | 전자장치 및 그 제어방법 | |
JP2002062900A (ja) | 収音装置及び受信装置 | |
WO2019103382A1 (ko) | 전자장치 및 그 제어방법 | |
EP3818725A1 (en) | Electronic device including a plurality of speakers | |
WO2022250387A1 (ko) | 음성을 처리하기 위한 음성 처리 장치, 음성 처리 시스템 및 음성 처리 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: 12789939 Country of ref document: EP Kind code of ref document: A2 |
|
ENP | Entry into the national phase |
Ref document number: 2014512772 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14119982 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 14/03/14) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 12789939 Country of ref document: EP Kind code of ref document: A2 |