Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04M—TELEPHONIC COMMUNICATION
  • H04M9/00—Arrangements for interconnection not involving centralised switching
  • H04M9/08—Two-way loud-speaking telephone systems with means for conditioning the signal, e.g. for suppressing echoes for one or both directions of traffic
  • H04M9/082—Two-way loud-speaking telephone systems with means for conditioning the signal, e.g. for suppressing echoes for one or both directions of traffic using echo cancellers
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; ELECTRIC HEARING AIDS; PUBLIC ADDRESS SYSTEMS
  • H04R3/00—Circuits for transducers
  • H04R3/02—Circuits for transducers for preventing acoustic reaction, i.e. acoustic oscillatory feedback

Definitions

• the present invention relates to an echo suppression device that suppresses echoes that circulate into a plurality of speaker power microphones.
• FIG. 3 is a block diagram of a conventional echo suppression device described in Patent Document 1. As shown in Figure 3, there are two speakers and two microphones. Hereinafter, each unit constituting the echo suppression device shown in FIG. 3 will be described.
• left and right stereo signals are input to the input terminals 301 and 302, and the DZ A converters 303 and 304, respectively.
• the DZ A converters 303 and 304 convert the left and right stereo signals into digital signals, and the speakers 305 and 306 output the signals converted into analog signals as sound.
• Sounds output from the speakers 305 and 306 are picked up by microphones 307 and 308 including echoes through spatial transmission paths hLL, hLR, hRL, and hRR, and are collected by the AZD converters 309 and 310. When the signal is converted to analog signal force digital signal.
• the sum / difference signal generation unit 321 calculates the sum signal and the difference signal of the signals from the input terminals 301 and 302 by the calorie calculator 322 and the subtractor 323, and the correlation detection unit 325 outputs the sum signal and the difference signal. Detect power correlation.
• Transfer function calculation section 324 estimates the spatial transfer characteristics between the speaker and the microphone. Specifically, the transfer function calculation unit 324 performs a cross spectrum calculation between the output signals of the subtractors 319 and 320 and the output signal of the sum / difference signal generation unit 321, and the correlation detection unit 325 reduces the correlation. Filter 313, 314 so that the estimation error of the spatial transfer characteristics is minimized.
• Each of the filters 313, 314, 315, and 316 uses a coefficient calculated by the transfer function calculation unit 324, and a spatial transfer characteristic in a spatial transfer path between the speakers 305 and 306 and the microphones 307 and 308. Simulate hLL, hLR, hRL, and hRR.
• Adder 317 adds the signals output from filters 313 and 314 to generate a pseudo echo signal
• adder 318 adds the signals output from filters 315 and 316 to generate a pseudo echo signal. Is generated.
• the subtractor 319 suppresses the echo by subtracting the pseudo echo signal of the adder 317 from the digital signal power of the AZD converter 309 and outputs it to the output terminal 311.
• the subtracter 320 outputs the digital signal of the AZ D converter 310.
• the signal power is also subtracted from the pseudo echo signal of the adder 318 to suppress the echo and output to the output terminal 312.
• Patent Document 1 Japanese Patent Laid-Open No. 2003-102085 (Page 56, Fig. 1)
• the present invention has been made to solve the conventional problems, and is capable of suppressing echoes corresponding to each spatial transmission path with a simple configuration as compared with a conventional echo suppression device.
• An echo suppressor capable of reducing the amount is provided. Means for solving the problem
• the echo suppressor of the present invention has a plurality of speakers and one or more microphones arranged at symmetrical positions so that the symmetry of arrangement with respect to a plurality of spatial transmission paths between the speakers and the microphones is improved.
• Target to suppress the echo A target signal generating unit that generates a signal, a reference signal generating unit that generates a reference signal corresponding to the target signal based on a signal before the speaker outputs the sound, and the target signal and the target signal And an echo suppression unit that suppresses the echo from the collected sound signal based on a reference signal.
• the echo suppression device of the present invention has a configuration in which the speakers are arranged symmetrically with respect to a center line and the microphones are arranged symmetrically with respect to the same center line.
• the echo suppression device of the present invention has a configuration in which the center speaker is arranged on the center line when the number of speakers is an odd number.
• the echo suppression device of the present invention has a configuration in which the center microphone is arranged on the center line when the number of the microphones is an odd number.
• the target signal generation unit in the echo suppressor of the present invention outputs one or more of an addition signal, a subtraction signal, or a through signal based on the sound pickup signal as the target signal. Have a configuration! / Speak.
• the reference signal generation unit includes an addition signal and a subtraction signal based on a signal before the speaker outputs the sound as the reference signal.
• the echo suppression unit in the echo suppression device of the present invention has a configuration using an adaptive filter.
• the echo suppression unit in the echo suppression device of the present invention has a configuration that performs processing related to suppression of the echo limited to a specific frequency band in the target signal and the reference signal.
• the echo suppression unit in the echo suppression apparatus of the present invention has a configuration for controlling the update of the filter coefficient of the adaptive filter in accordance with the presence or absence of an echo.
• the echo suppression apparatus of the present invention includes a signal synthesis unit that synthesizes a plurality of output signals output from the echo suppression unit, and the signal synthesis unit outputs each of the outputs output from the echo suppression unit. It has a configuration to synthesize by adding a predetermined constant to the addition signal and subtraction signal based on the signal.
• the present invention can suppress the echo corresponding to each spatial transmission path with a simple configuration as compared with the conventional echo suppressor, and can reduce the amount of calculation.
• a suppression device is provided.
• FIG. 1 is a block diagram of an echo suppression device according to a first embodiment of the present invention.
• FIG. 2 is a block diagram of an echo suppressor according to a second embodiment of the present invention.
• FIG.3 Block diagram of a conventional echo suppressor
• FIG. 1 is a block diagram of an echo suppressor according to the first embodiment of the present invention. As shown in Fig. 1, the number of speakers and microphones is two, and they are arranged symmetrically with respect to the same center line. Hereinafter, each unit constituting the echo suppression apparatus shown in FIG. 1 will be described.
• left and right stereo signals are input from the input terminals 101 and 102 to the DZ / converters 103 and 104, respectively.
• DZ A converters 103 and 104 convert left and right stereo signals.
• the digital signal power is also converted into an analog signal, and the speakers 105 and 106 output the signal converted into the analog signal as sound.
• the sound output from the speakers 105 and 106 is collected as an echo by the microphones 107 and 108 via the spatial transmission paths hLL, hLR, hRL, and hRR.
• the speaker and microphone shown in Fig. 1 are arranged symmetrically with the same center line as the boundary, so that each spatial transmission path can be approximated by the following equation.
• the AZD converters 109 and 110 convert the signal when the sound is picked up into an analog signal power into a digital signal.
• the target signal generation unit 111 includes an adder 117 and a subtractor 118, and generates a target signal from the digital signals converted by the A / D converters 109 and 110. Note that the target signal generation unit 111 may output the signals output from the AZD converters 109 and 110 as they are.
• the signal when the input signal is output as it is is referred to as a through signal.
• the reference signal generation unit 112 includes an adder 119 and a subtractor 120, calculates a sum signal and a difference signal of signals from the input terminals 101 and 102, and is referred to by the echo suppression unit 113. Is supposed to generate. Note that the reference signal generation unit 112 may output the through signal to the echo suppression unit 113.
• the echo suppression unit 113 includes adaptive filters 121 and 122 and subtractors 123 and 124 that estimate spatial transfer characteristics, and the target signal generated by the target signal generation unit 111 and the reference signal generation unit 112 generate Based on the reference signal, the echo is suppressed! /.
• the signal synthesis unit 114 includes an adder 125, a subtractor 126, and multipliers 127 and 128.
• the signal synthesis unit 114 synthesizes the signals output from the echo suppression unit 113 and outputs them to the output terminals 115 and 116. It is.
• the signal synthesis unit 114 may output the through signal to the output terminals 115 and 116.
• Stereo signal sL input from input terminals 101, 102 to DZA converters 103, 104 And sR are D / A converted by the D / A converters 103 and 104, and then output as speaker sounds 105 and 106.
• Sounds output from the speakers 105 and 106 are input as echo signals to the microphones 107 and 108 via the spatial transmission paths hLL, hLR, hRL, and hRR, and are AZD converted by the A / D converters 109 and 110. .
• Equation 1 and Equation 2 are signals that are originally desired to be picked up by the microphone
• the signals mL and mR output by the A / D converters 109 and 110 are respectively It is expressed by Equation 3 and Equation 4.
• * represents a convolution operation.
• the target signal generation unit 111 uses the adder 117 and the subtractor 118 to calculate the addition signal and the subtraction signal as the target signal of the echo suppression unit 113 as shown in the following Expression 5 and Expression 6.
• the reference signal generation unit 112 calculates an addition signal sL + sR and a subtraction signal sL-sR, which are reference signals for the echo suppression unit 113, using the adder 119 and the subtractor 120.
• the echo suppressor 113 calculates a pseudo echo signal by convolving the reference signal sL + sR and the filter coefficient with the adaptive filter 121, and the subtractor 123 calculates the pseudo echo signal with the target signal power of Equation 5. Echo is suppressed by subtraction.
• the echo suppression unit 113 calculates a pseudo echo signal by convolving the reference signal sL-sR with the filter coefficient by the adaptive filter 122, and subtracts the calculated pseudo echo signal by the subtractor 124 for the target signal force of Equation 6. This suppresses the echo.
• the output signals from the subtracters 123 and 124 corresponding to the error signals of the adaptive filters 121 and 122 are expressed by Expression 7 and Expression 8, respectively.
• the filter coefficients used by adaptive filters 121 and 122 are expressed by Equation 7 and Equation 8.
• the output signal is sequentially updated so that the root mean square value of the output signal is minimized. This corresponds to the adaptive filters 121 and 122 estimating the combined spatial transfer characteristic hLL + hLR and the combined spatial transfer characteristic hLL ⁇ hLR, respectively.
• various known adaptive algorithms such as the NLMS (Normalized Least Mean Square) algorithm are used.
• the signal synthesis unit 114 calculates an addition signal and a subtraction signal by the adder 125 and the subtractor 126 for the output signals eL and eR from the echo suppression unit 113, and then outputs 0. 0 by the multipliers 127 and 128, respectively. By multiplying by 5, the signals vL and vR that are originally desired to be collected are calculated and output to the output terminals 115 and 116.
• the echo suppressor according to the first embodiment of the present invention uses the symmetry of the arrangement of the speech force and the microphone, and the spatial transmission path to be estimated, and corresponding to this. Since the number of echoes to be reduced can be reduced, the echo corresponding to each spatial transmission path can be suppressed with a simple configuration and the amount of computation can be reduced.
• the number of speakers and the number of microphones are the same as described above.
• the present invention is not limited to this.
• a plurality of speakers and one or more microphones are symmetrical.
• the target signal generation unit 111, the reference signal generation unit 112, and the echo can be suppressed as long as they are arranged at a certain position, and so that echo suppression on a plurality of spatial transmission paths can be realized according to the arrangement of the speaker and the microphone.
• the suppression unit 113 and the signal synthesis unit 114 may be appropriately configured.
• the center speakers arranged symmetrically with respect to the same center line need only be installed on the center line.
• the center microphones arranged symmetrically with respect to the same center line need only be installed on the center line.
• the echo suppression unit 113 is used when the frequency band of the echo to be suppressed is limited to some extent (e.g., for a signal that is relatively low, such as speech, and where the frequency is concentrated), The processing may be limited to the frequency band necessary for echo suppression.
• FIG. 2 is a block diagram of an echo suppression apparatus according to the second embodiment of the present invention.
• Na Of the components constituting the echo suppressor according to the second embodiment of the present invention, the same components as those constituting the echo suppressor according to the first embodiment of the present invention are used. Are denoted by the same reference numerals, and description thereof is omitted.
• the echo suppression device shown in FIG. 2 includes an echo detection unit 201 in addition to the configuration of the echo suppression device shown in FIG.
• the echo detection unit 201 calculates the cross-correlation between the input signal of the input terminal 101 and the output signal of the adder 117, which is a constituent element of the target signal generation unit 111, and according to the magnitude of the cross-correlation value.
• the filter coefficients of the adaptive filters 121 and 122 that are components of the echo suppression unit 113 are updated.
• the echo suppression unit 113 updates the filter coefficients used by the adaptive filters 121 and 122 when the echo detection unit 201 determines that there is an echo.
• the echo suppression apparatus improves the estimation accuracy of the spatial transfer characteristics by controlling the filter coefficient update according to the presence or absence of an echo. Is possible.
• the echo detector 201 correlates the input signal of the input terminal 101 and the output signal of the adder 117 that is a component of the target signal generator 111.
• the level ratio may be calculated instead of the cross-correlation.
• the echo detection unit 201 may be configured to refer to the input signal of the input terminal 102 and the output signal of the adder 117 that is a component of the target signal generation unit 111.
• the same number of echo detectors 201 as the adaptive filters may be provided, and the adaptive filters may be configured to refer to appropriate signals so that the adaptive filters can control the filter coefficient with high accuracy.
• the echo suppressor according to the present invention has an effect that the echo corresponding to each spatial transmission path can be suppressed with a simple configuration as compared with the conventional echo suppressor, It is useful as an echo suppressor that suppresses echoes that circulate to a plurality of speaker power microphones, in addition to various acoustic devices having a plurality of speakers and one or more microphones.

Landscapes

• Engineering & Computer Science (AREA)
• Signal Processing (AREA)
• Health & Medical Sciences (AREA)
• General Health & Medical Sciences (AREA)
• Otolaryngology (AREA)
• Physics & Mathematics (AREA)
• Acoustics & Sound (AREA)
• Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
• Circuit For Audible Band Transducer (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (1)

Family

ID=38833252

Family Applications (1)

Country Status (5)

Cited By (1)

Families Citing this family (28)

Citations (2)

Family Cites Families (3)

• 2006
  • 2006-06-23 JP JP2006173750A patent/JP2008005293A/ja active Pending
• 2007
  • 2007-05-25 EP EP07744111A patent/EP2037699A1/en not_active Withdrawn
  • 2007-05-25 WO PCT/JP2007/060677 patent/WO2007148509A1/ja not_active Ceased
  • 2007-05-25 US US12/305,154 patent/US20090169027A1/en not_active Abandoned
  • 2007-05-25 CN CNA2007800236224A patent/CN101480063A/zh active Pending

Patent Citations (2)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events