JP2017157880A - Echo cancellation device and echo cancellation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To properly perform echo cancellation while tracking a change of an acoustic environment.SOLUTION: An echo cancellation device 101 of the present invention, comprises: an echo cancellation signal forming part 23 that forms an echo cancellation signal sc from a first sound signal s1 as a signal of a first sound which has to be outputted from a speaker 5; a delay part 9 that performs a delay processing to the echo cancellation signal sc; and a transmission side sound processing part 8 in which a second sound signal s2 as the signal of a second sound gathered by a microphone 6 and the echo cancellation signal sc to which the delay processing is to be processed are inputted, and which synthesizes both of them. Further, the device comprises: a time difference calculation part 10 that calculates a time difference of an input time corresponding to the transmission side sound processing part 8 of the echo cancellation signal sc and the input time for the transmission side sound processing part 8 of the second sound signal s2 corresponding to the first sound signal s1 of a formation source of the echo cancellation signal sc; and a delay time setting part 11 that sets the delay time of the delay processing on the basis of the time difference.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an echo cancellation apparatus used for a hands-free telephone.

  2. Description of the Related Art Conventionally, a hands-free call system capable of making a call with a partner without holding a telephone terminal has been proposed. The hands-free call system enables a call without using a mobile phone by using a speaker and a microphone in cooperation with the mobile phone. Specifically, the received voice from the other party is output to a speaker, and the own voice is collected by a microphone and transmitted to the other party.

  However, in a hands-free call system, since the speaker and the microphone are often arranged close to each other, the reception voice output from the speaker is collected by the microphone and mixed with the transmission voice to the other party as an echo voice. As a result, the echo phenomenon that the call quality deteriorates becomes a problem.

  In order to reduce the echo phenomenon, a hands-free call system having an echo cancellation function has been proposed. In this hands-free call system, the call quality is improved by using the voice input from the microphone minus the echo voice as the transmission voice. First, an echo characteristic is determined based on a signal output from a speaker and input to a microphone, and an echo signal to be subtracted is estimated based on the determined echo characteristic and a reference signal.

  Patent Document 1 discloses a hands-free call system as an in-vehicle device, and in particular, considering that the nature of the echo path varies depending on the vehicle model, the delay time and volume level of the echo signal are changed for each vehicle model. Has been.

JP 2001-119470 A

  However, since the echo cancellation system of Patent Document 1 performs echo cancellation by fixing the echo characteristics for each vehicle type, for example, an external amplifier serving as a delay element is attached to the system, or the attachment position of the external microphone is changed. When the acoustic environment changes, such as, or the like, there has been a problem that it is not possible to perform echo cancellation appropriately following this.

  The present invention has been made in view of the above-described problems, and an object thereof is to appropriately perform echo cancellation following a change in the acoustic environment.

  The echo canceling apparatus of the present invention is used in a hands-free telephone that transmits voice input to a microphone to a partner and reproduces the voice of the partner from a speaker, and is a first voice signal to be output from the speaker. An echo cancellation signal generating unit that generates an echo cancellation signal from one audio signal; a delay unit that applies a delay process to the echo cancellation signal; a second audio signal that is a second audio signal collected by a microphone; and a delay process An echo canceling device comprising an audio processing unit that receives the echo cancellation signal subjected to the processing and synthesizes the two, the input time of the echo canceling signal to the audio processing unit, and the first source of the echo cancellation signal creation source A time difference calculation unit that calculates a time difference between the input time of the second audio signal corresponding to one audio signal and the input time to the audio processing unit; Further comprising a delay time setting unit that sets, based time to time difference, the.

  The echo canceling method of the present invention is an echo canceling method used in a hands-free telephone that transmits voice input to a microphone to a partner and reproduces the voice of the partner from the speaker. An echo cancellation signal is created from the first audio signal that is an audio signal, the echo cancellation signal is subjected to delay processing, and the second audio signal that is the second audio signal collected by the microphone is subjected to delay processing. The echo processing unit obtains the echo cancellation signal and synthesizes them, and the input time of the echo cancellation signal to the speech processing unit and the speech processing of the second speech signal corresponding to the first speech signal from which the echo cancellation signal is created The time difference from the input time for the part is calculated. Then, the delay time in the delay process is set based on the time difference.

  The echo cancellation apparatus of the present invention calculates the time difference between the input time of the echo cancellation signal to the audio processing unit and the input time of the second audio signal corresponding to the first audio signal from which the echo cancellation signal is generated to the audio processing unit. And a delay time setting unit that sets the delay time of the delay process based on the time difference. Therefore, when the input time of the second audio signal to the audio processing unit changes due to a change in the acoustic environment, the echo cancellation is appropriately performed by adjusting the delay time of the delay processing for the echo cancellation signal accordingly. I can do it.

  In the echo cancellation method of the present invention, the time difference between the input time of the echo cancellation signal to the sound processing unit and the input time of the second sound signal corresponding to the first sound signal from which the echo cancellation signal is generated to the sound processing unit is calculated. To do. The delay time in the delay process is set based on the time difference. Therefore, when the input time of the second audio signal to the audio processing unit changes due to a change in the acoustic environment, the echo cancellation is appropriately performed by adjusting the delay time of the delay processing for the echo cancellation signal accordingly. I can do it.

1 is a configuration diagram of an echo cancellation apparatus according to Embodiment 1. FIG. 6 is a configuration diagram of an echo cancellation apparatus according to Embodiment 2. FIG. It is a flowchart which shows the operation | movement regarding an echo cancellation function. It is a flowchart which shows the operation | movement regarding adjustment of delay time. It is a figure which shows the relationship between the input time of a test signal and an echo cancellation signal. FIG. 5 is a hardware configuration diagram of an echo cancellation apparatus according to a second embodiment.

<A. Embodiment 1>
FIG. 1 is a configuration diagram of an echo cancellation apparatus 101 according to Embodiment 1 showing a configuration essential to the present invention. The echo cancel device 101 is used in a hands-free telephone that transmits the voice input to the microphone 6 to the other party and reproduces the other party's voice from the speaker 5. The echo cancellation apparatus 101 includes a communication unit 1 that transmits and receives audio signals to and from the mobile phone terminal 200, a reception side audio processing unit 2 (first audio processing unit), and a transmission side audio processing unit 8 (second audio processing unit). ). The receiving side audio processing unit 2 is a signal of the first audio to be output from the audio processing unit 21 that performs predetermined audio processing on the received audio signal sr acquired from the communication unit 1 and the speaker 5 including the other party's audio. An echo cancel signal creating unit 23 for creating an echo cancel signal sc from the first audio signal s1 is provided. In addition, the echo cancellation apparatus 101 includes a delay unit 9 that performs a delay process on the echo cancellation signal sc.

  The transmitter voice processing unit 8 includes an echo canceling unit 81. The echo cancellation unit 81 receives a second audio signal s2 that is a second audio signal collected by the microphone 6 and an echo cancellation signal sc that has been subjected to delay processing, and the echo cancellation unit 81 synthesizes both. To do.

  Further, the echo canceling apparatus 101 transmits the echo cancellation signal sc to the transmission side voice processing unit 8 and the transmission side of the second voice signal s2 corresponding to the first voice signal s1 from which the echo cancellation signal sc is created. A time difference calculation unit 10 that calculates a time difference from an input time for the audio processing unit 8 and a delay time setting unit 11 that sets a delay time of the delay process based on the time difference are provided. Therefore, when the transmission time of the echo path changes, the delay time of the echo cancellation signal sc can be adjusted accordingly. Therefore, appropriate echo cancellation processing can be performed following the change in echo characteristics.

<B. Second Embodiment>
FIG. 2 is a configuration diagram of the echo cancellation apparatus 102 according to Embodiment 2 of the present invention. Hereinafter, although the echo cancellation apparatus 102 is demonstrated as vehicle equipment, such as a car navigation apparatus, it is not restricted to this.

  The echo canceling apparatus 102 has a hands-free function for outputting the reception voice of the mobile phone terminal 200 from the speaker 5 and outputting the voice input from the microphone 6 to the mobile phone terminal 200 as the transmission voice. The speaker 5 and the microphone 6 are installed in the vehicle interior 201. As a result, the user in the vehicle interior 201 can make a call with the other party without having the mobile phone terminal 200 in hand.

  Hereinafter, the hands-free function of the echo cancellation apparatus 102 will be described. The echo cancellation apparatus 102 includes a communication unit 1, a reception side voice processing unit 2, a D / A conversion unit 3, a signal amplification unit 4, an A / D conversion unit 7 and a transmission side voice processing unit 8. The communication unit 1 is a communication interface that transmits and receives audio signals to and from the mobile phone terminal 200 in a wired or wireless manner, and is, for example, a Bluetooth (registered trademark) standard communication interface. When the communication unit 1 obtains the reception voice signal sr from the mobile phone terminal 200, the communication unit 1 sends it to the reception side voice processing unit 2. In the receiving side voice processing unit 2, the voice processing unit 21 performs predetermined voice processing on the received voice signal sr and sends it to the D / A conversion unit 3. The received voice signal sr is converted from a digital signal to an analog signal by the D / A converter 3, amplified by the signal amplifier 4, and output as a received voice from the speaker 5. In this way, the user can listen to the received voice of the mobile phone terminal 200 from the speaker 5.

  The microphone 6 collects a user's uttered voice and the like. The audio signal s2 input to the microphone 6 is converted from an analog signal to a digital signal by the A / D conversion unit 7 and sent to the transmission side audio processing unit 8. The transmitter voice processing unit 8 performs predetermined voice processing on the voice signal s 2, and then outputs the processed voice signal to the communication unit 1. The communication unit 1 transmits the audio signal acquired from the transmission side audio processing unit 8 to the mobile phone terminal 200. Thus, the user's utterance voice input to the microphone 6 can be used as the transmission voice from the mobile phone terminal 200.

  Such a hands-free function makes it possible to make a call without operating the mobile phone terminal 200 at hand even while driving the vehicle, and without removing the line of sight from the front, which is particularly useful for use during driving. However, since the speaker 5 and the microphone 6 are arranged close to each other, the output sound of the speaker 5 is input to the microphone 6, so that an echo phenomenon occurs. In this case, since the other party of the call hears his / her uttered voice with a delay, the call quality deteriorates. In particular, in the in-vehicle echo canceling apparatus 102, since the speaker 5 and the microphone 6 are provided in the same vehicle interior, the output sound of the speaker 5 easily enters the microphone 6 and an echo phenomenon is likely to occur.

  Therefore, the echo cancellation apparatus 102 has an echo cancellation function. Hereinafter, the echo cancellation function of the echo cancellation apparatus 102 will be described with reference to the flowchart of FIG. In this specification, an output audio signal from the speaker 5 included in the input signal of the microphone 6 is referred to as an echo signal or an echo component. The voice transmission path from the receiver side voice processor 2 to the transmitter side voice processor 8 via the speaker 5 and the microphone 6 is referred to as an echo path.

  In addition to the voice processing unit 21, the receiver side voice processing unit 2 includes an echo cancellation signal creation unit 23. When the communication unit 1 receives the reception voice signal sr from the mobile phone terminal 200 (step S1), the echo cancellation signal generation unit 23 acquires the reception voice signal sr from the communication unit 1 and generates the echo cancellation signal sc ( Step S2). The echo cancellation signal is a signal in which an echo signal to be eliminated in an echo cancellation process described later has an opposite phase. The echo signal is created by measuring the acoustic characteristics of the echo path in advance and estimating how the voice signal changes through the echo path. For example, in an ideal echo path in which attenuation of frequency components does not occur, the echo cancellation signal sc can be created by simply setting the received voice signal sr to an opposite phase.

  The echo cancellation signal creation unit 23 outputs the created echo cancellation signal sc to the delay unit 9. The delay unit 9 performs a delay process for a predetermined delay time on the echo cancellation signal sc (step S3).

  The transmitting side voice processing unit 8 includes an echo canceling unit 81. The echo cancel unit 81 acquires the delayed echo cancel signal sc from the delay unit 9, acquires the input audio signal s2 of the microphone 6 from the A / D conversion unit 7, and performs a synthesis process (echo cancel process) thereof. Perform (step S4).

  The signal (s2 + sc) is transmitted from the echo cancel unit 81 to the communication unit 1, and is transmitted from the communication unit 1 to the mobile phone terminal 200 as a transmission voice signal (step S5). And it is transmitted to the other party as a transmission voice signal of the mobile phone terminal 200. The reception voice signal sr output from the reception side voice processing unit 2 is input to the transmission side voice processing unit 8 through an echo path.

  The delay unit 9 sets a delay time so that the echo cancellation signal is input to the transmission side voice processing unit 8 at the same time as the reception voice signal input to the transmission side voice processing unit 8 via the echo path. Is done. The echo cancellation signal sc is generated so as to be an antiphase component of the received voice signal that is output from the speaker 5 and then input to the microphone 6 and input to the transmitting side voice processing unit 8. Accordingly, the echo component is removed from the signal (s2 + sc) obtained by the echo cancellation process.

  However, the transmission time of the echo path varies depending on factors such as attachment of an external amplifier having a delay time, or changes in the positions of the speaker 5 and the microphone 6. Therefore, if the delay time in the delay unit 9 is fixed, it is not possible to cancel echo following the fluctuation of the echo path. Therefore, the echo canceling device 102 adjusts the delay time of the echo cancellation signal in the delay unit 9 following the change in the transmission time in the echo path.

  Hereinafter, the configuration and operation of the echo cancellation apparatus 102 related to the adjustment of the delay time will be described. The echo cancellation apparatus 102 includes a time difference calculation unit 10 and a delay time setting unit 11. The reception side voice processing unit 2 includes a test signal creation unit 22, and the transmission side voice processing unit 8 includes a volume level detection unit 82.

  FIG. 4 is a flowchart showing the delay time adjustment operation in the echo cancellation apparatus 102. The delay time adjustment operation will be described below with reference to FIG. The flow of FIG. 4 is started by performing an operation input to the echo canceling apparatus 102 from an operation receiving unit (not shown in FIG. 2), for example. For this operation input, the echo cancellation apparatus may include an input interface such as a GUI.

  First, the test signal generator 22 generates a test signal st having a predetermined frequency and volume level (step S11). The test signal st generated by the test signal generator 22 is sent to the D / A converter 3 and the echo cancellation signal generator 23. The test signal st sent to the D / A conversion unit 3 is input to the transmission side voice processing unit 8 via an echo path.

  On the other hand, the echo cancellation signal creation unit 23 creates an echo cancellation signal sc from the test signal st (step S12). The method for creating the echo cancel signal sc is the same as the method for creating the echo cancel signal sc from the received voice signal sr. The echo cancellation signal sc created by the echo cancellation signal creation unit 23 is subjected to delay processing of a predetermined delay time by the delay unit 9 (step S13), and is input to the transmission side voice processing unit 8.

  The volume level detection unit 82 has a function of detecting the volume level of the input audio signal for each frequency, whereby the test signal st and the echo cancellation signal sc are input to the transmission side audio processing unit 8. Is detected. Then, the reception time T1 of the test signal st is recorded (step S14), and the input time T2 of the echo cancellation signal sc is recorded (step S15). Then, the volume level detector 82 outputs T1 and T2 to the time difference calculator 10. The time difference calculator 10 calculates the difference ΔT = | T2−T1 | between the input times T1 and T2 of the test signal st and the echo cancellation signal sc acquired from the volume level detector 82 as shown in FIG. (Step S16). In FIG. 5, the test signal generation time is shown as T0.

  Then, the echo cancellation apparatus 102 sleeps for a predetermined time (for example, several tens of ms) (step S17), and determines whether ΔT has been calculated N times (step S18). If not, the process returns to step S11, and the test signal creation unit 22 again generates a test signal and repeats the calculation of ΔT.

  When ΔT is calculated N times, the delay time setting unit 11 calculates an average value ΔTave of N ΔTs (step S19). Then, (T−ΔTave) obtained by subtracting ΔTave from the previous delay time T in the delay unit 9 is set as a new delay time T (step S20). Here, since the case where the test signal st is input to the volume level detection unit 82 prior to the echo cancellation signal sc is described, ΔTave is subtracted from the delay time T thus far, but the echo cancellation signal sc Is first input to the sound volume level detection unit 82, a value obtained by adding ΔTave from the previous delay time T (T + ΔTave) is set as a new delay time T.

  The flow in FIG. 4 may be performed in response to a user operation input as described above, or may be performed at any time, such as for a certain period, without a user operation input. As described above, the delay time of the echo path is adjusted by adjusting the delay time T in the delay unit 9 based on the difference between the input times T1 and T2 of the test signal st and the echo cancellation signal sc to the transmitting side voice processing unit 8. Even when the signal changes (for example, when an external amplifier is connected), the test signal st and the echo cancellation signal sc can be simultaneously input to the transmitting side voice processing unit 8. Therefore, it is possible to appropriately perform echo cancellation processing following changes in the echo path.

  FIG. 6 is a hardware configuration diagram of the echo cancellation apparatus 102. The echo cancellation apparatus 102 includes a processor 31, a memory 32, an A / D conversion circuit 33, an amplification circuit 34, a D / A conversion circuit 35, and a communication I / F 36. The D / A conversion unit 3 is realized by a D / A conversion circuit 35, and the A / D conversion unit 7 is realized by an A / D conversion circuit 33. The communication unit 1 is realized by the communication I / F 36. In addition, the receiving side voice processing unit 2, the sending side voice processing unit 8, the delay unit 9, the time difference calculation unit 10, and the delay time setting unit 11 are configured such that a processor 31 such as a CPU (Central Processing Unit) has a RAM (Random Access Memory). The function of the processor 31 is realized by executing a software program stored in the memory 32. However, these may be realized in cooperation with a plurality of processors 31, for example.

  In place of this, the receiving side voice processing unit 2, the sending side voice processing unit 8, the delay unit 9, the time difference calculation unit 10, and the delay time setting unit 11 perform signal processing for realizing the operation by an electrical circuit of hardware. It may be realized by a circuit. Software receiving side speech processing unit 2, transmitting side speech processing unit 8, delay unit 9, time difference calculation unit 10 and delay time setting unit 11, hardware receiving side speech processing unit 2, transmission side speech processing unit 8 As a concept combining the delay unit 9, the time difference calculation unit 10, and the delay time setting unit 11, the word “processing circuit” can be used instead of the word “part”.

  As described above, the echo cancellation apparatus 102 according to the second embodiment is used in a hands-free telephone that transmits the voice input to the microphone 6 to the other party and reproduces the other party's voice from the speaker 5. An echo cancellation signal creating unit 23 that creates an echo cancellation signal sc from a first audio signal s1 that is a first audio signal to be output from the speaker 5, and a delay unit 9 that performs a delay process on the echo cancellation signal sc; A second voice signal s2 that is a signal of the second voice collected by the microphone 6 and an echo cancel signal sc that has been subjected to delay processing are input, and a transmitting voice processing unit 8 (voice processing unit) that synthesizes both of them. And). Further, the echo canceling apparatus 101 transmits the echo cancellation signal sc to the transmission side voice processing unit 8 and the transmission side of the second voice signal s2 corresponding to the first voice signal s1 from which the echo cancellation signal sc is created. A time difference calculating unit that calculates a time difference from the input time to the audio processing unit 8; and a delay time setting unit 11 that sets a delay time of the delay process based on the time difference. Therefore, when the transmission time of the echo path changes, the delay time of the echo cancellation signal sc can be adjusted accordingly. Therefore, appropriate echo cancellation processing can be performed following the change in echo characteristics.

  In addition, the echo cancellation apparatus 102 includes a test signal creation unit 22 that creates a test signal st for adjusting the delay time of the delay processing in the delay unit 9. The first audio signal that is the audio signal to be output from the speaker 5 includes the test signal st, and the echo cancellation signal sc that the time difference calculation unit 10 calculates the time difference is an echo cancellation signal created from the test signal st. The second audio signal s2 for which the time difference calculator 10 calculates the time difference is the second audio signal s2 corresponding to the test signal st. By using a test signal st different from the received voice signal sr, for example, a test signal st having a specific frequency and a specific volume level, the test signal st input to the transmitting voice processing unit 8 via the echo path can be easily obtained. Can be identified.

  Further, when the microphone 6 and the speaker 5 are installed in the vehicle interior 210, since the output sound from the speaker 5 is easily input to the microphone 6, it is necessary to suppress the echo phenomenon. When the end user uses an external microphone, the distance to the speaker 5 may vary greatly depending on the mounting position of the microphone 6, and the delay time may vary. Therefore, the echo cancellation apparatus of the present invention is particularly effective.

  The echo canceling method of the present invention is an echo canceling method used for a hands-free telephone that transmits voice input to a microphone to the other party and reproduces the other party's voice from the speaker. The echo cancellation signal sc is generated from the first audio signal s1 that is a power signal, the delay processing is performed on the echo cancellation signal sc, and the second audio signal s2 that is the audio signal collected by the microphone 6 and the delay processing The echo cancellation signal sc subjected to the processing is acquired by the transmission side voice processing unit 8 and synthesized, and the input time of the echo cancellation signal sc to the transmission side voice processing unit 8 and the origin of the echo cancellation signal sc The time difference between the second audio signal s2 corresponding to the first audio signal s1 and the input time of the second audio signal s2 to the transmitting side audio processing unit 8 is calculated. The delay time in the delay process is adjusted based on the time difference. Therefore, when the transmission time of the echo path changes, the delay time of the echo cancellation signal sc can be adjusted accordingly. Therefore, appropriate echo cancellation processing can be performed following the change in echo characteristics.

  It should be noted that the present invention can be freely combined with each other within the scope of the invention, and each embodiment can be appropriately modified or omitted.

  DESCRIPTION OF SYMBOLS 1 Communication part, 2 receiver side voice processing part, 3 D / A conversion part, 4 signal amplification part, 5 speaker, 6 microphone, 7 A / D conversion part, 8 transmission side voice processing part, 9 delay part, 10 time difference Calculation unit, 11 delay time setting unit, 21 sound processing unit, 22 test signal creation unit, 23 echo cancellation signal creation unit, 81 echo cancellation unit, 82 volume level detection unit, 200 mobile phone terminal, 201 vehicle interior.

Claims (4)

Used for hands-free telephones that transmit the voice input to the microphone to the other party and play the other party's voice from the speaker.
An echo cancellation signal creating unit that creates an echo cancellation signal from a first audio signal that is a first audio signal to be output from the speaker;
A delay unit that performs a delay process on the echo cancellation signal;
A second audio signal that is a second audio signal collected by the microphone, and the echo cancellation signal that has been subjected to the delay processing, and a voice processing unit that synthesizes both;
An echo canceling device comprising:
A time difference for calculating a time difference between an input time of the echo cancellation signal to the audio processing unit and an input time of the second audio signal corresponding to the first audio signal from which the echo cancellation signal is generated to the audio processing unit A calculation unit;
A delay time setting unit for setting a delay time of the delay processing based on the time difference;
Further comprising
Echo cancellation device. A test signal creating unit for creating a test signal for adjusting the delay time of the delay processing;
The first audio signal includes the test signal;
The echo cancellation signal for calculating the time difference by the time difference calculation unit is the echo cancellation signal created from the test signal,
The second audio signal for which the time difference calculation unit calculates the time difference is the second audio signal corresponding to the test signal.
The echo cancellation apparatus according to claim 1. The microphone and the speaker are installed in a vehicle interior.
The echo cancellation apparatus according to claim 1 or 2. An echo canceling method used for a hands-free telephone that transmits voice input to a microphone to a destination and plays back the voice of the destination from a speaker,
Creating an echo cancellation signal from the first audio signal that is the first audio signal to be output from the speaker;
Applying a delay process to the echo cancellation signal,
A second audio signal that is a signal of the second audio collected by the microphone and the echo cancellation signal that has been subjected to the delay processing are acquired by an audio processing unit, and both are combined,
Calculating a time difference between an input time of the echo cancellation signal to the audio processing unit and an input time of the second audio signal corresponding to the first audio signal from which the echo cancellation signal is generated to the audio processing unit;
Setting a delay time of the delay processing based on the time difference;
Echo cancellation method.

Images