JP2944310B2 - Echo canceller - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an echo canceller, and more particularly, to an echo canceller suitable for use in a hands-free telephone.

[0002]

2. Description of the Related Art In a hands-free telephone, an omnidirectional microphone is applied to a microphone so as to increase the degree of freedom of the position of a near-end speaker. It cannot be avoided that it is captured and is transmitted as an echo towards the far-end talker. To avoid such inconveniences, many hands-free telephones have an echo canceller.

FIG. 2 shows an example of a conventional echo canceller (JP-A-3-80628). FIG. 2 shows a basic configuration, in which an analog / digital converter, a digital / analog converter, and the like are omitted.

[0004] In FIG. 2, a reception signal R relating to the utterance of a far-end speaker is provided to a speaker 104 to be sounded, and is also input to an adaptive filter 101. Speaker 104
The received voice (echo) R2 uttered from the main microphone 102 wraps around the non-directional main microphone 102,
2 and is provided to adder 105 as a subtracted input. When the received signal R is input, the adaptive filter 101 performs a convolution operation on the filter coefficient (estimated value of the echo path gain) at that time and supplies the operation value to the subtraction input terminal of the adder 105 as a pseudo echo signal. Thus,
The adder 105 outputs a signal from which the echo component has been removed from the signal captured by the microphone 102, and the signal is transmitted and supplied to the coefficient control circuit 106.
As described later, the coefficient control circuit 106 controls the adaptive filter 1 based on the signal output from the adder 105 during a period in which the update of the coefficient of the adaptive filter 101 is permitted.
The coefficient of 01 is updated.

[0005] The echo becomes a problem when the reception signal R is being received. Such cases include the case where only the reception signal R is being received and the case where the reception signal is being received. There is a case of so-called double talk in which the near-end talker 108 is emitting voice. In the former case, the adder 105 outputs a residual signal from which the echo component has been removed, which is suitable for updating the filter coefficient. On the other hand, in the latter case, the adder 105 outputs a transmission signal related to the near-end speaker in addition to the remaining echo-removed signal. It is generally not suitable for updating because the coefficient is updated based on it.

Therefore, a configuration is provided to prohibit updating of the coefficient of the adaptive filter 101 during double talk. That is, the sub microphone 103 having directivity and the double talk detection circuit 107 are provided. The directivity of the microphone 103 is determined so that the expressions (1) to (4) described later are satisfied, and the installation position is determined. Double talk detection circuit 107
Detects a double talk state from the captured signal of the main microphone 102 and the captured signal of the sub microphone 103, and when the double talk state is detected, causes the coefficient control circuit 106 to prohibit coefficient updating.

Hereinafter, a principle of detecting a double talk state in the conventional example will be described. Here, the input levels of the reception signal component, the transmission signal component, and the noise (background noise) component among the signals captured by the omnidirectional microphone 102 are R2, S2, and N2, respectively, and the directional microphone 103
, The input levels of the received signal component, the transmitted signal component, and the noise component are R3, S3, and N3, respectively.
It will be described as.

By appropriately selecting the directivity of the sub microphone 103 and the positional relationship between the microphones 102 and 103 and the speaker 104, the following expressions (1) to (4) are satisfied. That is, in this conventional example,
It is assumed that equations (1) to (4) hold.

R2 + N2> R3 + N3 (1) S3 + N3> S2 + N2 (2) R2> R3 (3) S3> S2 (4) In the transmitting state where only the near-end talker 108 is emitting a voice, The signals captured by the microphones 102 and 103 include transmission signal components S2 and S3 and noise components N2 and N3. Therefore, the power ratio Ssend of the captured signals of both microphones 102 and 103 in the transmission state can be expressed by the following equation.

Send = (S3 + N3) / (S2 + N2) (5) Similarly, when the captured signal is divided into components, the capturing of both microphones 102 and 103 in the receiving state in which only the far-end speaker is emitting voice The signal power ratio Srec is (6)
The power ratio Sdouble of the captured signals of the two microphones 102 and 103 in the double talk state in which both the near-end speaker 108 and the far-end speaker are emitting voices.
Can be expressed by equation (7).

Srec = (R3 + N3) / R2 + N2 (6) Sdouble = (R3 + S3 + N3) / (R2 + S2 + N2) (7) The power ratio Ssend in the transmission state, the power ratio Srec in the reception state, and the power ratio S in the double talk state.
When the magnitude relation of the double is obtained by using the above-described equations (1) to (4), the following equation is obtained.

Srec <Sdouble <Ssend (8) Accordingly, a power ratio between the captured signals of the two microphones 102 and 103 is obtained, and this power ratio is compared with a two-stage threshold value to obtain one of the three states. It can be determined whether there is.

[0013]

However, the conventional echo canceller shown in FIG. 2 has the following drawbacks in detecting double talk.

(1) When the noise source is large, such as when the noise source is located near the extension of the directional microphone and the near-end speaker, the above-mentioned equations (1) to (4) are satisfied. And double-talk detection performance is degraded.

(2) If the noise is large, even if there is no received signal, it is misunderstood that the receiving state is due to the noise only, and the coefficient of the adaptive filter is updated. This may cause the erasure amount to deteriorate.

(3) The difference between the powers of the signals captured by both microphones is caused by the fact that the voice from one near-end speaker is picked up by one directional microphone. Therefore, when there are a plurality of speakers at the near end and speaking at the same time, double talk detection performance deteriorates.

(4) A secondary microphone is required in addition to the main microphone, and since the directivity, the installation position, and the like are defined, the size of the entire hands-free telephone becomes large.

The present invention has been made in view of the above points, and it is possible to determine various call states including a double talk state with high accuracy irrespective of the presence of noise, thereby improving the echo canceling accuracy. It is an object of the present invention to provide a small echo canceller that can be operated.

[0019]

In order to solve this problem, according to the present invention, an adder removes an echo signal circulating from a speaker to a microphone by using a pseudo echo signal formed by an adaptive filter. The following means are provided in the echo canceller provided in the telephone.

That is, based on the autocorrelation of the signal on the transmission signal line, a significant detection signal is output in the double talk state and the transmission state, while the reception state and the silent state are output. A transmission voice detector that outputs a non-significant detection signal at times, and provided on a reception signal line,
Based on the signal on the signal line, the receiving voice detector outputs a significant detection signal in the double talk state and the receiving state, and outputs a non-significant detection signal in the transmitting state and the silent state. A call state determiner that determines a call state based on detection signals from a transmission sound detector and a reception sound detector; and a coefficient control unit that controls updating of a coefficient of an adaptive filter based on the determined call state. And provided.

In this case, the echo elimination components composed of the adaptive filter and the adder are multiplexed, and the coefficients in the adaptive filters of the respective echo elimination components are created by the coefficient control unit at different times. A power comparator that determines a post-removal signal to be a transmission signal in the double talk state by comparing the power of the post-removal signal output from the unit, and selects a post-removal signal determined by the power comparator in the double talk state as a transmission signal. Transmission signal selecting means for selecting a signal after removal by a predetermined echo removal component in a reception state as a transmission signal, and selecting a signal captured by a microphone as a transmission signal in a transmission state and a silent state. preferable.

In this preferred embodiment, the coefficients of the adaptive filters in each echo cancellation component are fixed in the double talk state, except for a predetermined echo cancellation component whose output signal is selected as a transmission signal in the receiving state. Is more preferable.

[0023]

In the present invention, the transmission voice detector outputs a significant detection signal in the double talk state and the transmission state based on the autocorrelation of the signal on the transmission signal line,
A non-significant detection signal is output in the receiving state and the silent state, and the receiving voice detector outputs a significant detection signal in the double talk state and the receiving state based on the signal on the receiving signal line, An insignificant detection signal is output in the transmission state and the silent state. Then, the call state determiner determines a call state based on the detection signals from the transmitted voice detector and the received voice detector, and the coefficient control unit determines a coefficient of the adaptive filter based on the determined call state. Control updates.

Here, as the signal on the transmission signal line, various combination signals of a transmission signal component, an echo component, and a noise component can be considered. Since the detection operation is performed based on the correlation, the detection accuracy is good. Note that a receiver using autocorrelation can be applied to the received voice detector.

As described above, since the transmitted voice detector and the received voice detector are provided, the silent state can be detected independently, and the operation suitable for the silent state can be executed. Further, a microphone used only for the call state determination is not required.

In order to properly control the echo canceling operation in accordance with the four communication states, the echo canceling unit composed of the adaptive filter and the adder is multiplexed, and the coefficient in the adaptive filter of each echo canceling unit is changed by a coefficient. A power comparator that determines a post-removal signal to be used as a transmission signal in a double-talk state by comparing powers of the post-removal signals output from the respective echo rejection components; In the state, the post-removal signal determined by the power comparator is selected as the transmission signal, and in the reception state, the post-removal signal by the predetermined echo removal component is selected as the transmission signal, and the microphone is captured in the transmission state and the silent state. It is preferable to provide transmission signal selection means for selecting the signal as a transmission signal.

Here, in the double talk state, each echo canceling component except for a predetermined echo canceling component whose output signal is selected as a transmission signal in the receiving state so that stable coefficients in the past can be used. Is preferably fixed in the double talk state.

[0028]

FIG. 1 shows a conceptual configuration of an embodiment of the present invention. First, the conceptual configuration and operation of the embodiment will be described with reference to FIG. FIG. 1 does not exactly show the configuration of the embodiment, but is conceptually shown so as to clarify its largest feature. The exact configuration of the embodiment will be apparent through the description of FIG.

In this embodiment, the basic configuration and operation for removing echoes are the same as in the prior art. That is, the reception signal is provided to the speaker 2 to be sounded, and is input to the adaptive filter 4. The received voice (echo) y emitted from the speaker 2 goes around the omnidirectional microphone 1, is captured by the microphone 1, and is given to the adder 3 as a subtracted input. When the reception signal is input, the adaptive filter 4 performs a convolution operation with the filter coefficient (estimated value of the echo path gain) at that time, and supplies the operation value to the subtraction input terminal of the adder 3 as the pseudo echo signal Y. Thus, the adder 3 outputs a signal e from which the echo component has been removed from the signal captured by the microphone 1, which is transmitted and supplied to the coefficient control circuit 6. The coefficient control circuit 6 updates the coefficients of the adaptive filter 4 based on the signal e output from the adder 6 during the period in which the update of the coefficients of the adaptive filter 4 is permitted.

However, the configuration for controlling whether or not to update the coefficient of the adaptive filter 4 is different from the conventional configuration.

That is, a transmitted voice detector 8, a received voice detector 10, and a call state determiner 11 are provided. The transmission voice detector 8 detects whether or not there is a transmission signal based on the output signal e from the adder 3 by using autocorrelation, and provides a detection result to the call state determination unit 11. The received voice detector 10 detects whether or not there is a received signal using the autocorrelation, and gives the detection result to the call state determiner 11. The call state determination unit 11 determines whether the call state is a double talk state, a transmission state, a reception state, or a silent state based on the detection results of the transmission sound detector 8 and the reception sound detector 10. The coefficient control unit 6 is in the receiving state.
, The coefficient updating operation by the coefficient control unit 6 is permitted or prohibited in the double talk state, and the coefficient updating operation by the coefficient control unit 6 is prohibited in the other state.

Conventionally, a silent state cannot be determined by itself, but has been determined as a receiving state.

Transmission voice detector 8 and reception voice detector 10
Detects the presence of a transmission signal or a reception signal using autocorrelation as described above. This detection method will be described in detail.

Assuming that a speaker signal (transmitting signal or receiving signal) at the sampling period i is ai, the autocorrelation function S (j) when shifted by j times the sampling period is expressed by the following equation.

S (j) = Σai · ai + j / Σai · ai (9) where the sum of the two is from 1 to N (the period longer than the maximum possible pitch of the speaker signal ai). Value corresponding to). The transmitted voice detector 8 and the received voice detector 10 change a plurality of autocorrelation functions S (j) by changing a parameter j within a predetermined range centered on a value corresponding to an average pitch of speaker signals. Ask. Then, for example, a maximum value Smax (j) of the plurality of autocorrelation functions S (j) is obtained,
The maximum value Smax (j) is compared with a predetermined threshold value. If the maximum value Smax (j) is large, it is determined that a speaker signal (transmitted signal or received signal) exists, and the maximum value Smax (j) is determined.
Is smaller, it is determined that there is no speaker signal (transmitted signal or received signal).

As is well known, an audio signal has periodicity, and the autocorrelation function corresponding to the period has a very large value. On the other hand, noise is generally random, and a plurality of autocorrelation functions obtained by changing the parameter j take small values. Therefore, it is possible to distinguish between the noise and the speaker signal by using the autocorrelation function.

Here, the signal input to the transmission voice detector 8 changes depending on the call state or switching of the call state. For example, immediately after changing from the transmission state to the double talk state, when the call state determination unit 11 continuously determines the transmission state, a signal in which the echo component of the reception signal is added to the transmission signal is detected by the detector 8. Is input to Both the transmitted signal and the received signal have a high autocorrelation, and the signal obtained by adding the echo component of the received signal to the transmitted signal also slightly decreases in value, but the autocorrelation is still high because the power of the transmitted signal is large. It is. Also, for example, immediately after the change from the silence state to the receiving state, when the call state judging unit 11 still judges that the state is the silence state, a signal to which the echo component of the receiving signal is added is input to the detector 8. Since there are strictly many echo paths and the power ratio between the echo component and the noise component is small, the autocorrelation of the echo component is small even if the received signal has a pitch characteristic. Based on the same consideration, the transmitted voice detector 8 can easily output a detection signal in the transmitted state and the double talk state by appropriately selecting the threshold.

Further, even when a plurality of near-end speakers speak simultaneously, the synthesized speech has periodicity, and the transmitted voice detector 8 can detect the presence of the transmitted signal.

On the other hand, since the received voice detector 10 has only two states, that is, whether the received voice signal is input or not, the determination can be made easily.

A known method, for example, an algorithm such as a learning identification method can be applied to the method of updating the tap coefficients executed by the coefficient control unit 6 described above. FIG. 3 shows an example of a change in the amount of echo removal when there is only a received signal. Note that this is an example in which the echo removal amount is defined as 10 · log (input power of the adder 3 / output power of the adder 3). A constant value v1 where the echo consumption is constant from the start-up from the silent state
Until the tap coefficient is calculated, the tap coefficients are updated by performing the sequential processing by a known algorithm such as a learning identification method.
Even after the echo consumption reaches a value near the constant value v1, the tap coefficients are updated every sampling period.

FIG. 4 shows the components of the adder 3 and the adaptive filter 4 according to this embodiment in detail. In FIG. 4, the adaptive filter portion of this embodiment is adapted to generate two types of pseudo echo signals Y1 and Y2 simultaneously or selectively, which is also a feature of this embodiment.

In order to generate the first pseudo echo signal Y1, the delay element group 55 (551-55 (n-1)), the coefficient multiplier group 54 (541-54n) and the adder group 52 (5
21 to 52 (n-1)). A switch 6 for opening and closing the path of the coefficient given to the coefficient multiplier group 54
3 and a switch 62 for opening and closing the output path of the first pseudo echo signal.

In order to generate the second pseudo echo signal Y2, the delay element group 55 and the coefficient multiplier group 53 (531-53)
n) and an adder group 51 (511-51 (n-1)) are provided, and the delay element group 55 is common to the configuration for generating the first pseudo echo signal. Also, the second pseudo echo signal Y
In order to stop the occurrence of 2, the switch group 56 (561 to 561)
56n) are provided.

The switch groups 54 and 56 and the switch 6
The switches 2 and 63 and the switches 60 and 61 described below are controlled by the coefficient control unit 6 and / or the power comparator 59 described later.

A received signal (received signal sequence data) from a far-end speaker is input to one sampling period delay elements 551 to 55 (n-1) connected in cascade, and one sampling extracted from n taps. Each tap data having a different timing for each cycle is input to the corresponding variable coefficient multiplier 541 to
54n. Variable coefficient multipliers 541 to 54n
Multiplies each input tap data by the coefficient set at that time, and adds the corresponding tap data to the corresponding adders 521 to 52 (n−
Give to 1). The adders 521 to 52 (n-1) are connected in cascade, and the sum of tap data multiplied by a coefficient is obtained by the entire adder group 52. The sum data output from -1) is supplied as a first pseudo echo signal Y1 to the echo removing adder 58 as a subtraction input when the switch 62 is closed. The coefficients of the variable coefficient multipliers 541 to 54n are updated by the coefficient control unit 6 at each sampling cycle when the switch 63 is closed (a part of the double talk state and the receiving state).

Switches 561 to 5 that open and close in conjunction with each other
6n is closed, the delay element group 55
The tap data extracted at different timings by one sampling period and extracted from the n taps of each of the taps is supplied to the corresponding one of the variable coefficient multipliers 531 through 5n through the corresponding switch 561 through 56n.
3n. Variable coefficient multipliers 531 to 53n
Multiplies each input tap data by the coefficient set at that time, and adds the corresponding tap data to the corresponding adders 511 to 51 (n−
Give to 1). The adders 511 to 51 (n-1) are connected in cascade, and the sum of the tap data multiplied by the coefficient is obtained by the entire adder group 51. -1) is used as the second pseudo echo signal Y2 as the total sum data output from the echo removing adder 5.
7 as a subtraction input.

Here, in the receiving state, the coefficients are transferred to the variable coefficient multipliers 531 to 53n from the variable coefficient multipliers 541 to 54n for forming the first pseudo echo signal Y1. This transfer cycle is a predetermined time T
It is twice as long as 2. In another call state, each of the variable coefficient multipliers 531 to 53n maintains the set coefficient.

FIG. 5 is an explanatory diagram showing the relationship between the detection response time T1 of the transmission voice detector 8 and the predetermined time T2 defining the transfer cycle. This shows the case where migration has occurred. Even if the state changes to the double talk state at the time point B, the transmitted voice detector 8 cannot detect it immediately, and detects it for the first time at a time point A which is slightly delayed. A predetermined time is set to a time T2 longer than the detection response time T1 (which slightly changes depending on the signal content), and in the receiving state, the variable coefficient multiplier coefficients 541 to 54n are set every 2 × T2 times this time. The transfer operation is performed so as to set the coefficient to the variable coefficient multipliers 531 to 53n. That is, in the period T2 (k + 1) before the transfer operation, the period T2 (k + 1)
The coefficients transferred from the variable coefficient multipliers 541 to 54n in -1) are set in the variable coefficient multipliers 531 to 53n. Therefore, the variable coefficient multiplier 5 is switched by state switching or the like.
Even if the coefficients of 41 to 54n are disturbed, the variable coefficient multipliers 531 to 53n
The coefficient of 53n can be stabilized.

As described above, the two types of pseudo echo signals Y
The reason why 1, Y2 is generated is that even in a call state where at least one of the coefficients is not updated, an attempt is made to eliminate echo by a good pseudo echo signal.

The signal P captured by the microphone 1 is subtracted by the echo removing adder 58 to which the first pseudo echo signal Y1 is input and the echo removing adder 57 to which the second pseudo echo signal Y2 is input. Provided as an input, the echo removing adder 58 or 57 subtracts the first or second pseudo echo signals Y1 and Y2 from the captured signal P and outputs the subtracted signal as the first and second transmission candidate signals. Output as E1 and E2.

The first and second transmission candidate signals E1 and E2 are converted into transmission signals via the switch 60 or 61 which is closed at that time, and supplied to the transmission line and the transmission voice detector 8. .

The first and second transmission candidate signals E1 and E2 are supplied to a power comparator 59. The power comparator 59 calculates the average power of the first and second transmission candidate signals E1 and E2, and calculates a value obtained by adding an offset to the average power of the first transmission candidate signal and the average power of the second transmission candidate signal. Is compared with, and according to the comparison result,
The switches 60, 61, and 63 are controlled to open and close. Note that the switch control by the power comparator 59 functions effectively only when the result of the determination by the call state determiner 11 is a double talk state.

Hereinafter, the operation of the echo canceller according to the embodiment having the above-described configuration will be described by dividing the operation into a case of a talking state.

(A) Receiving state In a receiving state in which there is only a receiving signal, the receiving voice detector 10 gives a significant detection signal to the calling state determiner 11,
The transmission voice detector 8 does not output a significant detection signal.

Thus, the call state determiner 11 determines that the telephone is in the receiving state, and supplies the result of the determination to the coefficient control unit 6. At this time, the coefficient control unit 6 closes the switches 61 to 63 and opens the switch 60 and the switch group 56. Therefore, in the receiving state, only the first pseudo echo signal Y1 is formed and supplied to the adder 58, and the microphone 1
The first pseudo echo signal Y1 is subtracted from the captured signal P from the first to remove the echo, and the signal E1 after removal (including only the removal residual) is transmitted as a transmission signal. The transmission voice detector 8 to which the transmission signal E1 is given continuously outputs an insignificant detection signal.

The signal E1 after removal including only the removal residual is:
Provided to the coefficient control unit 6, the coefficient control unit 6 updates the coefficients of the variable coefficient multipliers 541 to 54n every sampling cycle. The coefficients of the variable coefficient multipliers 541 to 54n are
The variable coefficient multiplier 531 to the variable coefficient multiplier 531 every two times the predetermined time T2
The coefficient is transferred to 53n, and the coefficient previously set in the variable coefficient multipliers 541 to 54n is set in the multipliers 531 to 53n.

As described above, in the receiving state, the first
The echo removing operation using the pseudo echo signal Y1 and the coefficient updating operation are repeatedly executed.

(B) Double Talk State In the double talk state in which both the transmission signal and the reception signal are present, the reception voice detector 10 gives a significant detection signal to the communication state determination unit 11 and the transmission voice detector 8 Also gives a significant detection signal to the call state determination unit 11. Although the echo signal is included in the signal P captured by the microphone 1, the transmitted voice detector 8 performs a detection operation using the autocorrelation function as described above, and therefore outputs a significant detection signal.

Thus, the call state determination unit 11 determines that the state is a double talk state, and provides the determination result to the coefficient control unit 6. The coefficient control unit 6 opens the switch 63 immediately after the determination as the double talk state to prevent the coefficient update from being performed, and closes the switch 62 and the switch group 56 to perform the first and second pseudo-controls. Echo signal Y
Both 1 and Y2 are supplied to the adders 58 and 57. In the double talk state, the switch 60
And 61 are delegated to the power comparator 59. When the initial state is exceeded, the control of the switch 63 is also delegated to the power comparator 59.

Therefore, immediately after switching to the double talk state, the first transmission candidate signal (removal residual and transmission) is obtained by subtracting the first pseudo echo signal Y1 from the signal P captured by the microphone 1 and removing the echo. E1)
And a second transmission candidate signal E2 (which consists of a removed residual and a transmission signal component) obtained by subtracting the second pseudo echo signal Y2 from the signal P captured by the microphone 1 and removing the echo, to the power comparator 59. Is entered. The power comparator 59 compares the average power of the first transmission candidate signal E1 with a value obtained by adding an offset to the average power of the second transmission candidate signal E2. When the former is larger, it is determined that the removal residual component included in the first transmission candidate signal E1 is larger, and the switch 60 is closed, and the switch 61 is closed.
And the second transmission candidate signal E2 having the smaller removal residual component is selected as the transmission signal. Conversely, when the latter is large, it is determined that the removal residual component included in the first transmission candidate signal E1 is smaller, and the switch 61 is closed and the switch 60 is opened to remove the removal residual component. A first transmission candidate signal E1 having a small component is selected as a transmission signal. When the first transmission candidate signal E1 is thus selected, the switch 63 is closed and the variable coefficient multiplier 5
The coefficients 41 to 54n are updated. However, the coefficient update of the variable coefficient multipliers 531 to 53n is not performed.

As described above, the switch 63 is opened in the initial state, but thereafter, the switch 63 is closed or opened according to the comparison result of the power comparator 59.

It is to be noted that the magnitude is determined by adding the offset because the coefficient relating to the latest estimated value of the impulse response of the echo path is set in the variable coefficient multipliers 541 to 54n. This is because E1 is given some priority over the second pseudo echo signal E2. Also, the first
The transmission candidate signal E1 is mainly composed of a transmission signal component in the double talk state, but the coefficient of the variable coefficient multipliers 541 to 54n may be updated based on this signal in the following case. For reasons. This is because it is possible to instantaneously improve the echo removal characteristic before the state is determined as the receiving state by the change from the double-talk state to the receiving state. In addition, when the double-talk period becomes long, the variable coefficient multiplier 531 is used. This is because the coefficients of 係数 53n and 541１〜54n are no longer valid and updating in this way may provide a better output signal.
This is to prevent the howling of alternately and repeatedly selecting the transmission candidate signal E1 and the second transmission candidate signal E2 in a short cycle.

Since most of the transmission signals selected and formed in the double talk state are transmission signals, the transmission voice detector 8 continuously outputs a significant detection signal during this state.

As described above, in the double talk state, the echo removing operation using the first pseudo echo signal Y1 and the echo removing operation using the second pseudo echo signal Y2 are performed in parallel. A signal with good removal is selected to be a transmission signal, and a coefficient updating operation is controlled according to the selected content.

(C) Transmission state In a transmission state in which only a transmission signal is present, the reception voice detector 10 gives a non-significant detection signal to the call state determination unit 11, and the transmission voice detector 8 outputs a significant signal. The detection signal is provided to the call state determiner 11.

Thus, the call state determination unit 11 determines that the state is the transmission state, and supplies the determination result to the coefficient control unit 6. At this time, the coefficient control unit 6 closes only the switch 61 and opens the other switches 60, 62, 63 and the switch group 56. Therefore, the first pseudo echo signal Y1 input to the adder 58 is 0, and the signal P (only the transmission signal) captured by the microphone 1 becomes a transmission signal as it is via the adder 58 and the switch 61. . Therefore, the transmission voice detector 8 continuously supplies a significant detection signal to the call state determination unit 11. In the transmitting state, there is no echo component in the received signal, so there is no need to perform echo removal.

Here, since all the switches other than the switch 61 are open, the coefficient updating operation is not performed, and the variable coefficient multipliers 531 to 53n, 541 to 541 are opened by opening the other switches. The coefficient held at 54n is protected without being affected by noise or a received signal.

As described above, in the transmission state, the captured signal P of the microphone 1 is directly used as a transmission signal without performing echo cancellation, and the variable coefficient multiplier 531 is used.
The coefficient update operation for .about.53n, 541-54n is stopped.

(D) Silence state In a silence state in which neither a transmission signal nor a reception signal exists,
The received voice detector 10 provides a non-significant detection signal to the call state determiner 11, and the transmitted voice detector 8 also provides a non-significant detection signal to the call state determiner 11.

Thus, the call state determiner 11 determines that there is no sound, and supplies the determination result to the coefficient control unit 6. At this time, the coefficient control unit 6 closes only the switch 61 and opens the other switches 60, 62, 63 and the switch group 56. That is, the same control as in the transmission state is performed. Therefore, the signal captured by the microphone 1 (only noise)
P becomes a transmission signal as it is via the adder 58 and the switch 61. Therefore, the transmission voice detector 8 continuously supplies the insignificant detection signal to the call state determination unit 11. Even in the silent state, there is no need to perform echo removal because there is no echo component of the received signal. Also in this case, the switch 61
Since the other switches are all open, the coefficient updating operation is not performed, and the variable coefficient multipliers 531 to 53n, 54 are opened by opening the other switches.
The coefficients held in 1 to 54n are protected without being affected by noise or a received signal.

In the prior art described above, the silent state and the receiving state are treated the same (although the silent state is also determined to be the receiving state). In this embodiment, the silent state is processed in the same manner as the transmitting state. I have. Both the silent state and the transmission state have a common feature that the signal P captured by the microphone 1 has no echo component, and therefore can be handled similarly.
When the silent state is handled in the same manner as the receiving state, there is a disadvantage that the coefficient is updated, so that it is preferable to use this embodiment.

As described above, in the silent state, the captured signal P of the microphone 1 is directly used as a transmission signal without performing echo removal, and the variable coefficient multiplier 531 is used.
The coefficient update operation for .about.53n, 541-54n is stopped.

According to the above embodiment, the received voice detector 10
And the transmitted voice detector 8 detects the presence of the received signal or the transmitted signal using the autocorrelation function, and the communication state determination unit 11 determines the communication state based on these detection signals. Regardless of this, the call state can be determined with high accuracy, the echo removal operation can be appropriately controlled, and the echo removal characteristics can be improved.

Further, according to the above-described embodiment, in the double talk state, the transmission signal is determined by selecting from two types of post-removal signals obtained by performing echo removal using two types of pseudo echo signals. Even when the coefficient shifts from the receiving state or the transmitting state to the double talk state or the coefficient is disturbed during the double talk state, the echo can be removed by the pseudo echo signal using the stable past coefficient. The echo removal characteristics can be improved.

Further, according to the above-described embodiment, a microphone used only for determining the call state is not required, and the hands-free telephone can be downsized, and the positional flexibility of the microphone 1 provided can be increased.

Further, according to the above-described embodiment, the silent state is distinguished from other call states, and the updating of the adaptive filter coefficients is stopped in the silent state. It is possible to prevent the coefficient from being unnecessarily updated by noise in a silent state.

In the above embodiment, the adaptive filter has a double configuration. However, the present invention can be applied to a single configuration having an adaptive filter. The present invention can also be applied to

In the above embodiment, both the received voice detector 10 and the transmitted voice detector 8 detect the presence of a received signal or a transmitted signal using an autocorrelation function. For the received voice detector 10 in which the combination of the components is simple, a device that does not use the autocorrelation function may be applied. For example, the power of the input signal may be obtained, and the power may be compared with a detection threshold for detection.

Further, in the above embodiment, the coefficients of the variable coefficient multipliers 541 to 54n may be updated in the double talk state, but the coefficient updates of the variable coefficient multipliers 541 to 54n are completely updated in the double talk state. It may be prohibited.

Further, the variable coefficient multipliers 541 to 54n in the reception state are changed to variable coefficient multipliers 531 to 53n.
The coefficient transfer cycle to the variable coefficient multipliers 531 to 53n is not limited to the above-described embodiment as long as the coefficients set in the variable coefficient multipliers 531 to 53n can be stabilized. Further, a large-capacity register may be provided in the coefficient control unit 6 to transfer the data from the coefficient control unit 6. In this case, the transfer cycle is set to one sampling cycle (however, the transferred coefficient is considerably earlier). Can also be.

[0081]

As described above, according to the present invention, a significant detection signal is provided in the double talk state and the talk state based on the autocorrelation of the signal on the talk signal line. And a transmitting voice detector that outputs a non-significant detection signal in a receiving state and a silent state, and a double talk state and a receiving state provided on a receiving signal line based on the signal on the signal line. And a non-significant detection signal is output in the transmission state and the silent state, and a detection signal from the transmission sound detector and the reception sound detector is output. A call state determiner that determines a call state based on the call state, and a coefficient control unit that controls coefficient update of the adaptive filter based on the determined call state. The Can be determined with high accuracy regardless of the presence of the figure can be realized compact echo canceler can be improved echo cancellation accuracy.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a conceptual overall configuration of an embodiment.

FIG. 2 is a block diagram showing a conventional configuration.

FIG. 3 is a diagram for explaining a coefficient updating method according to the embodiment;

FIG. 4 is a block diagram illustrating a detailed configuration around an adaptive filter and an adder according to the embodiment.

FIG. 5 is an explanatory diagram of a detection response time of a transmission signal and a time difference between both coefficients in the embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Microphone, 2 ... Speaker, 3, 57, 58 ... Adder, 4 ... Adaptive filter, 6 ... Coefficient control part, 8 ... Transmission voice detector, 10 ... Reception voice detector, 11 ... Talk state determination device, 51, 52 ... adder group, 53, 54 ... variable coefficient multiplier group, 55 ... delay element group, 56 ... switch group, 59 ... power comparator, 60-63 ... switch.

Continuation of front page (56) References JP-A-64-27325 (JP, A) JP-A-3-80628 (JP, A) JP-A-6-14100 (JP, A) JP-A-62-24726 (JP) (A) JP-A-1-126839 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) H04B 1/76-3/44 H04B 3/50-3/60 H04B 7/005 -7/015 H04M 1/00 H04M 1/24 H04M 1/58-1/62 H04M 1/66-1/78

Claims (3)

(57) [Claims] An echo canceller provided in a hands-free telephone for removing an echo signal from a speaker to a microphone using a pseudo echo signal formed by an adaptive filter, the echo canceller being provided on a transmission signal line. Based on the autocorrelation of the signal on the signal line, outputting a significant detection signal in the double talk state and the transmission state, and outputting a non-significant detection signal in the reception state and the silent state. A voice detector, provided on a reception signal line, and outputs a significant detection signal in a double talk state and a reception state based on a signal on the signal line, and is insignificant in a transmission state and a silent state based on the signal on the signal line. A receiving voice detector for outputting a detection signal, and a communication state for determining a call state based on detection signals from the transmitting voice detector and the receiving voice detector. An echo canceller comprising: a talk state determiner; and a coefficient control unit that controls updating of the coefficient of the adaptive filter based on the determined call state. 2. The apparatus according to claim 1, wherein the adaptive filter and the adder are multiplexed with an echo removing component, and the coefficients in the adaptive filters of the respective echo removing components are created by the coefficient control unit at different times. A power comparator that determines a post-removal signal to be used as a transmission signal in the double talk state by comparing the power of the post-removal signal output from each echo removal component, and a post-removal signal determined by the power comparator in the double talk state. Transmission signal selection for selecting a transmission signal, selecting a signal after removal by a predetermined echo removal component in a reception state as a transmission signal, and selecting a signal captured by the microphone as a transmission signal in a transmission state and a silent state. The echo canceller according to claim 1, further comprising: 3. The adaptive filter coefficient in each of the echo canceling components is fixed in the double talk state except for the predetermined echo canceling component in which an output signal is selected as a transmission signal in a receiving state. The echo canceller according to claim 2, wherein: