JP2003125070A - Call signal processing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To apply a call signal processing for executing specific signal processing to a call signal to a diversified terminal, without making the configuration complex and to stably and reliably avoid the generation of echoes and howling, in a terminal apparatus for transmitting and receiving call signals via a full duplex transmission line. SOLUTION: The call signal-processing apparatus comprises an acoustic conversion means 11 for converting a call signal that is received via a transmission line 10 to a acoustic signals, an acoustic coupling monitoring means 13 for converting voice that is generated by a talker and for monitoring the degree of acoustic coupling to a signal-conversion means 12, where the relative position to the acoustic conversion means 11 may be changed, and a reverberation- inhibiting means 14 for relaxing the variation in a gain that is generated, according to the increase/decrease of the degree of acoustic coupling that is monitored by the acoustic coupling-monitoring means 13, in an echo path to the transmission line 10 via the signal-converting means 12 from the acoustic conversion means 11.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a call signal processing device for maintaining call quality by performing a predetermined signal processing on a call signal in a terminal device for transmitting and receiving a call signal through a full-duplex transmission line. Regarding

[0002]

2. Description of the Related Art In recent years, a terminal accommodated in a mobile communication system or other telephone network is provided with a function capable of providing various services. For example, a received call signal is transmitted through a speaker to an operator. Hands-free call service is possible by being transmitted to. Figure 5
It is a figure which shows the structural example of the terminal device which can talk by hands-free. In the figure, a microphone 71 is an A / D converter 7
2 to the upstream voice input of the voice switch 73, and the upstream voice output of the voice switch 73 is connected to the transmitter / receiver 7
4 modulation inputs. The transmission / reception unit 74 is connected to a predetermined mobile communication network 75 via a wireless line, and the demodulation output of the transmission / reception unit 74 is connected to the downstream voice input of the voice switch 73. The downstream audio output of the audio switch 73 is connected to the speaker 77 via the D / A converter 76.

The voice switch 73 is composed of the following elements. -Suppression processing unit 81u whose input and output are connected to the above-described upstream audio input and upstream audio output, respectively-Suppression processing unit 81d whose input and output are connected to the above-described downstream audio input and downstream audio output, respectively The suppression control unit 8 having two outputs respectively connected to the control inputs of the suppression processing unit 81u and the suppression processing unit 81d.
2 • Level detection unit 83u whose input and output are connected to the above-mentioned upstream voice input and one input of the suppression control unit 82, respectively • Input and output to the above-mentioned downstream voice input and the other input of the suppression control unit 82, respectively Level detection unit 83d to which output is connected In the terminal device having such a configuration, the upstream call signal output by the microphone 71 according to the voice (sending sound) emitted by the speaker is output by the A / D converter 72. It is converted into a digital upstream call signal.

The transmitter / receiver 74 also receives a digital downlink call signal indicating a voice (received sound) received from the caller of such a speaker via the mobile communication network 75 and made by the caller. Output. Level detector 83
u and 83d measure the levels Lu and Ld of the digital upstream call signal and the digital downstream call signal described above.

The suppression control unit 82 controls these levels Lu,
The suppression processing unit 81 is as follows according to the combination of Ld.
The operation of u and 81d is controlled. When both the levels Lu and Ld are larger than the predetermined threshold value, the suppression processing unit 81u is instructed to prevent the transmission / reception unit 84 from transmitting the upstream digital speech signal, and the downlink digital speech signal to the D / A converter 76 is transmitted. The suppression processing unit 81d is instructed to permit transmission.

When the level Lu is below the above-mentioned threshold value, regardless of the value of the level Ld, the suppression processing unit 81u is instructed to block the transmission of the upstream digital call signal to the transmission / reception unit 84, and D / Suppression processing unit 8 for permitting transmission of the downstream digital call signal to the A converter 76
Command 1d. When the level Lu exceeds the above threshold value and the level Ld falls below the threshold value, the suppression processing unit 18u is instructed to permit transmission of the upstream digital call signal to the transmission / reception unit 84, and the D / A converter 76 is also instructed. The suppression processing unit 81d is instructed to prevent the transmission of the downlink digital call signal.

That is, the level Ld of the downlink digital call signal is large enough to exceed the above-mentioned threshold value, or the level Ld and the level Lu of the uplink digital call signal Lu.
If both are smaller than this threshold value, the transmission of the call signal to the mobile communication system 75 (call partner) is suspended. Therefore, even when the speaker 77 and the microphone 71 are located close to each other, howling and echo generation due to the close acoustic coupling between the speaker 77 and the microphone 71 are avoided.

As a prior art related to the present invention, for example, as disclosed in Japanese Patent Application Laid-Open No. 8-9005, in a folding type portable terminal, between a member in which a speaker and a microphone are arranged, respectively. There is a technique (hereinafter, simply referred to as "prior art") in which the filtering characteristic of the echo canceller is changed according to the angle formed by the hinge interposed in the.

[0009]

By the way, in the above-mentioned conventional example, either one of the transmission and the reception is restricted regardless of the degree of acoustic coupling between the speaker 77 and the microphone 71, or By suppressing the level of a call signal transmitted or received, howling and echo are prevented from occurring. Therefore, the background sound of the talker is not necessarily transmitted to the listener, and the mood based on such background sound is impaired or the accurate identification of the situation based on the sound is impeded, so that Service quality was poor.

However, in recent years, various added values have been incorporated into communication terminals, and aesthetic designs have been adopted. For example, foldable type, flipper type and telescopic type terminals used for mobile communication services have been adopted. As described above, many terminals in which the distance (acoustic coupling) between the speaker 77 and the microphone 71 described above can be appropriately changed by the operator have been put into practical use.

Also, during a hands-free call through such a terminal, the received sound is output from the speaker 77 at a level large enough to be heard by a speaker located far from the terminal. The voice emitted by this speaker is amplified to a level large enough to be transmitted to the other party at a sufficient level. Therefore, during a hands-free call, the speaker 77 and the microphone 7
There is a high possibility that the suppression of echo and the prevention of howling cannot be sufficiently achieved because the acoustic coupling with 1 is dense.

Further, in the above-mentioned prior art, of the filtering processes performed by the echo canceller based on the adaptive algorithm, the form of signal processing performed only on the speech signal to be transmitted is variable according to the above-mentioned angle. . However, the accuracy of estimation and identification of echo to be canceled decreases as the background noise superimposed on the speech signal increases, and this echo is not always sufficiently suppressed.

It is an object of the present invention to provide a speech signal processing device which can be applied to various terminals without complicating the structure and which can stably and accurately avoid the occurrence of echo and howling. .

[0014]

FIG. 1 is a principle block diagram of a speech signal processing apparatus according to the present invention. Claim 1
In the invention described in (1), the acoustic coupling monitoring means 13 includes
Acoustic conversion means 11 for converting a call signal received via 0 into an acoustic signal, and a voice signal emitted by a talker into a call signal to be transmitted to the transmission line 10, and this acoustic conversion means 11 To monitor the degree of acoustic coupling with the signal conversion means 12 whose position relative to can be changed. The echo suppressing means 14 is provided in the echo path from the acoustic converting means 11 to the transmission path 10 via the signal converting means 12 described above.
The fluctuation of the gain caused by the increase or decrease of the acoustic coupling degree monitored by the acoustic coupling monitoring means 13 is mitigated.

That is, a substantial variation of the acoustic coupling degree between the acoustic converting means 11 and the signal converting means 12 is quickly compressed or suppressed by the variable gain that can be realized without the intervention of the adaptive algorithm. . Therefore, the occurrence of howling and echo due to the excessive gain of the echo path and the deterioration of the call quality due to the excessive gain are avoided with high accuracy without complicating the configuration. To be done.

According to the second aspect of the invention, the echo suppressing means 14 has the gain of the acoustic converting means 11 and / or the signal converting means 12 which are updated without interposing the echo suppressing means 14 and / or one of them. , Alleviating a variation in gain occurring in the echo path in response to an increase or decrease in all or a part of the level of the speech signal received via the transmission path 10 and the level of the voice uttered by the speaker.

That is, the level of the call signal at which the call is to be transmitted or received by the operator of the communication apparatus provided with the call signal processing apparatus according to the present invention and / or the other party of the call of the operator. Even when the value is changed, the occurrence of howling or echo and the deterioration of the call quality are avoided. According to the third aspect of the invention, the adaptive filtering means 16 applies a signal based on an adaptive algorithm to the acoustic signal obtained by the acoustic converting means 11 and / or the speech signal obtained by the signal converting means 12 based on the adaptive algorithm. Apply processing.

These fluctuations in the level of the acoustic signal and the speech signal are, prior to the above-described signal processing, "to the extent that howling and echo are prevented from occurring due to the increase or decrease in the gain of the echo path". Suppressed. Therefore, the signal processing procedure to be performed by the adaptive filtering means 16,
The adaptive filtering means 16 and the echo suppressing means 14 are subject to no severe restrictions on algorithms and accuracy.
In cooperation with and, it is possible to avoid or suppress the occurrence of howling and echo with high accuracy.

According to the fourth aspect of the invention, the adaptive filtering means 16 restarts the signal processing based on the adaptive algorithm when the acoustic coupling degree monitored by the acoustic coupling monitoring means 13 changes. That is, after the time when the degree of acoustic coupling between the acoustic converting means 11 and the signal converting means 12 changes, an error caused by the change in the acoustic coupling degree in the process of signal processing performed by the adaptive filtering means 16. The occurrence of is surely avoided.

Therefore, the restrictions on the above-mentioned adaptive algorithm and the coefficients to be applied to the signal processing performed based on the adaptive algorithm are greatly relaxed. In the invention described in claim 5, the processing mode storage means 1
7 is obtained by the call conversion signal received through the transmission path 10 and the signal conversion means 12 under the individual values that the acoustic coupling degree between the sound conversion means 11 and the signal conversion means 12 can actually take. The form of the signal processing to be performed by the adaptive filtering means 16 is stored in advance in both and / or one of the call signals. The adaptive filtering means 16 applies the acoustic coupling degree monitored by the acoustic coupling monitoring means 13 to the speech signal received through the transmission line 10 and / or the speech signal obtained by the signal converting means 12. The signal processing of the form stored in the processing form storage means 17 is performed corresponding to.

That is, the form of signal processing to be performed by the adaptive filtering means 16 is maintained in a form adapted to the degree of acoustic coupling between the acoustic converting means 11 and the signal converting means 12. Therefore, the occurrence of howling and echo is avoided with higher accuracy than in the case where the form of such signal processing is constant regardless of the acoustic coupling degree described above.

In the invention of the first subordinate concept of the invention described in claim 1, the acoustic coupling monitoring means 13 comprises the acoustic conversion means 11.
The acoustic coupling degree is monitored as a conversion value corresponding to both and / or the shape and / or the size of the member that supports both the signal conversion means 12 and the signal conversion means 12. That is, the acoustic conversion means 1
The degree of acoustic coupling between 1 and the signal conversion means 12 is as long as the above-mentioned shapes and dimensions can be identified with desired accuracy and speed.
Even when both the acoustic conversion means 11 and the signal conversion means 12 are supported by a single member, they are reliably monitored.

Therefore, the flexible adaptation to the standardization of the structure and the reduction of the number of parts is secured, and the occurrence of the above-mentioned howling and echo is avoided with high accuracy. Claim 1
According to the second sub-invention of the invention described in (1), the acoustic coupling monitoring means 13 has a three-dimensional shape and / or a dimension of a solid body composed of members that individually support the acoustic conversion means 11 and the signal conversion means 12. The acoustic coupling degree is monitored as a conversion value corresponding to either one.

That is, as for the degree of acoustic coupling between the acoustic conversion means 11 and the signal conversion means 12, as long as the above-mentioned shapes and dimensions can be identified with desired accuracy and speed, these acoustic conversion means 11 and the signal conversion means. It is reliably monitored even when both 12 and 12 are supported on separate members. Therefore, the standardization of the structure and the flexible adaptation to the variety of mechanical mechanisms are secured, and the occurrence of the above-mentioned howling and echo is highly avoided.

In the invention of the third subordinate concept of the invention described in claim 1, the acoustic coupling monitoring means 13 comprises the acoustic conversion means 11.
The pilot signal which can be regarded as uncorrelated to the speech signal is superimposed on the speech signal converted into the acoustic signal described above, and the pilot signal and the speech signal obtained by the signal converting means 12 are superimposed. The degree of acoustic coupling is monitored as a cross-correlation with the pilot signal.

That is, in this way, the acoustic coupling degree is
The relative distance between the acoustic conversion means 11 and the signal conversion means 12 and not just the standard characteristics of the acoustic propagation path formed between them, but these acoustic conversion means 11 and signal conversion means 12 It is monitored as a substantive value that is flexibly adapted to the characteristics of an object or medium that is located around and between and that acts as an acoustic reflector, absorber, scatterer, or the like.
Therefore, the acoustic conversion means 11 and the signal conversion means 12
The restrictions on the characteristics of the space where the can be located and the environment are alleviated, and the occurrence of the above-mentioned howling and echo is highly accurately and stably avoided.

In the first invention related to the first and second aspects of the invention, the storage means 15 has individual values that the acoustic coupling degree between the acoustic conversion means 11 and the signal conversion means 12 can actually take. The fluctuation amount of the gain of the echo path that occurs below is stored in advance. The echo suppression unit 14 reduces the fluctuation of the gain of the echo path over the fluctuation amount stored in the storage unit 15 corresponding to the degree of acoustic coupling monitored by the acoustic coupling monitoring unit 13.

That is, the gain of the echo path is the sum of the time required for accessing the storage means 15 and the response time of the echo suppressing means 14 even if the acoustic coupling degree between the acoustic converting means 11 and the signal converting means 12 increases or decreases. It is kept constant with a certain accuracy without significant delays over equal time. Therefore, as long as the fluctuation stored in the storage unit 15 is appropriate, the occurrence of howling or echo due to the fluctuation of the gain of the echo path and the deterioration of the communication quality can be stably avoided with high accuracy.

In the second invention related to the first and second aspects of the invention, the gain storage means 15A has individual acoustic coupling degrees between the acoustic conversion means 11 and the signal conversion means 12. The echo path gain that results in a value is pre-stored. The echo suppressing unit 14 is the acoustic coupling monitoring unit 1.
The variation of the gain of the echo path is mitigated over the deviation of the gain stored in the gain storage means 15A corresponding to the degree of acoustic coupling monitored by 3.

That is, the gain of the echo path is the time required for accessing the gain storing means 15A and the response time of the echo suppressing means 14 even if the acoustic coupling degree between the acoustic converting means 11 and the signal converting means 12 increases or decreases. It is kept constant with a certain accuracy without delay for a time significantly exceeding the sum of Therefore, the gain storage means 15A
As long as the gain stored in is appropriate, the occurrence of howling or echo due to the variation of the gain of the echo path and the deterioration of the communication quality are stably avoided with high accuracy.

[0031]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings. FIG. 2 is a diagram showing the first and second embodiments of the present invention. The feature of this embodiment is that the microphone 71 and the speaker 77 are arranged in a foldable terminal, and the housing of the terminal is interlocked with the rotation of a hinge (not shown) that enables folding of the housing. The suppression amount determination unit 20 that performs a process described later according to an angle (hereinafter, referred to as an “opening degree”) θ formed by a member in which the microphone 71 and the speaker 77 are respectively arranged.
The point is that the voice switch 21 is provided instead of the voice switch 73 shown in FIG.

The voice switch 21 is constructed by providing a suppression control unit 22 having a control input connected to the output of the suppression amount determination unit 20 in place of the suppression control unit 82 shown in FIG. The operation of the first embodiment of the present invention will be described below with reference to FIG. Generally, the degree of acoustic coupling between the speaker 77 and the microphone 71 becomes denser as the distance between them becomes shorter.

The suppression amount determination unit 20 monitors the above-described opening degree θ, and, for example, as shown in FIG. 3, gives a larger amount of attenuation L to the suppression control unit 22 as the opening degree θ decreases. The suppression control unit 22 obtains, for example, two attenuation amounts L1 and L2 that satisfy the following relational expressions (1) and (2) for this attenuation amount L. L = L1 + L2 (1) L1> L2 (2) Further, the suppression control unit 22 includes the level detection units 83u and 83.
The levels Lu and Ld of the upstream digital communication signal and the downstream digital communication signal measured in parallel by d are identified, and these levels Lu and Ld and the above-described attenuation amounts L and L are identified.
The following processing is performed according to 1 and L2.

When the level Lu is larger than the level Ld with a predetermined accuracy, the gain of the suppression processing unit 81u is set to a small value over the attenuation amount L2, and the suppression processing unit 81d
The gain is set to a small value over the attenuation amount L1. When the level Lu is smaller than the level Ld with a predetermined accuracy, the gain of the suppression processing unit 81u is set to a small value over the attenuation amount L1, and the gain of the suppression processing unit 81d is set small over the attenuation amount L2. Set to the value.

When the levels Lu and Ld are substantially the same, the gains of the suppression processing units 81u and 81d are both set to a small value over the half value of the attenuation amount L. That is, even if the degree of acoustic coupling between the speaker 77 and the microphone 71 increases, the suppression processing units 81d, D
The increase in the gain of the echo path reaching the modulation input of the transmission / reception unit 74 via the A / A converter 76, the speaker 77, the microphone 71, the A / D converter 72 and the suppression control unit 81u is mitigated or suppressed.

As described above, according to this embodiment, the gains of the suppression processing units 81u and 81d determined by the suppression control unit 22 are increased without the hardware configuration being significantly complicated as compared with the conventional example. Since not only the combination of the digital call signal and the downlink digital call signal but also the actual acoustic coupling degree between the speaker 77 and the microphone 71 is varied, echo and howling can be stably avoided.

In the present embodiment, the above-mentioned attenuation amount L
The calculation procedure and algorithm to be performed by the suppression amount determination unit 20 to obtain 1 and L2 are not disclosed in detail. However, the attenuation amounts L1 and L2 may be set to any of the following values as long as the echo and the howling are avoided with desired accuracy and the deterioration of the call quality is allowed.

A constant that is uniquely determined with respect to the attenuation amount L. A constant obtained by dividing the attenuation amount L by a predetermined ratio, regardless of the attenuation amount L. The level (or difference) between the digital call signals and / or the levels of these digital call signals, and the predetermined information identified during the channel control process (for example, the form of the service to be provided). ) Variables specified based on all or part of the following) and the second embodiment of the present invention will be described below.

The feature of this embodiment is that, as shown by the broken line in FIG. 2, there is provided a level detection section 23 whose input is connected to the output of the D / A converter 76, and the output of the level detection section 23 and the level. The output of the detection unit 83u is connected to the corresponding input of the suppression amount determination unit 20. Below, FIG.
The operation of the second embodiment of the present invention will be described with reference to FIG.

The level detecting section 23 measures the level Ld1 of the downlink call signal obtained from the demodulated output of the transmitting / receiving section 74 via the voice switch 21 and the D / A converter 76.
The suppression amount determination unit 20 individually integrates the level Ld1 and the level Lu of the upstream digital speech signal measured by the level detection unit 83u individually to obtain the average values Ld1a and Lua of these levels Ld1 and Lu. The following processing is performed while seeking.

When the level Ld1 increases above the average value Ld1a over a predetermined upper limit value, the suppression processing unit has a large value over a value equal to the difference Δ1 (= Ld1−Ld1a) between the two. The attenuation amount of 81d is updated. When the level Ld1 decreases below the average value Ld1a over a predetermined upper limit value, the difference Δ between the two
The attenuation amount of the suppression processing unit 81d is updated to a small value over a value equal to 2 (= Ld1a−Ld1).

When the level Lu increases above the average value Lua over a predetermined upper limit value, the suppression processing unit has a large value over a value equal to the difference Δ3 (= Lu-Lua) between the two. Update the attenuation amount of 81u. When the level Lu decreases below the average value Lua over a predetermined upper limit value, the difference Δ4 between them
The attenuation amount of the suppression processing unit 81u is updated to a small value over a value equal to (= Lua-Lu).

Therefore, according to this embodiment, the level of the call signal is adjusted by the caller according to the sensitivity of the ear of the caller, the voice volume, the level of background noise in the place where the caller is located, and the like. Even in this case, the level of the call signal sent back to the call partner via the acoustic coupling path between the speaker 77 and the microphone 71 is appropriately maintained without becoming excessive.

In this embodiment, the level of the call signal adjusted by the caller is the average value Ld1a,
It is calculated as a deviation from Lua. However, the present invention is not limited to such a configuration, and, for example, the level L of the downlink digital call signal output from the transmission / reception unit 74.
d may be applied instead of the average value Ld1a, and the level of the upstream call signal output from the microphone 71 may be applied instead of the average value Lua.

Further, in this embodiment, the difference Δ1 to
Δ4 is appropriately calculated based on the level of the actually measured call signal. However, the present invention is not limited to such a configuration, and, for example, when the level of the received sound or the level of the transmitted sound is changed based on the caller's instruction or the procedure of channel control, these Difference Δ1 to Δ4
May be directly given by a processor or a control unit that takes the initiative in processing related to the man-machine interface or channel control.

FIG. 4 is a diagram showing the third and fourth embodiments of the present invention. The feature of the present embodiment is that an echo canceller 41 is provided between the A / D converter 72 and the D / A converter 76 and the audio switch 21, and the opening detection unit 42 is used instead of the suppression amount determination unit 20. Is provided. The echo canceller 41 is composed of the following elements.

A / D converter 72 and suppression processing unit 81u
And an adaptive filter 43 having a monitoring input connected to the output of the suppression processing unit 81d provided in the voice switch 21. The input is connected to the corresponding output of the opening degree detection unit 42,
The coefficient controller 44 whose output is connected to the coefficient input of the above-described adaptive filter 43 will be described below with reference to FIG. 4 in accordance with the operation of the third embodiment of the present invention.

The coefficient control unit 44 registers in advance a "coefficient that has been obtained in advance as a value adapted to the above-mentioned opening degree θ and should be applied to an adaptive algorithm for realizing the signal processing to be performed by the adaptive filter 43". A coefficient table (not shown). Further, the opening degree detection unit 42
Performs the same processing as the processing performed by the suppression amount determination unit 20 in the above-described first or second embodiment in cooperation with the switch 21, and performs the following processing in parallel.

It is determined whether the opening θ has changed. If the result of the determination is true, the coefficient control unit 4
4 is notified of the latest opening θ. The coefficient control unit 44 acquires the coefficient registered in the above-described coefficient table corresponding to the latest opening degree θ, and gives the coefficient to the adaptive filter 43.

That is, the echo canceller 41 is provided together with the voice switch 21, and the voice switch 21 separately prevents the acoustic coupling between the speaker 77 and the microphone 71 from being excessively dense. As described above, according to the present embodiment, the echo canceller 4 as in the conventional example is used.
As compared with the case where the 1 (adaptive filter 43) is provided alone, the range of the instantaneous values of the upstream digital call signal and the downstream digital call signal, which are the objects of calculation by the echo canceller 41, is significantly compressed.

Therefore, high accuracy and responsiveness are realized, and even if complicated echo signal processing is not performed by the echo canceller 41, the occurrence of echo and howling is stably and accurately avoided. The operation of the fourth embodiment of the present invention will be described below with reference to FIG. The feature of this embodiment lies in the procedure of the following processing performed by the opening degree detection unit 42.

The opening detection unit 42 monitors the above-described opening θ and, in parallel, performs the same processing as that of the suppression amount determination unit 20 in the above-described second embodiment to thereby detect the downlink call signal. Along with the average values Ld1a and Lua of the level Ld1 and the level Lu of the upstream digital call signal, the difference Δ
1 (= Ld1−Ld1a) and Δ3 (= Lu−Lua) are calculated.

Further, when the opening degree θ and any of the differences Δ1 and Δ3 change with a predetermined accuracy, the opening degree detecting section 42 outputs a signal which the adaptive filter 43 performs based on the adaptive algorithm. Restart the process. That is, when the opening degree θ and the difference Δ described above change, the estimation error of the pseudo echo performed based on the adaptive algorithm is quickly initialized.

Therefore, according to the present embodiment, the call quality is further improved and maintained stable. In the present embodiment, the signal processing performed by the adaptive filter 43 based on the adaptive algorithm is performed by the above-described opening degree θ and difference Δ.
It is restarted when any one of 1 and Δ3 changes with a predetermined accuracy.

However, the present invention is not limited to such a configuration. For example, as long as a desired call quality is secured or deterioration of the call quality is allowed, the opening θ
And only the difference Δ1 or Δ3 changes with a desired accuracy, the signal processing described above is restarted, or the condition under which such initialization should be performed is based on channel control or other procedures. It may be changed appropriately.

In the third and fourth embodiments described above, the coefficients to be applied in the process of signal processing to be performed by the adaptive filter 43 are registered in advance in the coefficient table described above and are appropriately selected. There is. However, the present invention is not limited to such a configuration, and for example, as long as a desired call quality is secured or deterioration of the call quality is allowed, the above-described opening degree θ and the differences Δ1, Δ3 are set. A constant coefficient may be applied regardless of the above, or a coefficient of a value calculated as a function of both or one of the opening degree θ and the differences Δ and Δ3 may be applied.

Further, in each of the above-described embodiments, the present invention is applied to the folding type terminal. However, the present invention is not limited to such a terminal, and means such as a speaker that outputs a received call signal as an acoustic signal, a microphone that converts a voice uttered by a speaker into a predetermined call signal, and the like. If the degree of acoustic coupling with the means can change, not only flipper type or telescopic type terminals, but also various communication devices and information terminals that realize voice call service via a desired network. Is applicable to.

Further, in each of the above-mentioned embodiments, the opening θ
The configuration of the hardware or software that realizes the detection of is not specifically shown. However, such opening θ,
Alternatively, monitoring of the degree of acoustic coupling between the speaker 77 and the microphone 71 is performed by, for example, the technique disclosed in the above-mentioned prior art,
Since it can be realized by applying various other known techniques and is not a feature of the present invention, description thereof will be omitted here.

The method of obtaining the degree of acoustic coupling described above is not limited to the conversion of the "relative distance or position between the speaker 77 and the microphone 71" obtained directly or indirectly. , A pilot signal that can be regarded as uncorrelated to the call signal emitted as a sound signal from the speaker 77 is superimposed on the call signal, and the pilot signal and the pilot signal superimposed on the call signal output by the microphone 71 These speaker 77 and microphone 7 are directly obtained as the cross correlation of
Substantial transfer characteristics of the sound field formed between 1 and 1 may be reflected.

Further, in each of the above-described embodiments, the gain (attenuation amount) to be set in the suppression processing units 81u and 81d.
Is determined based on the levels of the upstream digital call signal and the downstream digital call signal measured by the level detectors 83u and 83d, respectively. However, the present invention is not limited to such a configuration, and the levels of these upstream digital call signals and downstream digital call signals are
For example, as listed below, various signals identified in the process of channel control matching the channel arrangement or multiple access scheme or information represented by the signals may be substituted.

A signal indicating a period during which transmission or reception should be performed for a predetermined wireless channel.-A signal provided by hardware that encodes or modulates a call signal to be transmitted, and indicates the presence or absence or level of the call signal. Signal indicating desired accuracy-Signal provided by hardware for decoding and demodulating received call signal, signal indicating presence or absence or level of the call signal with desired accuracy Further, the present invention relates to the above-described embodiment. The present invention is not limited to the above, and various embodiments are possible within the scope of the present invention, and any or all of the constituent elements may be improved.

The configurations of the invention disclosed in each of the above-described embodiments will be arranged in a hierarchical and multifaceted manner, and will be sequentially listed as additional items. (Supplementary Note 1) Acoustic conversion means 11 for converting a call signal received via the transmission line 10 into an acoustic signal, and a voice uttered by a speaker into a call signal to be transmitted to the transmission line 10. , The acoustic coupling monitoring means 13 for monitoring the degree of acoustic coupling with the signal converting means 12 whose relative position to the acoustic converting means 11 can be changed, and the transmission from the acoustic converting means 11 via the signal converting means 12. Road 10
A call signal processing device, characterized in that the echo path leading to (1) is provided with an echo suppressing means (14) for mitigating a gain variation caused by an increase / decrease in acoustic coupling degree monitored by the acoustic coupling monitoring means (13).

(Supplementary Note 2) In the speech signal processing device according to Supplementary Note 1, the acoustic coupling monitoring means 13 is configured to have a shape and a size of a member that supports both the acoustic conversion means 11 and the signal conversion means 12. A call signal processing device, wherein the acoustic coupling degree is monitored as a conversion value corresponding to both or either one.

(Supplementary Note 3) In the call signal processing apparatus according to Supplementary Note 1, the acoustic coupling monitoring means 13 has a three-dimensional shape composed of members that individually support the acoustic conversion means 11 and the signal conversion means 12. A speech signal processing device, wherein the acoustic coupling degree is monitored as a conversion value corresponding to both or one of the dimensions.

(Supplementary Note 4) In the speech signal processing device according to Supplementary Note 1, the acoustic coupling monitoring means 13 does not correlate the speech signal converted into the acoustic signal by the acoustic conversion means 11 with the speech signal. A telephone call characterized by superimposing a pilot signal that can be regarded as a signal, and monitoring the acoustic coupling degree as a cross-correlation between the pilot signal and the pilot signal superimposed on the telephone call signal obtained by the signal conversion means 12. Signal processing device.

(Supplementary Note 5) In the speech signal processing device according to any one of Supplementary Notes 1 to 4, the echo suppressing means 14 is updated by the acoustic converting means without the interposition of the echo suppressing means 14. 11 and / or the signal converting means 12 and / or gain, the level of the call signal received through the transmission line 10, and the level of the voice uttered by the speaker, all or one of A call signal processing device, which alleviates a change in gain occurring in the echo path according to an increase or decrease in the number of parts.

(Supplementary Note 6) In the speech signal processing device according to any one of Supplementary Notes 1 to 5, the acoustic coupling degree between the acoustic converting means 11 and the signal converting means 12 can be set to an individual value that can be actually taken. A storage unit 15 is prestored with a variation in the gain of the echo path that occurs below, and the echo suppression unit 14 is stored in the storage unit 15 in correspondence with the acoustic coupling degree monitored by the acoustic coupling monitoring unit 13. A speech signal processing device, characterized in that variation in gain of the echo path is mitigated over the stored variation.

(Supplementary Note 7) In the call signal processing device according to any one of Supplementary Notes 1 to 5, the acoustic coupling degree between the acoustic converting means 11 and the signal converting means 12 can be set to an individual value that can be actually taken. A gain storage means 15A in which the gain of the echo path generated below is stored in advance is provided, and the echo suppression means 14 corresponds to the acoustic coupling degree monitored by the acoustic coupling monitoring means 13 and the gain storage means 15 is provided.
A speech signal processing apparatus, wherein the variation in gain of the echo path is alleviated over the gain deviation stored in A.

(Supplementary Note 8) In the speech signal processing device according to any one of Supplementary Notes 1 to 7, the acoustic signal obtained by the acoustic conversion means 11 and the speech signal obtained by the signal conversion means 12 And / or one of them is provided with adaptive filtering means 16 for performing signal processing based on an adaptive algorithm.

(Supplementary note 9) In the speech signal processing apparatus according to supplementary note 8, the adaptive filtering means 16 is a signal based on the adaptive algorithm when the acoustic coupling degree monitored by the acoustic coupling monitoring means 13 changes. A call signal processing device characterized by restarting processing.

(Supplementary Note 10) In the speech signal processing apparatus according to Supplementary Note 8 or Supplementary Note 9, the transmission is performed under individual values that the acoustic coupling degree between the acoustic converting means 11 and the signal converting means 12 can actually take. The adaptive filtering means 16 is provided for the call signal received via the path 10 and / or the call signal obtained by the signal converting means 12.
The adaptive filtering means 16 is provided with a processing mode storage means 17 in which a mode of signal processing to be performed by
Is the acoustic coupling degree monitored by the acoustic coupling monitoring means 13 for both or either of the speech signal received via the transmission line 10 and the speech signal obtained by the signal converting means 12. Correspondingly, a call signal processing device which performs signal processing of the form stored in the processing form storage means 17.

[0072]

As described above, according to the first aspect of the invention, howling and echo are generated due to the excessive gain of the echo path, and the speech quality is reduced due to the side effect when the gain is suppressed to an excessively low level. Degradation is avoided with high accuracy without complicating the configuration. Further, in the invention according to claim 2, transmission and reception are performed by an operator of a communication device provided with the call signal processing device according to the present invention and / or a call partner of the operator. Even if the level of the call signal to be transmitted is changed, howling and echo generation and the deterioration of the call quality are avoided.

Further, according to the invention described in claim 3, avoiding or suppressing the occurrence of howling or echo can be performed without imposing severe restrictions on the procedure, algorithm and accuracy of signal processing to be performed by the adaptive filtering means. To be Further, in the invention described in claim 4, the restrictions on the signal processing performed based on the adaptive algorithm and the coefficient to be referred to based on the adaptive algorithm are significantly relaxed.

Furthermore, in the invention described in claim 5, the occurrence of howling and echo is avoided with higher accuracy than in the case where the form of signal processing is constant regardless of the acoustic coupling degree described above. . In the invention of the first subordinate concept and the invention of the second subordinate concept of the invention described in claim 1, flexible adaptation to the standardization of the configuration and the reduction of the number of parts is ensured, and howling and echo are generated. Is avoided with high accuracy.

Furthermore, in the invention of the third subordinate concept of the invention described in claim 1, restrictions on the characteristics of the space where the acoustic conversion means and the signal conversion means can be located and the environment are alleviated, and The occurrence of howling and echo is highly accurate and stably avoided. In addition, in the first invention and the second invention related to the inventions described in claims 1 and 2, the occurrence of howling or echo due to the variation of the gain of the echo path and the deterioration of the call quality are stable. It is avoided with high accuracy.

Therefore, in the communication device to which these inventions are applied, it is possible to maintain high service quality at low cost without lowering reliability and to flexibly adapt to various services.

[Brief description of drawings]

FIG. 1 is a principle block diagram of a call signal processing apparatus according to the present invention.

FIG. 2 is a diagram showing first and second embodiments of the present invention.

FIG. 3 is a diagram showing an example of an attenuation amount L adapted to an opening degree θ.

FIG. 4 is a diagram showing third and fourth embodiments of the present invention.

FIG. 5 is a diagram showing a configuration example of a terminal device capable of hands-free communication.

[Explanation of symbols]

11 Acoustic conversion means 12 Signal conversion means 13 Acoustic coupling monitoring means 14 Echo suppression means 15 storage means 15A gain storage means 16 Adaptive filtering means 17 Processing mode storage means 20 Suppression amount determination unit 21,73 voice switch 22, 82 Suppression control unit 23, 83u, 83d Level detector 41 Echo Canceller 42 Opening detector 43 Adaptive filter 44 coefficient control unit 71 microphone 72 A / D converter 74 Transmitter / receiver 75 Mobile communication network 76 D / A converter 77 speakers 81u, 81d Suppression processing unit

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5K027 AA10 AA11 BB03 DD08 DD10                       DD18                 5K046 AA01 AA05 BB01 HH05 HH47                       HH68 HH78

Claims (5)

[Claims] 1. An acoustic conversion means for converting a call signal received through a transmission line into an acoustic signal, and a voice uttered by a talker into a call signal to be transmitted through the transmission line. Acoustic coupling monitoring means for monitoring a degree of acoustic coupling with a signal converting means whose relative position to the acoustic converting means can be changed; and an echo path from the acoustic converting means to the transmission path via the signal converting means, A call signal processing device, comprising: echo suppressing means for alleviating fluctuations in gain caused by an increase or decrease in the degree of acoustic coupling monitored by the acoustic coupling monitoring means. 2. The call signal processing device according to claim 1, wherein the echo suppressing unit is updated by the acoustic converting unit and / or the signal converting unit without the intervention of the echo suppressing unit. One of the gains, the level of the speech signal received through the transmission line, and the gain fluctuation occurring in the echo path in response to the increase or decrease of all or part of the level of the voice uttered by the speaker A call signal processing device characterized by easing. 3. The call signal processing device according to claim 1, wherein the acoustic signal obtained by the acoustic conversion unit and / or the speech signal obtained by the signal conversion unit are both or either. A speech signal processing device, characterized in that it is provided with adaptive filtering means for performing signal processing based on an adaptive algorithm. 4. The speech signal processing device according to claim 3, wherein the adaptive filtering unit re-executes signal processing based on the adaptive algorithm when the acoustic coupling degree monitored by the acoustic coupling monitoring unit changes. A call signal processing device characterized by being activated. 5. The call signal processing device according to claim 3, wherein the transmission path is set under individual values that can be actually taken by the acoustic coupling degree between the acoustic conversion means and the signal conversion means. Processing mode storage means for preliminarily storing the mode of signal processing to be performed by the adaptive filtering means in both or either of the call signal received via the call signal and the call signal obtained by the signal converting means. The adaptive filtering means may include both or one of a call signal received through the transmission path and a call signal obtained by the signal converting means, and the sound monitored by the acoustic coupling monitoring means. A call signal processing device, which performs signal processing in a form stored in the processing form storage means according to a degree of coupling.