JP5780096B2 - Audio processing apparatus, audio processing method, and audio processing program - Google Patents

Description

  The present invention relates to a sound processing device, a sound processing method, and a sound processing program that can perform sound input / output processing and can communicate with another sound processing device.

  An audio processing device that transmits audio collected by a microphone to another audio processing device installed at a remote location via the network, and simultaneously receives audio from the remote location via the network and outputs it from a speaker. Are known. Such a voice processing apparatus is widely used in a remote conference system or the like. A speakerphone is an example of the sound processing device. An audio processing device is usually equipped with an echo removal function. The sound processing apparatus can suppress the occurrence of echoes and howling by making echo cancellation function. For example, Patent Document 1 discloses a technique for suppressing the occurrence of echoes and howling in a system including two or more speakers and microphones by applying an echo removal function to reduce the gain of acoustic coupling.

  2. Description of the Related Art There is known a speech processing device that can connect and use a plurality of speech processing devices within the same site. In such an audio processing device, the audio processing device can be spread over a wide area in the base, thereby expanding the audible range of the conference audio and the sound collection range of the uttered sound.

JP 2001-95084 A

  Conventional echo cancellation is set to work when a speaker and microphone of a single audio processing device is used. For this reason, when a plurality of speech processing devices are connected in the same site, there is a possibility that the conventional echo removal cannot sufficiently suppress the occurrence of echoes and howling. On the other hand, when the technique described in Patent Document 1 is used, it is assumed that the occurrence of echoes and howling can be suppressed even when a plurality of voice processing devices are connected in the same site. In this case, however, a complex algorithm for controlling the gain of acoustic coupling must be implemented in the acoustic processing apparatus. Therefore, there is a problem that the voice processing device cannot be provided at a low cost because the configuration of the voice processing device becomes complicated or the number of parts included in the device increases.

  An object of the present invention is to suppress the occurrence of echoes and howling when a plurality of audio processing devices are connected to each other while suppressing the complexity of the device configuration and the increase in the number of parts included in the device. A voice processing apparatus, a voice processing method, and a voice processing program.

An audio processing apparatus according to a first aspect of the present invention is an audio processing apparatus including a microphone and a speaker, and communication control means for communicating with another audio processing apparatus, and the other audio processing by the communication control means. By communicating with a device, a first specifying means for specifying a timing for outputting a predetermined sound from the speaker, and outputting the predetermined sound from the speaker at a predetermined volume at the timing specified by the first specifying means. The predetermined sound obtained when the predetermined sound output at the predetermined volume from the other speaker connected to the other sound processing device and the predetermined sound volume is acquired via the microphone. A second specifying means for specifying the volume and the direction of arrival of the sound, and the sound processing device based on the volume and the direction of arrival specified by the second specifying means. First adjustment means for adjusting the overall directivity, which is a directivity of sensitivity to incoming sound, wherein the overall directivity is adjusted so that sensitivity to sound coming from a specific direction toward the voice processing device is attenuated; A first adjustment means for adjusting, and a change notification for notifying information for specifying the changed volume when the volume when the sound is output from the speaker is changed, to the other audio processing device Transmitting means for transmitting to, receiving means for receiving the change notification transmitted by the other audio processing device, and the information notified by the change notification when the change notification is received by the receiving means. And second adjusting means for readjusting the overall directivity adjusted by the first adjusting means .

According to the first aspect, even when a plurality of sound processing devices are used in a state of being connected to each other, echoes and howling are caused by the sound output from other sound processing devices. Can be deterred. In addition, since the well-known echo removal function that is normally provided in the speech processing apparatus can be used as it is, it is not necessary to newly install a special function for echo removal in the speech processing apparatus. For this reason, it is possible to efficiently suppress the occurrence of echoes and howling when a plurality of sound processing devices are connected while suppressing the complexity of the structure of the sound processing device and the increase in the number of components included in the device. it can. In addition, when the volume of the sound output from the speaker of another audio processing device is changed, the overall directivity necessary to suppress the occurrence of echo and howling needs to be adjusted again. On the other hand, in this aspect, the sound coming from the direction of arrival with respect to the sound processing device is readjusted by re-adjusting the overall directivity based on the volume after the change of the sound output from the speaker of the other sound processing device. The sensitivity to can be attenuated. As a result, the speech processing apparatus can readjust the overall directivity to an optimum state. Also, since the overall directivity is automatically adjusted, it is possible to save the trouble of resetting the overall directivity from the beginning by outputting a predetermined sound from the speaker of another audio processing device again.

  In the first aspect, the first adjustment means attenuates the sensitivity in the direction of arrival corresponding to the volume when the volume specified by the second specifying means is equal to or greater than a predetermined threshold, thereby The directivity may be adjusted. By selectively attenuating the input level of sound output from other sound processing devices, it is possible to effectively suppress the occurrence of echoes and howling.

  In the first aspect, the first adjustment means attenuates the sensitivity in the direction of arrival corresponding to the volume so that the volume when the predetermined sound is acquired via the microphone is within the predetermined threshold. The overall directivity may be adjusted as follows. As a result, the input level of the sound output from another audio processing device can be attenuated to a level at which no echo or howling occurs. As a result, the occurrence of echoes and howling can be more effectively suppressed.

  In the first aspect, the transmission unit calculates the difference in the volume before and after the change, transmits the change notification for notifying the difference, and the second adjustment unit is notified by the change notification. Based on the difference, the overall directivity adjusted by the first adjusting means may be readjusted. The sound processing device may readjust the overall directivity based on the difference between before and after the change in the volume of the sound output from the speaker of the other sound processing device, and attenuate the sensitivity to the sound coming from the direction of arrival. it can. Thus, the sound processing device can adjust the overall directivity to an optimal state according to the degree of change in the volume of the sound output from the speaker of the other sound processing device.

The audio processing apparatus according to the second aspect of the present invention is an audio processing apparatus including a microphone and a speaker, and communication control means for communicating with another audio processing apparatus, and the other audio processing by the communication control means. By communicating with a device, a first specifying means for specifying a timing for outputting a predetermined sound from the speaker, and outputting the predetermined sound from the speaker at a predetermined volume at the timing specified by the first specifying means. The predetermined sound obtained when the predetermined sound output at the predetermined volume from the other speaker connected to the other sound processing device and the predetermined sound volume is acquired via the microphone. A second specifying means for specifying the volume and the direction of arrival of the sound, and the sound processing device based on the volume and the direction of arrival specified by the second specifying means. First adjustment means for adjusting the overall directivity, which is a directivity of sensitivity to incoming sound, wherein the overall directivity is adjusted so that sensitivity to sound coming from a specific direction toward the voice processing device is attenuated; When the sensitivity when the microphone collects sound is changed , the first adjustment means to adjust, the difference of the sensitivity before and after the change is calculated, and the first adjustment based on the calculated difference And second adjusting means for readjusting the overall directivity adjusted by the means. According to the second aspect, even when a plurality of sound processing devices are used in a state of being connected to each other, echoes and howling are caused by the sound output from other sound processing devices. Can be deterred. In addition, since the well-known echo removal function that is normally provided in the speech processing apparatus can be used as it is, it is not necessary to newly install a special function for echo removal in the speech processing apparatus. For this reason, it is possible to efficiently suppress the occurrence of echoes and howling when a plurality of sound processing devices are connected while suppressing the complexity of the structure of the sound processing device and the increase in the number of components included in the device. it can. In addition, when the sensitivity of the microphone is changed, it is necessary to adjust the overall directivity necessary for suppressing the occurrence of echoes and howling again. On the other hand, in the present invention, the sensitivity with respect to the sound coming from the arrival direction can be attenuated by calculating the difference before and after the sensitivity change and readjusting the overall directivity based on the difference. As a result, the speech processing apparatus can readjust the overall directivity to an optimum state according to the degree of change in the sensitivity of the microphone. Also, since the overall directivity is automatically adjusted according to the difference, it is possible to save the trouble of resetting the overall directivity from the beginning by outputting a predetermined sound again from the speaker of another audio processing device. .

In the first aspect and the second aspect , the first specifying unit determines the timing based on identification information sequentially assigned to the voice processing device and the other voice processing device by the communication control unit. May be. The voice processing device can easily determine the timing for outputting the predetermined sound. The voice processing device independently determines the timing without requiring the user to use the voice processing device or the control device that controls the voice processing device to output the predetermined sound from the voice processing device at the predetermined timing. A predetermined sound can be output.

The speech processing method according to the third aspect of the present invention includes a communication control step for communicating with another speech processing device, and a speaker for the speech processing device by communicating with the other speech processing device through the communication control step. A first specifying step for specifying a timing for outputting a predetermined sound from the output control step for performing control for outputting the predetermined sound at a predetermined volume from the speaker at the timing specified by the first specifying step; When the predetermined sound output at the predetermined volume from another speaker connected to another audio processing apparatus is acquired via the microphone of the audio processing apparatus, the volume and direction of arrival of the acquired predetermined sound are indicated. Based on the second specifying step to specify, the volume and the direction of arrival specified by the second specifying step, the sound processing device A first adjustment step of adjusting the overall directivity, which is a directivity of sensitivity to an incoming sound, wherein the overall directivity is set so that sensitivity to a sound coming from a specific direction toward the voice processing device is attenuated. A first adjustment step for adjusting, and a change notification for notifying information for specifying the changed volume when the volume when the sound is output from the speaker is changed, to the other audio processing device A transmitting step for transmitting to the receiving device; a receiving step for receiving the changing notification transmitted by the other audio processing device; and the information notified by the changing notification when the changing notification is received by the receiving step. And a second adjustment step for readjusting the overall directivity adjusted by the first adjustment step . According to the 3rd aspect, there can exist an effect similar to a 1st aspect.

The audio processing program according to the fourth aspect of the present invention includes a communication control step for communicating with another audio processing device, and a speaker of the audio processing device by communicating with the other audio processing device through the communication control step. A first specifying step for specifying a timing for outputting a predetermined sound from the output control step for performing control for outputting the predetermined sound at a predetermined volume from the speaker at the timing specified by the first specifying step; Acquired when the predetermined sound output at the predetermined volume from another speaker connected to the other audio processing device is acquired through the microphone of the audio processing device under the control of the other audio processing device. A second specifying step for specifying the volume and the direction of arrival of the predetermined sound, and the volume specified by the second specifying step. And a first adjustment step for adjusting the overall directivity, which is the directivity of sensitivity to the sound arriving at the sound processing device, based on the arrival direction, and the sound arriving from a specific direction toward the sound processing device A first adjustment step for adjusting the overall directivity so that the sensitivity to the sound is attenuated, and notification of information for specifying the changed sound volume when the sound volume when the sound is output from the speaker is changed Transmitting the change notification to the other voice processing device, receiving the change notification transmitted by the other voice processing device, and receiving the change notification by the receiving step. A second adjustment for readjusting the global directivity adjusted by the first adjustment step based on the information notified by the change notification. To execute the steps in the computer of the audio processing device. According to the 4th aspect, there can exist an effect similar to a 1st aspect.

1 is a diagram showing an outline of a conference system 1 including a voice processing device 10 and an electrical configuration of the voice processing device 10. FIG. 6 is a schematic diagram showing a first example of a state table 231. FIG. It is a flowchart which shows a main process. 4 is a flowchart showing main processing, which is a continuation of FIG. 3. 5 is a flowchart showing main processing, which is a continuation of FIG. 4. It is a figure which shows the pattern of whole directivity. 6 is a schematic diagram illustrating a second example of a state table 231. FIG. 1 is a diagram showing a network configuration of a voice processing device 10. FIG.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. These drawings are used to explain technical features that can be adopted by the present invention. The configuration of the apparatus, the flowcharts of various processes, and the like that are described are not intended to be limited to only that, but are merely illustrative examples.

  An overview of the conference system 1 will be described with reference to FIG. The conference system 1 includes voice processing apparatuses 11, 12, 13 and a PC 15. The voice processing apparatuses 11, 12, 13 and the PC 15 are installed in the same base (hereinafter also referred to as own base). The audio processing devices 11, 12, 13 and the PC 15 are daisy chain connected by a communication cable. Daisy chain refers to a connection method for connecting a plurality of devices in a daisy chain. The voice processing device 11 is connected to the PC 15 and the voice processing device 12. The audio processing device 12 is connected to the audio processing devices 11 and 13. The PC 15 is also connected to the Internet network 16. Therefore, in the conference system 1, the (Internet network 16) PC 15 and the audio processing devices 11, 12, and 13 are connected in this order. Hereinafter, when the voice processing devices 11, 12, and 13 are not distinguished or collectively referred to, they are referred to as the voice processing device 10.

  The voice processing apparatus 10 can communicate with other PCs (not shown) and other voice processing apparatuses (not shown) installed at other sites different from the own site via the PC 15 and the Internet network 16. The voice processing apparatus 10 transmits voice data collected by a microphone 25 (described later) to another voice processing apparatus via the Internet network 16 and simultaneously transmits voice data from the other voice processing apparatus via the Internet network 16. Data is received and sound is output from the speaker 24. A user at his / her own site using the voice processing device 10 can perform a voice remote conference with another user at another site using another voice processing device.

  In the conference system 1, the voice processing devices 11, 12, and 13 can be scattered over a wide area in the own base. And the sound based on the audio | voice data transmitted from the other speech processing apparatus installed in another base can be output from the speaker 24 (after-mentioned) of the speech processing apparatuses 11,12,13. Thereby, the sound output from the speaker 24 can be heard in a wide area. In addition, the voice processing device 10 can collect the voice in the local site to every corner, and can transmit the voice data to other voice processing devices installed in other bases.

  The system configuration in FIG. 1 is an example of the present invention, and other system configurations may be used. For example, various known external communication networks such as a fixed telephone network, a mobile telephone network, and a dedicated communication network may be used in place of the Internet network 16, and communication may be performed between the own base and another base. The voice processing device 10 may be connected to an external communication network via various devices (a fixed telephone, a mobile phone, a router, a modem, etc.) other than the voice processing device 10 instead of the PC 15. The voice processing device 10 may be directly connected to an external communication network.

  In the conference system 1, a display and a camera may be connected to the PC 15. The PC 15 transmits the video data of the local site taken by the camera to the other PC via the Internet network 16 and simultaneously receives the video data from the other PC via the Internet network 16 and displays the video on the display. May be displayed. As a result, the user at the local site can hold a video and audio remote conference with other users at other sites.

  The electrical configuration of the audio processing device 10 will be described. The voice processing device 10 includes a CPU 20 that controls the voice processing device 10. The CPU 20 is electrically connected to the ROM 21, RAM 22, flash memory 23, speaker 24, microphone 25, communication interface (hereinafter referred to as communication I / F) 26, and input unit 27. The ROM 21 stores a boot program, BIOS, OS, and the like. The RAM 22 stores a timer, a counter, and temporary data. The RAM 22 stores a sound volume setting value output from the speaker 24 and a sensitivity setting value of the microphone 25. Hereinafter, the information stored in the RAM 22 as the set value of the volume of the sound output from the speaker 24 is referred to as the speaker volume. The information stored in the RAM 22 as the sensitivity setting value of the microphone 25 is referred to as microphone sensitivity. The user can set the speaker volume and the microphone sensitivity by operating the input unit 27 (described later).

  The flash memory 23 stores a control program for the CPU 20. The flash memory 23 stores a state table 231 (see FIG. 2) described later. The flash memory 23 stores test sound data, which will be described later, and a test sound volume that is the sound volume of the test sound. The speaker 24 can output sound at the speaker volume stored in the RAM 22. The microphone 25 can collect sound with the microphone sensitivity stored in the RAM 22. A plurality of microphones 25 are provided around the sound processing apparatus 10. The CPU 20 can specify the arrival direction of the collected sound based on the volume of the sound collected by each microphone 25 and the time difference between the sound collection timings of the sound in each microphone 25. Further, the CPU 20 can adjust the directivity of the microphone 25 as a whole by adjusting the amplification degree of the sound collected by each microphone 25. Hereinafter, the directivity of the microphone 25 as a whole is referred to as overall directivity. The communication I / F 26 is an interface for communicating with the other voice processing apparatus 10 and the PC 15. Note that the audio processing device 10 realizes daisy chain connection by connecting to two different devices. For this reason, two or more communication I / Fs 26 are provided in order to communicate with two different other apparatuses. The input unit 27 is a button for setting speaker volume and microphone sensitivity.

  The sound processing apparatus 10 is equipped with a known echo removal function. The sound processing device 10 can suppress echo and howling generated by the microphone 25 collecting sound output from the speaker 24 of the sound processing device 10 by making echo removal function. However, for example, when the sound output from the speaker 24 of the sound processing devices 12 and 13 is collected by the microphone 25 of the sound processing device 11, the sound processing device 11 may not be able to function echo removal effectively. is there. The echo removal function is based on the sound collected by the microphone 25 and adjusts the sound output from the speaker 24 of its own device by feedback control to suppress echo and howling. Therefore, when the sound output from the speakers 24 of the other connected sound processing devices 12 and 13 is a cause of echo or howling, the sound processing device 11 outputs from the speakers 24 of the sound processing devices 12 and 13. Since the sound to be played cannot be adjusted, echo and howling cannot be suppressed.

  On the other hand, the sound processing apparatus 11 according to the present embodiment adjusts the overall directivity, and reduces the sound volume when the sound output from the speaker 24 of the sound processing apparatuses 12 and 13 is collected by the microphone 25. As a result, the sound processing device 11 can suppress the occurrence of echoes and howling due to the sound output from the speaker 24 of the sound processing devices 12 and 13. In addition, the sound processing device 11 can suppress the occurrence of echo and howling due to the sound output from the speaker 24 of the sound processing device 11 by using a known echo removal function. Therefore, the voice processing apparatus 11 can perform a remote conference in a clear voice environment free from echo and howling even when the voice processing apparatuses 12 and 13 connected to each other are installed in the local site. The details will be described below.

  The state table 231 stored in the flash memory 23 of the sound processing device 11 will be described with reference to FIG. The state table 231 is a table that is referred to when the speech processing apparatus 11 adjusts the overall directivity. In the state table 231, information indicating the reception volume and the arrival direction of the sound is stored in association with the ID of the sound processing device 10. The reception volume is a volume when sound that has arrived from the direction of arrival is collected. The arrival direction is indicated by an angle with respect to a predetermined direction of the speech processing device 10. Details of the creation method and the reference method of the state table 231 will be described later.

  With reference to FIG. 3 to FIG. 5, main processing executed by the voice processing device 10 will be described. The main process described below is executed by the CPU 20 of the sound processing apparatus 10 according to the sound processing program stored in the flash memory 23. The main processing is started when the main processing program stored in the flash memory 23 is activated when the power of the audio processing device 10 is turned on. Then, the CPU 20 executes this program. In addition to the main process, the CPU 20 executes various known acoustic processes (echo cancellation process, etc.) in parallel. These descriptions are omitted below. In the RAM 22, variable IDs, X, and M are defined when the main process is started.

  When the main process is started, the CPU 20 initializes the ID by storing 0 in the ID (S11). The CPU 20 determines whether another audio processing device 10 or the PC 15 is connected via the communication I / F 26 (see FIG. 1) (S13). If it is determined that the other voice processing apparatus 10 or PC 15 is not connected (S13: NO), the process returns to S13 in order to continuously monitor the connection of the other voice processing apparatus 10 or PC 15 . On the other hand, when it is determined that the voice processing device 10 or the PC 15 is connected (S13: YES), the CPU 20 determines whether the PC 15 is connected (S15). Specifically, it is determined whether or not the PC 15 is connected by detecting an electrical state (for example, a pull-up state) of a signal included in the communication cable. The method for determining whether the PC 15 is connected is not limited to this method. For example, when the apparatuses are connected via a communication cable, initial communication for notifying each other's type may be executed. The CPU 20 may determine whether or not the PC 15 is connected by initial communication. When it is determined that the PC 15 is connected (S15: YES), the CPU 20 stores 1 in the ID (S17). Thereby, the ID of the voice processing device 10 is determined. The CPU 20 transmits ID notification data for notifying the determined ID to the other voice processing apparatus 10 connected via the communication cable (S19). The process proceeds to S29.

  On the other hand, when it is determined in S15 that the PC 15 is not connected (S15: NO), the other voice processing device 10 is connected to the voice processing device 10. The CPU 20 determines whether or not the ID notification data transmitted from the connected PC 15 or other voice processing device 10 has been received (S21). When it is determined that the ID notification data has been received (S21: YES), the CPU 20 stores a value obtained by adding 1 to the ID notified by the received ID notification data in the ID (S23). Thereby, the ID of the voice processing device 10 is determined. The CPU 20 transmits ID notification data for notifying the determined ID to the other voice processing apparatus 10 connected via the communication cable (S25). The process proceeds to S29.

  On the other hand, if it is determined in S21 that the ID notification data has not been received (S21: NO), the CPU 20 determines whether or not it is connected to the end of the daisy chain connection (S27). When the other audio processing device 10 or the PC 15 is connected to one side of the communication I / F 26 and neither the other audio processing device 10 nor the PC 15 is connected to the other side, the CPU 20 terminates the daisy chain connection. It is determined that it is in a connected state. If it is determined that the terminal is connected to the end of the daisy chain connection (S27: YES), the process returns to S13 in order to continuously monitor the connection of another audio processing device 10 or the PC 15 to the terminal. . On the other hand, when it is determined that the terminal is not connected to the end of the daisy chain connection (S27: NO), in order to continuously monitor the reception of the ID notification data transmitted from the PC 15 or the other voice processing device 10, The process returns to S21.

  In S29, the CPU 20 determines whether a predetermined time has elapsed since the ID notification data was transmitted in S19 or S25 (S29). If it is determined that the predetermined time has not elapsed (S29: NO), the process returns to S29. If it is determined that the predetermined time has elapsed (S29: YES), the process proceeds to S31 (see FIG. 4).

  For example, when the voice processing apparatus 10 is daisy chain connected as shown in FIG. 1, the voice processing apparatus 11 determines 1 as the ID (S17) when the PC 15 is connected (S15: YES), and the communication. ID notification data is transmitted to the audio processing device 12 connected via the cable (S19). The voice processing device 12 receives the ID notification data from the voice processing device 11 (S21: YES), and determines 2 as the ID (S23). The voice processing device 12 transmits the ID notification data to the voice processing device 13 connected via the communication cable (S25). The voice processing device 13 receives the ID notification data from the voice processing device 12 (S21: YES), and determines 3 as the ID (S23). As described above, the IDs of the sound processing apparatuses 11, 12, and 13 are determined as 1, 2, and 3, respectively.

  As shown in FIG. 4, the CPU 20 initializes by storing 1 in X (S31). The CPU 20 deletes the ID, reception volume, and arrival direction stored in the state table 231 (see FIG. 2), and clears the state table 231 (S33). CPU20 memorize | stores 0 in the speaker volume memorize | stored in RAM22 (S35).

  The CPU 20 determines whether X is equal to the total number of audio processing devices 10 connected in a daisy chain (S37). When X is smaller than the total number (S37: NO), the CPU 20 determines whether X and ID match (S39). If it is determined that X and ID match (S39: YES), the CPU 20 stores the test volume of the flash memory 23 in the speaker volume of the RAM 22 (S41). The CPU 20 outputs a test sound from the speaker 24 based on the test sound data in the flash memory 23 (S43). The test sound is output from the speaker 24 at the test volume. After the output of the test sound from the speaker 24 is completed, the CPU 20 stores 0 in the speaker volume of the RAM 22 (S45). The CPU 20 updates X by adding 1 to X (S47). The process returns to S37.

  On the other hand, if it is determined in S39 that X and ID do not match (S39: NO), the CPU 20 determines whether or not a test sound is output from the speaker 24 of another audio processing device 10 (S49). When the test sound cannot be received via the microphone 25, the CPU 20 determines that the other sound processing apparatus 10 does not output the test sound (S49: NO). In this case, in order to continuously monitor the reception of the test sound, the process returns to S49. On the other hand, when the test sound is received through the microphone 25, it is determined that the test sound is output from the speaker 24 of the other sound processing apparatus 10 (S49: YES). The CPU 20 specifies the reception volume and arrival direction of the test sound received via the microphone 25 (S51). The CPU 20 specifies X at this point as the ID of the other voice processing apparatus 10 that has output the test sound. The CPU 20 stores the specified reception volume and arrival direction in the state table 231 (see FIG. 2) in association with the ID of the other specified voice processing device 10 (S53). The CPU 20 updates X by adding 1 to X (S55). The process returns to S37.

  When X is repeatedly updated in S47 and S55 and X becomes equal to the total number (S37: YES), the process proceeds to S61 (see FIG. 5).

  For example, in FIG. 1, when the IDs of the sound processing devices 11, 12, and 13 are determined as 1, 2, and 3 respectively, first, the test sound is the test volume from the speaker 24 of the sound processing device 11 with ID 1. Is output (S43). The voice processing devices 12 and 13 receive the test sound output from the speaker 24 of the voice processing device 11 via the microphone 25, and specify the reception volume and direction of arrival (S51). The voice processing devices 12 and 13 store the identified reception volume and arrival direction in the state table 231 (see FIG. 2) of the flash memory 23 in association with ID 1. Next, a test sound is output at a test volume from the speaker 24 of the audio processing device 12 of ID 2 (S43). The audio processing devices 11 and 13 receive the test sound output from the speaker 24 of the audio processing device 12 via the microphone 25, and specify the reception volume and arrival direction (S51). The audio processing devices 11 and 13 store the identified reception volume and arrival direction in the state table 231 (see FIG. 2) in association with ID 2. Similar processing is repeated when a test sound is output from the speaker 24 of the sound processing device 13.

  As shown in FIG. 5, the CPU 20 initializes by storing 1 in M (S61). The CPU 20 determines whether or not the reception volume corresponding to ID M in the state table 231 is equal to or higher than a predetermined value (for example, 70 dB) (S63). Note that the predetermined value is set to the minimum volume that may generate howling or echo when the sound output from the speaker 24 of another audio processing device 10 is received via the microphone 25. When the reception volume corresponding to ID M is less than the predetermined value (S63: NO), the volume of sound output from the speaker 24 of the audio processing device 10 of ID M is sufficiently small, which causes generation of echo and howling. Unlikely. The CPU 20 updates M by adding 1 to M (S69). In order to repeatedly execute the determination of the reception volume based on the updated M, the process returns to S63.

  On the other hand, if the reception volume corresponding to ID M in the state table 231 is greater than or equal to a predetermined value (S63: YES), the volume of the sound output from the speaker 24 of the audio processing apparatus 10 of ID M is large, and echo or howling is performed. Is likely to be the cause of Therefore, it is necessary to suppress the occurrence of echoes and howling by reducing the reception volume when the sound output from the speaker 24 of the IDM audio processing apparatus 10 is collected by the microphone 25. The CPU 20 adjusts the overall directivity by reducing the sensitivity of the arrival direction corresponding to ID M in the state table 231 (S65). The CPU 20 determines whether M is equal to the total number of audio processing devices 10 connected in a daisy chain (S67). When M is smaller than the total number (S67: NO), the CPU 20 updates M by adding 1 to M (S69). Since the determination of the reception volume is repeatedly performed based on the updated M, the process returns to S63. When M is repeatedly updated in S69 and M becomes equal to the total number (S67: YES), the adjustment of the overall directivity is completed. The process proceeds to S71.

  FIG. 6 shows the pattern of overall directivity of the sound processing devices 11 to 13 adjusted by the processing of S61 to S69. Assume that the sound processing apparatuses 11 (ID: 1), 12 (ID: 2), and 13 (ID: 3) are arranged in a straight line in order from the left to the right. Then, it is assumed that the state table 231 shown in FIG. Since the distance between the voice processing devices 11 and 12 is small, as shown in FIG. 2, when the test sound output from the voice processing device 12 (ID: 2) is collected by the voice processing device 11 The reception volume of is increased (75 dB). Since this value is larger than the predetermined value (70 dB) (S63: YES, see FIG. 5), the overall directivity is adjusted (S65, see FIG. 5). The specific method of adjustment is as follows. The position of the sound processing device 11 is close to the arrival direction (0 deg, see FIG. 2) so that the reception volume (75 dB) when the test sound transmitted from the sound processing device 12 is received is equal to or less than the predetermined value (70 dB). Attenuating the sensitivity of the microphone 25 arranged in the. As a result, the sensitivity of the sound coming from the arrival direction (0 deg) is attenuated by 5 dB with respect to the speech processing device 11.

  On the other hand, since the distance between the speech processing devices 11 and 13 is sufficiently large, the reception volume when the test sound output from the speech processing device 13 (ID: 3) is collected by the speech processing device 11 is small. (63 dB, see FIG. 2). Since this value is smaller than the predetermined value (70 dB) (S63: NO, see FIG. 5), the overall directivity is not adjusted. As a result of the above, as shown in FIG. 6, the overall directivity pattern 31 of the voice processing device 11 has a sensitivity in the direction in which the voice processing device 12 is disposed (right direction in FIG. 6). However, it is 5 dB smaller than the sensitivity in other directions.

  On the other hand, in the voice processing device 12, both the distance to the voice processing device 11 and the distance to the voice processing device 13 are small. Accordingly, the omnidirectional pattern 32 of the voice processing device 12 is in the direction in which the voice processing device 11 is arranged with respect to the voice processing device 12 (left direction in FIG. 6) and the voice processing with respect to the voice processing device 12. Both sensitivities in the direction in which the device 13 is arranged (right direction in FIG. 6) are smaller than those in the other directions.

  As described above, the sound processing device 10 attenuates the sensitivity to sound coming from a specific direction toward the sound processing device 10, thereby receiving the received sound volume when the sound output from the other sound processing device 10 is received. Can be selectively attenuated. As a result, the sound processing device 10 can effectively suppress the occurrence of echoes and howling due to the sound output from the other sound processing devices 10. In addition, the sound processing device 10 adjusts the sensitivity of the microphone 25 disposed in a position close to a specific direction so that the reception volume when receiving the sound output from the other sound processing device 10 is equal to or less than a predetermined value. And adjust the overall directivity. As a result, the reception volume can be efficiently attenuated to a level at which no echo or howling occurs. As a result, the speech processing apparatus 10 can more effectively suppress the occurrence of echoes and howling.

  As shown in FIG. 5, after the overall directivity is adjusted, the CPU 20 starts voice remote conference by communicating voice data with another voice processing device installed at another base. To do. After the start of the remote conference, the CPU 20 determines whether an operation for changing the speaker volume has been performed by the user via the input unit 27 (S71). When it is determined that an operation for changing the speaker volume has been performed (S71: YES), the CPU 20 stores the changed speaker volume in the RAM 22. The CPU 20 changes the volume of the sound output from the speaker 24 based on the speaker volume stored in the RAM 22.

  When the speaker volume of the voice processing device 10 is changed, the other voice processing devices 10 need to readjust the overall directivity adjusted in S65 based on the changed speaker volume. The CPU 20 transmits the change notification data for notifying the other audio processing device 10 of the difference between the speaker volume before the change and the speaker volume after the change, to the other audio processing device 10 connected via the communication cable. (S73). The process returns to S61.

  If it is determined in S71 that the operation for changing the speaker volume has not been performed (S71: NO), the CPU 20 determines whether or not change notification data has been received from another audio processing device 10 (S75). If it is determined that the change notification data has been received (S75: YES), the CPU 20 relays and transfers the change notification data so that the change notification data reaches all the daisy chain-connected voice processing devices 10. For example, when the voice processing device 12 (see FIG. 1) receives the change notification data from the voice processing device 11 (see FIG. 1), the CPU 20 of the voice processing device 12 sends the received change notification data to the voice processing device 13 (see FIG. 1). Relay) to (see). Next, the CPU 20 updates the contents of the state table 231 based on the difference notified by the received change notification data. Specifically, it is as follows.

  The CPU 20 selects a reception volume corresponding to the ID of the voice processing apparatus 10 that has transmitted the change notification data from the state table 231. The CPU 20 updates the state table 231 by adding the difference notified by the received change notification data to the selected reception volume (S79). For example, it is assumed that the speaker volume has been changed to be increased by 10 dB in another audio processing apparatus 10. In this case, the difference notified by the change notification data is +10 dB. The CPU 20 updates the state table 231 by adding the notified difference +10 dB to the reception volume selected from the state table 231. The process returns to S61.

  Based on the updated state table 231, the CPU 20 determines whether or not the reception volume corresponding to each ID is a predetermined value or more (S63). When the reception volume is less than the predetermined value (S63: NO), the volume of the sound field output from the speaker 24 of the other sound processing apparatus 10 is sufficiently small with the speaker volume after the change, which may cause echo and howling. The nature is low. In this case, the overall directivity is not readjusted. On the other hand, when the reception volume is equal to or higher than the predetermined value (S63: YES), the volume of the sound output from the speaker 24 of the other audio processing device 10 is large at the changed speaker volume, which is a cause of occurrence of echo and howling. Is likely to be. Therefore, in order to suppress the occurrence of echoes and howling, the CPU 20 readjusts the overall directivity (S65). Specifically, the direction in which the other sound processing device 10 with the changed speaker volume is arranged is specified by the arrival direction of the state table 231, and the sensitivity of the microphone 25 arranged at a position close to the specified direction is set. Readjust the global directivity by attenuating. The details are the same as the overall directivity adjustment method executed using the test sound, and thus the description thereof is omitted.

  When the speaker volume of another audio processing device 10 is changed, the overall directivity necessary to suppress the occurrence of echoes and howling needs to be adjusted again. On the other hand, in this embodiment, the overall directivity can be readjusted based on the speaker volume after the change in the other audio processing device 10. As a result, the speech processing apparatus 10 can readjust the overall directivity to an optimum state. In addition, since the overall directivity is adjusted according to the difference between before and after the change of the speaker volume, the overall directivity is reset from the beginning by outputting the test sound from the speaker 24 of another audio processing device 10 again. Save time and effort. Therefore, the speech processing apparatus 10 can quickly readjust the overall directivity.

  Furthermore, the audio processing device 10 readjusts the sound output from the speaker 24 of the other audio processing device 10 by readjusting the overall directivity based on the difference before and after the change in the speaker volume of the other audio processing device 10. The overall directivity can be adjusted to an optimum state according to the degree of change in volume.

  If it is determined in S75 that the change notification data has not been received (S75: NO), the CPU 20 determines whether an operation for changing the microphone sensitivity has been performed by the user via the input unit 27 (S77). If it is determined that the operation for changing the microphone sensitivity is not performed (S77: NO), the process proceeds to S81. On the other hand, if it is determined that an operation for changing the microphone sensitivity has been performed (S77: YES), the CPU 20 stores the changed microphone sensitivity in the RAM 22. The CPU 20 changes the sensitivity when collecting sound via the microphone 25 based on the microphone sensitivity stored in the RAM 22.

  When the microphone sensitivity of the sound processing device 10 is changed, it is necessary to readjust the overall directivity adjusted in S65 based on the changed microphone sensitivity. The CPU 10 calculates the difference in microphone sensitivity before and after the change. The CPU 20 updates the state table 231 by adding the calculated difference to all the received sound volumes stored in the state table 231 (S79). For example, it is assumed that the sensitivity of the microphone is changed so as to be reduced by 10 dB. In this case, the difference in microphone sensitivity before and after the change is −10 dB. The CPU 20 updates the state table 231 by adding the calculated difference −10 dB to all received sound volumes stored in the state table 231. The process returns to S61.

  Based on the updated state table 231, the CPU 20 determines whether or not the reception volume corresponding to each ID is a predetermined value or more (S63). When the received sound volume is less than the predetermined value (S63: NO), when the sound output from the speaker 24 of the other sound processing apparatus 10 is received with the changed microphone sensitivity, the received sound volume is sufficiently small, and echo or howling is not performed. It is unlikely to be a cause of occurrence. In this case, the overall directivity is not readjusted. On the other hand, when the received sound volume is equal to or higher than the predetermined value (S63: YES), when the sound output from the speaker 24 of the other sound processing device 10 is received with the changed microphone sensitivity, the received sound volume increases, and echo or This is likely to be a factor in howling. Therefore, in order to suppress the occurrence of echoes and howling, the CPU 20 readjusts the overall directivity (S65). Specifically, the overall directivity is readjusted by attenuating the sensitivity of the microphone 25 arranged in the vicinity of the arrival direction corresponding to the reception volume larger than the predetermined value in the state table 231. The details are the same as the directivity adjustment method of the microphone 25 performed using the test sound, and thus the description thereof is omitted.

  When the sensitivity of the microphone 25 is changed, the overall directivity necessary to suppress the occurrence of echo and howling needs to be adjusted again. On the other hand, in this embodiment, the difference before and after the sensitivity change can be calculated, and the overall directivity can be readjusted based on the difference. As a result, the speech processing apparatus 10 can readjust the overall directivity to an optimal state according to the degree of change in sensitivity of the microphone 25. Further, since the overall directivity is adjusted according to the difference, it is possible to save the trouble of resetting the overall directivity from the beginning by outputting the test sound from the speaker 24 of the other audio processing device 10 again. Therefore, the speech processing apparatus 10 can quickly readjust the overall directivity.

  In S81, the CPU 20 determines whether another audio processing device 10 is newly connected to the audio processing device 10 in the daisy chain connection state (S81). If it is determined that another audio processing device 10 is connected (S81: YES), the process proceeds to S31 (see FIG. 4) in order to execute the process of adjusting the overall directivity by outputting a test sound from the beginning. Return. If it is determined that another audio processing device 10 is not connected (S81: NO), the process returns to S71.

  As described above, the voice processing device 10 outputs from other voice processing devices 10 even when the plurality of voice processing devices 10 are connected to each other and used in the same location. It is possible to suppress the occurrence of echo and howling due to the generated sound. Further, since the well-known echo removal function that is normally provided in the voice processing apparatus 10 can be used as it is, it is not necessary to newly install a special function for echo removal in the voice processing apparatus 10. For this reason, generation | occurrence | production of the echo and howling when the several audio | voice processing apparatus 10 is connected can be suppressed efficiently, suppressing the cost of the audio | voice processing apparatus 10. FIG.

  The voice processing apparatus 10 can easily determine the timing for outputting the test sound by communicating the ID notification data. Therefore, the voice processing apparatus 10 does not need to perform control to output a test sound from the voice processing apparatus 10 at a predetermined timing by a user who uses the voice processing apparatus 10 or a control device that controls the voice processing apparatus 10. Can output the test sound by judging the timing independently.

  The CPU 20 that performs the processes of S19 and S25 corresponds to the “communication control means” of the present invention. The CPU 20 that performs the process of S39 corresponds to the “first specifying means” of the present invention. The CPU 20 that performs the process of S43 corresponds to the “output control means” of the present invention. The CPU 20 that performs the process of S51 corresponds to the “second specifying means” of the present invention. The CPU 20 that performs the process of S65 corresponds to the “first adjustment unit” and the “second adjustment unit” of the present invention. The CPU 20 that performs the process of S73 corresponds to the “transmission means” of the present invention. The CPU 20 that performs the process of S75 corresponds to the “reception unit” of the present invention. The processes of S19 and S25 correspond to the “communication control step” of the present invention. The process of S23 corresponds to the “first specific step” of the present invention. The process of S43 corresponds to the “output control step” of the present invention. The process of S51 corresponds to the “second specifying step” of the present invention. The process of S65 corresponds to the “first adjustment step” of the present invention.

  In addition, this invention is not limited to the above-mentioned embodiment, A various change is possible. In the above description, the voice processing apparatus 10 is daisy chain connected. However, the voice processing apparatus 10 may be connected in other connection forms such as a tree type, a ring type, and a star type. The audio processing devices 10 are connected to each other via a communication cable, but the audio processing devices 10 may be connected to each other wirelessly. The audio processing device 10 may be configured not to include the speaker 24 and the microphone 25, or may be configured to be able to connect an external speaker and microphone.

  In the above description, each voice processing device 10 determines an ID by communicating ID notification data between the voice processing devices 10. In addition, the voice processing device 10 transmits test sounds in the order of the determined IDs. On the other hand, the PC 15 may collectively determine the ID of the voice processing device 10 by transmitting the ID notification data to the voice processing device 10. Further, the timing at which each voice processing device 10 outputs a test sound may be controlled by the PC 15 transmitting a control signal to the voice processing device 10. For example, the voice processing device 10 may output a test sound at the timing notified from the PC 15.

  For example, the audio processing device 10 may perform only control for storing the test volume in the speaker volume of the RAM 22 at the timing of outputting the test sound. The PC 15 may continuously transmit test sound data to the sound processing device 10. The sound processing apparatus 10 may output the test sound from the speaker 24 only when the test sound volume is stored in the speaker sound volume based on the test sound data received from the PC 15.

  The audio processing device 10 may add the reception volume stored in the state table 231 for each arrival direction. The audio processing device 10 may adjust the overall directivity so that the sensitivity in the corresponding arrival direction is attenuated when the added reception volume is equal to or higher than a predetermined value.

  The method of adjusting the overall directivity is not limited to the above-described embodiment. In the above description, the sound processing apparatus 10 adjusts the sensitivity of the microphone 25 so that the reception volume becomes a predetermined value when the reception volume of the state table 231 is equal to or higher than the predetermined value. On the other hand, the sensitivity of the microphone 25 may be adjusted such that a predetermined margin is subtracted from the predetermined value and the received sound volume becomes this value.

  In the above description, the audio processing device 10 transmits change notification data for notifying the difference in speaker volume to the other audio processing devices 10. On the other hand, the audio processing device 10 may transmit change notification data for notifying the changed speaker volume to the other audio processing devices 10. Further, when such change notification data is received from another voice processing device 10, the voice processing device 10 calculates a difference by subtracting the changed speaker volume notified by the received change notification data from the test volume. May be. The voice processing device 10 may readjust the overall directivity based on the calculated difference.

  Based on the information stored in the state table 231, the network configuration of the voice processing device 10 may be specified. For example, it is assumed that the state table 231 shown in FIG. The reception volume (85 dB) when the test sound output from the audio processing device 13 with ID 3 is collected is received when the test sound output from the audio processing device 12 with ID 2 is collected. It is larger than the volume (75 dB). In this case, the voice processing device 13 is arranged closer to the voice processing device 11 than the voice processing device 12. The arrival direction of the test sound output from the voice processing device 13 with ID 3 is 45 deg. Therefore, as shown in FIG. 9, the voice processing device 11 is arranged side by side with the voice processing devices 11 and 12, so that the voice processing device 13 approaches the voice processing device 11 side. It recognizes that it is arranged at a 45 degree position. The voice processing device 11 can adjust the overall directivity based on the recognized network configuration. As described above, the voice processing device 11 specifies the network configuration of the voice processing devices 11, 12, and 13 based on the reception volume and the arrival direction of the test sound, and adjusts the overall directivity based on the specified network configuration. Can do.

1 Conference system 10, 11, 12, 13 Audio processing device 23 Flash memory 24 Speaker 25 Microphone 231 Status table

Claims (8)

An audio processing device including a microphone and a speaker,
Communication control means for communicating with other audio processing devices;
First specifying means for specifying timing for outputting a predetermined sound from the speaker by communicating with the other sound processing device by the communication control means;
Output control means for performing control to output the predetermined sound at a predetermined volume from the speaker at the timing specified by the first specifying means;
When the predetermined sound output at the predetermined volume from another speaker connected to the other audio processing device is acquired via the microphone, the volume and direction of arrival of the acquired predetermined sound are specified. Two specific means;
First adjusting means for adjusting overall directivity, which is directivity of sensitivity to sound arriving at the sound processing device, based on the volume and the arrival direction specified by the second specifying means, the sound processing; First adjusting means for adjusting the overall directivity so that sensitivity to sound coming from a specific direction toward the device is attenuated ;
A transmission means for transmitting a change notification for notifying information for specifying the volume after the change to the other audio processing device when the volume when the sound is output from the speaker is changed;
Receiving means for receiving the change notification transmitted by the other audio processing device;
Second adjustment means for readjusting the overall directivity adjusted by the first adjustment means based on the information notified by the change notification when the change notification is received by the reception means; A speech processing apparatus comprising: The first adjusting means includes
The overall directivity is adjusted by attenuating sensitivity in the direction of arrival corresponding to the volume when the volume specified by the second specifying unit is equal to or greater than a predetermined threshold. The speech processing apparatus according to 1. The first adjusting means includes
The global directivity is adjusted by attenuating the sensitivity in the arrival direction corresponding to the volume so that the volume when the predetermined sound is acquired via the microphone is within the predetermined threshold. The speech processing apparatus according to claim 2. The transmission means calculates the difference in the volume before and after the change, transmits the change notification for notifying the difference,
The second adjusting means is
Based on the difference notified by the change notification, a voice processing device according to any one of claims 1 to 3, characterized in that the re-adjust the whole directional adjusted by the first adjusting means. An audio processing device including a microphone and a speaker,
Communication control means for communicating with other audio processing devices;
First specifying means for specifying timing for outputting a predetermined sound from the speaker by communicating with the other sound processing device by the communication control means;
Output control means for performing control to output the predetermined sound at a predetermined volume from the speaker at the timing specified by the first specifying means;
When the predetermined sound output at the predetermined volume from another speaker connected to the other audio processing device is acquired via the microphone, the volume and direction of arrival of the acquired predetermined sound are specified. Two specific means;
First adjusting means for adjusting overall directivity, which is directivity of sensitivity to sound arriving at the sound processing device, based on the volume and the arrival direction specified by the second specifying means, the sound processing; First adjusting means for adjusting the overall directivity so that sensitivity to sound coming from a specific direction toward the device is attenuated;
When the sensitivity when the microphone collects sound is changed, the difference between the sensitivity before and after the change is calculated, and the omnidirectional adjusted by the first adjustment unit based on the calculated difference features and to Ruoto voice processing apparatus further comprising a <br/> a second adjusting means for readjusting sex. The first specifying means includes
By the communication control means, based on the identification information sequentially assigned to the voice processing device and the other speech processing apparatus, according to any one of claims 1 to 5, characterized in that identifying the timing Voice processing device. A communication control step for communicating with other audio processing devices;
A first specifying step of specifying a timing for outputting a predetermined sound from a speaker of the sound processing device by performing communication with the other sound processing device in the communication control step;
An output control step for performing control to output the predetermined sound from the speaker at a predetermined volume at the timing specified by the first specifying step;
When the predetermined sound output at the predetermined volume from another speaker connected to the other audio processing apparatus is acquired via the microphone of the audio processing apparatus, the volume and direction of arrival of the acquired predetermined sound A second identification step for identifying
A first adjustment step of adjusting a global directivity, which is a directivity of sensitivity to sound arriving at the sound processing device, based on the volume and the arrival direction specified by the second specifying step, the sound processing A first adjustment step of adjusting the global directivity so that sensitivity to sound coming from a specific direction toward the device is attenuated ;
A transmission step of transmitting, to the other audio processing device, a change notification for notifying information for specifying the volume after the change when the volume when the sound is output from the speaker is changed;
A receiving step of receiving the change notification transmitted by the other audio processing device;
A second adjustment step for readjusting the global directivity adjusted by the first adjustment step based on the information notified by the change notification when the change notification is received by the reception step; A speech processing method characterized by comprising: A communication control step for communicating with other audio processing devices;
A first specifying step of specifying a timing for outputting a predetermined sound from a speaker of the sound processing device by performing communication with the other sound processing device in the communication control step;
An output control step for performing control to output the predetermined sound from the speaker at a predetermined volume at the timing specified by the first specifying step;
When the predetermined sound output at the predetermined volume from another speaker connected to the other sound processing device is acquired through the microphone of the sound processing device under the control of the other sound processing device. A second specifying step of specifying the volume and direction of arrival of the acquired predetermined sound;
A first adjustment step of adjusting a global directivity, which is a directivity of sensitivity to sound arriving at the sound processing device, based on the volume and the arrival direction specified by the second specifying step, the sound processing A first adjustment step of adjusting the global directivity so that sensitivity to sound coming from a specific direction toward the device is attenuated ;
A transmission step of transmitting, to the other audio processing device, a change notification for notifying information for specifying the volume after the change when the volume when the sound is output from the speaker is changed;
A receiving step of receiving the change notification transmitted by the other audio processing device;
A second adjustment step for readjusting the global directivity adjusted by the first adjustment step based on the information notified by the change notification when the change notification is received by the reception step; Is a voice processing program for causing the computer of the voice processing apparatus to execute the program.

Images