Audio input / output system and audio input / output method

Abstract

To remove stress of operators who respond by voice in contact center business and to remove the burden of maintenance of devices by the operators.SOLUTION: An operator communication device 200 selected from a plurality of operator communication devices 200 based on an automatic call distribution processing unit (ACD) of a contact center server 300 includes: an audio output unit 201 for outputting audio received from a customer communication device 400 from a speaker 111 whose audio output directivity is adjusted to an operator U10 using the operator communication device 200; and an audio input unit 202 for transmitting, to the customer communication device 400, audio input by the operator U10 to a microphone array unit 121 whose audio input directivity is adjusted.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention ,sound The present invention relates to a voice input / output system and a voice input / output method. [Background technology]

[0002] Traditionally, operators working at contact centers, which are businesses that specialize in handling customer phone calls, answer calls wearing so-called headsets, which are headphones worn on the head with a small microphone attached, or so-called hands-free microphones, which are earphones that are worn on the auricle or placed in close contact with the entrance of the ear canal with a small microphone attached.The use of headsets and hands-free microphones has the advantages of eliminating the need to hold the telephone handset, allowing both hands to be used freely, ensuring stable sound pickup by the microphone because the distance between the mouth (sound source) and the microphone does not change, and allowing customers' voices to be heard at an optimal volume.

[0003] Patent Document 1 discloses an invention that aims to improve work efficiency by determining the state of an operator from the state of a headset used by the operator and automatically changing the state of the operator. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2004-282154 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the case of a headset, for example, there is a problem that the pressure applied to the operator's head needs to be greater than a certain amount in order to stabilize the headset for long periods of time during work, and the weight of the headset itself can cause stress to the operator. Furthermore, headsets and hands-free microphones are used in close contact with the skin or the entrance to the ear canal. For this reason, from a hygienic standpoint, the devices had to be dedicated to each user and kept clean. This meant that the devices needed to be maintained by an operator. It is also possible to use the microphone and speaker of a PC or telephone instead of a headset, but in a space where many operators gather to work, the microphone picks up many surrounding sounds other than what the operator is saying, and the speaker broadcasts the sound widely to the surrounding area, making this method unsuitable for practical use.

[0006] In view of the above circumstances, an object of the present invention is to relieve the stress of operators who handle voice responses in contact center operations, and to relieve the operators of the burden of maintaining devices. [Means for solving the problem]

[0010] The present invention for solving the above-mentioned problems provides a processing device selected from a plurality of processing devices used by a plurality of operators based on an automatic call distribution processing unit of a contact center device, Multiple Speakers and Multiple and a voice input / output device equipped with a microphone, wherein each of the processing devices in the plurality of processing devices comprises a voice output unit that transmits first voice information received from a customer device to the voice input / output device, and a voice input unit that transmits second voice information to the customer device, and the voice input / output device transmits the first voice information received from each of the processing devices as follows: Each based on ear position information of the plurality of operators acquired from the imaging device, in order to output from the speaker to each operator using each of the processing devices, Multiple By adjusting the angle of directivity of the audio output from the speaker, the sound transmission area of ​​the audio signal is corrected so that the audio signal reaches the ears of each of the operators in the target sound transmission area. Eachan output control unit for directing the speaker; Each In order to transmit the second voice information input by each of the operators to the microphone to each of the processing devices used by each of the operators, based on mouth position information of the plurality of operators acquired from the imaging device, Multiple By adjusting the angle of the directivity of the voice input to the microphone, the sound pickup area of ​​the voice signal is corrected to focus on the mouth of each operator in the target sound pickup area. Each a sound collection unit to which the microphone is directed; When the distance between adjacent operators is equal to or less than a predetermined distance, the output control unit outputs the first audio information by directing a first speaker of the plurality of speakers toward the ears of a target operator among the adjacent operators, and outputs audio with a phase shift of half a wavelength of the wavelength of the output first audio information by directing a second speaker of the plurality of speakers toward the ears of the neighboring operator, and the sound collection unit directs two or more microphones of the plurality of microphones having different sound collection directions toward the mouth of the target operator, and extracts the audio component of the second audio information by subtracting the audio component contained in both of the audio collected by the two or more microphones using a spectral subtraction method. It is an audio input / output system.

[0011] The present invention also provides a call distribution system for a call center device, the call distribution system comprising: a processing device selected from a plurality of processing devices used by a plurality of operators based on an automatic call distribution processing unit of the contact center device; Multiple Speakers and Multiple a voice input / output device equipped with a microphone, wherein each of the plurality of processing devices executes a voice output step of transmitting first voice information received from a customer device to the voice input / output device, and a voice input step of transmitting second voice information to the customer device, and the voice input / output device transmits the first voice information received from each of the processing devices as follows: Each based on ear position information of the plurality of operators acquired from the imaging device, in order to output from the speaker to each operator using each of the processing devices, Multiple By adjusting the angle of directivity of the audio output from the speaker, the sound transmission area of ​​the audio signal is corrected so that the audio signal reaches the ears of each of the operators in the target sound transmission area. Each an output control step of directing the speaker; Each In order to transmit the second voice information input by each of the operators to the microphone to each of the processing devices used by each of the operators, based on mouth position information of the plurality of operators acquired from the imaging device, MultipleBy adjusting the angle of the directivity of the voice input to the microphone, the sound pickup area of ​​the voice signal is corrected to focus on the mouth of each operator in the target sound pickup area. Each A sound pickup step of pointing the microphone. In the output control step, when the distance between adjacent operators is equal to or less than a predetermined distance, a first speaker of the plurality of speakers is directed toward the ears of a target operator among the adjacent operators to output the first audio information, and a second speaker of the plurality of speakers is directed toward the ears of the adjacent operator to output audio with a phase shift of half a wavelength of the wavelength of the output first audio information; and in the audio collection step, two or more microphones of the plurality of microphones having different audio collection directions are directed toward the mouth of the target operator, and audio components of the second audio information are extracted by subtracting audio components contained in both of the audio collected by the two or more microphones using a spectral subtraction method. This is an audio input / output method. [Effects of the Invention]

[0012] According to the present invention, it is possible to relieve the stress of operators who handle voice calls in contact center operations, and also to relieve the operators of the burden of maintaining devices. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is an explanatory diagram of a work space where an operator responds to voice calls. [Figure 2] FIG. 1 is a functional configuration diagram of a voice input / output system according to a first embodiment. [Figure 3] 10 is a flowchart of the voice input / output process in the first embodiment, where (a) is a case where a customer speaks, and (b) is a case where an operator speaks. [Figure 4] FIG. 10 is a functional configuration diagram of a voice input / output system according to a second embodiment. [Figure 5] 10 is a flowchart of a voice input / output process according to the second embodiment, where (a) is a case where a customer speaks, and (b) is a case where an operator speaks. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Each drawing is merely a schematic illustration to allow a sufficient understanding of the present invention. Therefore, the present invention is not limited to the illustrated examples. Furthermore, in the drawings, the dimensions of components constituting the present invention may be exaggerated for clarity. In addition, in each drawing, common or similar components are designated by the same reference numerals, and redundant explanations thereof will be omitted.

[0015] [First embodiment] <Configuration> 1, the voice input / output system of the first embodiment is a system in which an operator U10 performing contact center work communicates with customers by voice within a work space. The voice input / output system includes a speaker 111, a microphone array unit 121, and an imaging device 131. Note that the first embodiment does not use the imaging device 131, and therefore a detailed description of the imaging device 131 will be provided later.

[0016] (Speaker 111) The speakers 111 are directional speakers that output the customer's voice (voice signal) to the operator U10 who is within a specific range. Directional speakers are disclosed, for example, at URL: https: / / bae.dentsutec.co.jp / articles / speaker / . Each of the speakers 111 is placed on the ceiling of the work space and above the head of the operator U10 who is working. The speakers 111 are directed toward the operator U10 who is directly below them, and can output the customer's voice only to the operator U10 who is directly below them.

[0017] When the distance between adjacent operators U10 is less than a predetermined distance, the sound output from the speaker 111 is heard not only by the operator U10 directly below, but also by the operator U10 next to him. To the adjacent operator U10, this sound is merely noise, so it is preferable to keep the distance between adjacent operators U10 at a predetermined distance or more. However, if the design of the work space necessitates that the distance between adjacent operators U10 be kept less than a predetermined distance, the following measure can be taken. That is, when sound is output from the speaker 111 above the head of the target operator U10, sound with a phase shift of half a wavelength from the wavelength of the sound can be output from the speaker 111 above the head of the adjacent operator U10. This makes it possible to mute the sound due to sound interference within the range of the adjacent operator U10.

[0018] (Microphone array part 121) The microphone array unit 121 is a directional sound collection device that collects the voices (voice signals) of the operators U10 within a specific range. Directional sound collection devices are disclosed in Japanese Patent Application Laid-Open No. 2014-072708. Each of the microphone array units 121 is arranged on a side wall of the work space and is directed toward the operators U10 who are working. Two microphone array units 121 with different sound collection directions are provided for one operator U10.

[0019] When collecting the voice of the target operator U10, two microphone array units 121, 121, which are directed toward the target operator U10 and have different sound pickup directions, form directional beams 123, 123 using beamforming (BF) toward the target area where the target operator U10 is located. The directional beams 123, 123 of each BF of the microphone array units 121, 121 contain non-target area sound components (the voice of the operator U10 next to the target operator U10) in the target area sound direction. However, sound components of the target area are also contained in both of the directional beams 123, 123 of the microphone array units 121, 121. Therefore, for example, spectral subtraction (SS) can be used to extract only the sound components of the target area. In other words, even if all the operators U10 in the work space are talking to customers, only the voice of the target operator U10 can be collected.

[0020] (Functional configuration) As shown in FIG. 2, the voice input / output system of the second embodiment includes a voice input / output device 100, an operator communication device 200, and a contact center server 300.

[0021] (Audio input / output device 100) The voice input / output device 100 is a device that outputs the voice of the customer to the operator U10 and inputs the voice signal of the operator U10. The voice input / output device 100 includes a directional speaker function 110 and an area sound collection device 120. The directional speaker function 110 controls the operation of the speaker 111. The directional speaker function 110 includes the speaker 111, which has already been described, and an output control unit 112. The output control unit 112 controls the output of audio output by the speaker 111. The area sound collection device 120 controls the operation of the microphone array units 121, 121. The area sound collection device 120 includes the microphone array units 121, 121 already described, and a sound collection unit 122. The sound collection unit 122 extracts only the sound of the target area (the sound of the operator U10 in the target area) from the sounds collected by the microphone array units 121, 121.

[0022] (Operator communication device 200) The operator communication device 200 is a device that provides the operator U10 with voice calls with customers connected via the contact center server 300. An operator communication device 200 is provided for each operator U10. The operator communication device 200 can be, for example, but is not limited to, a notebook PC used by the operator U10. The operator communication device 200 is communicatively connected to the contact center server 300 via the in-house communication network 500. The operator communication device 200 includes an audio output unit 201 and an audio input unit 202. The operator communication device 200 also includes a function unit (not shown) for performing well-known control functions for performing voice communication. The voice output unit 201 outputs voice from the customer and voice signals (e.g., voice signals of automated voice guidance) from the contact center server 300 via at least the contact center server 300. The voice output unit 201 is connected to the output control unit 112 so as to be able to communicate with it. The voice input unit 202 sends a voice signal from the operator U10 to the contact center server 300. The voice input unit 202 is connected to the sound collection unit 122 so as to be able to communicate with each other.

[0023] (Contact Center Server 300) The contact center server 300 is a device that manages contact center operations. The contact center server 300 is communicatively connected to one or more operator communication devices 200 via an in-house communication network 500. The contact center server 300 is also communicatively connected to one or more customer communication devices 400 via a public network 600. The customer communication device 400 is an information processing device used by a customer, such as, but not limited to, a PC or a smartphone.

[0024] The contact center server 300 includes an IVR (Interactive Voice Response) processing unit 301 , a call control processing unit 302 , and a communication processing unit 303 . The IVR processing unit 301 is a functional unit that processes automatic voice responses. The call control processor 302 functions as an ACD (Automatic Call Distribution) processor that appropriately distributes incoming calls to specific operators U10. The call control processor 302 automatically allocates incoming calls according to, for example, the caller number, the called number, the skills, working hours, and availability of the operators U10. When allocating an incoming call, an appropriate operator U10 is selected. The communication processing unit 303 functions as a PBX (Private Branch eXchange) processing unit that enables extension calls between multiple telephones within the facility and connects to outside lines.

[0025] The local area network 500 is, for example, a LAN (Local Area Network), but is not limited to this. The public network 600 is, for example, an IP (Internet Protocol) network, but is not limited to this. The operator communication device 200, the contact center server 300, and the customer communication device 400 are configured as computers having a CPU (Central Processing Unit), memory, storage means (storage unit) such as a hard disk, and a network interface. The CPU executes programs loaded into the memory, thereby realizing various functions of the computer.

[0026] <Operation> The voice input / output processing of the first embodiment will be described with reference to Figures 3(a) and 3(b). This processing starts when the contact center server 300 distributes a call from a customer to a specific operator U10 using the call control processing unit 302, and a session is established between the customer communication device 400 and the operator communication device 200. As a premise, at the seat of the operator U10 to whom the call is distributed, the voice output destination of the directional speaker function 110 is adjusted to the audible range of the operator U10. In addition, the range of sound collection by the area sound collection device 120 is adjusted to include the seat position of the operator U10.

[0027] First, when a customer speaks, the contact center server 300 transmits the voice signal of the customer's speech from the customer communication device 400 to the operator communication device 200 of the assigned operator U10 (step A1). Next, the voice output unit 201 of the operator communication device 200 transmits the voice signal received from the contact center server 300 to the output control unit 112 of the voice input / output device 100 (step A2). Next, the voice input / output device 100 outputs the voice signal from the speaker 111 above the assigned operator U10 (step A3). Note that not only the customer's voice signal but also various voice signals generated by the contact center server 300 (e.g., ringtones, mechanical voices supporting voice calls with the customer) are output from the speaker 111 in the same way.

[0028] Furthermore, when the operator U10 speaks, the microphone array units 121, 121 facing the operator U10 acquire the voice signal of the operator U10's speech (step A4). Next, the sound pickup unit 122 of the voice input / output device 100 extracts voice within a pre-adjusted range and transmits it to the voice input unit 202 of the operator communication device 200 as the voice signal of the operator U10 (step A5). Next, the operator communication device 200 transmits the voice signal received from the voice input / output device 100 to the contact center server 300 (step A6). When the contact center server 300 transmits the voice signal to the customer communication device 400, the customer communication device 400 plays it back to the customer.

[0029] According to the first embodiment, the operator U10 can make voice calls with customers without wearing a headset or a hands-free microphone, which relieves the stress of the operator U10 who makes voice calls in contact center work and also relieves the operator from the burden of equipment maintenance. Furthermore, the customer's voice can be output only to the specific operator U10 to whom the call is distributed by the call control processing unit 302. This makes it possible to implement noise countermeasures for other operators U10 who do not make voice calls with the customer. Furthermore, since the operator U10 operates the operator communication device 200 to log in and perform contact center operations, if the identifier of the operator communication device 200, the identifier of the speaker 111, and the identifier of the microphone array units 121, 121 are linked, the customer's voice can be output from the speaker 111 only to the specific operator U10 to whom the call was distributed by the call control processing unit 302, and the voice of only that operator U10 can be input from the microphone array units 121, 121, even without managing the location information of the operator U10.

[0030] [Second embodiment] In describing the second embodiment, differences from the first embodiment will be mainly described, and a description of overlapping points will be omitted. The second embodiment uses an imaging device 131 shown in FIG.

[0031] The imaging device 131 is a device that captures an image of the face of the operator U10. The imaging device 131 is preferably placed on a side wall or ceiling of the work space so as to be able to capture at least the mouth of the operator U110, which is a sound source, and the ears, which are sound receiving parts.

[0032] (Functional configuration) As shown in Fig. 4, the voice input / output system of the second embodiment includes a voice input / output device 100, an operator communication device 200, and a contact center server 300. Compared to Fig. 2, the voice input / output device 100 further includes an image processing device 130 in addition to the directional speaker function 110 and the area sound collection device 120 already described.

[0033] (Image processing device 130) The image processing device 130 is a device that processes images captured by the imaging device 131. The image processing device 130 includes the imaging device 131 already described and an image processing recognition unit 132. The image processing recognition unit 132 has the function of extracting the positions of the ears and mouth from the video of the operator U10 captured by the imaging device 131, sending the ear position information to the output control unit 112, and sending the mouth position information to the sound collection unit 122.

[0034] The output control unit 112 corrects the sound transmission area of ​​the audio signal (directivity of the audio output) as needed based on the position information received from the image processing recognition unit 132. Specifically, the output control unit 112 adjusts the angle of the directivity of the audio output from the speaker 111, and directs the speaker 111 toward the ears of the operator U10 who is directly below. The sound collection unit 122 corrects the sound collection area of ​​the audio signal (directivity of the audio input) as needed based on the position information received from the image processing recognition unit 132. Specifically, the sound collection unit 122 adjusts the angle of the directivity of the audio input in the microphone array units 121, 121, and points the microphone array units 121, 121 toward the mouth of the operator U10 who is in the target area.

[0035] <Operation> The audio input / output process of the second embodiment will be described with reference to Figures 5(a) and 5(b). The start conditions and premise of this process are the same as those of Figures 3(a) and 3(b). In addition, the image processing recognition unit 132 processes images captured by the imaging device 131 as needed.

[0036] When a session is established between the customer communication device 400 and the operator communication device 200, the image processing recognition unit 132 recognizes the positions of the mouth and ears of the operator U10 to whom the call is distributed from the image to be processed. The image processing recognition unit 132 transmits the ear position information to the output control unit 112. The image processing recognition unit 132 also transmits the mouth position information to the sound collection unit 122.

[0037] When a customer speaks, the output control unit 112 corrects the orientation of the speaker 111 above the head of the responsible operator U10 based on the ear position information received from the image processing recognition unit 132 (step B1). Thereafter, the processes of steps A1 to A3 in FIG. 3(a) are performed.

[0038] Furthermore, when the operator U10 speaks, the sound collection unit 122 corrects the orientation of the microphone array units 121, 121 directed toward the operator U10 in charge based on the mouth position information received from the image processing recognition unit 132 (step B2). Thereafter, the processes of steps A4 to A6 in FIG. 3(b) are performed.

[0039] According to the second embodiment, in addition to the effects of the first embodiment, the sound transmission area and sound collection area can be adjusted at any time in accordance with the movement of the operator U10, so that even if the position of the operator U10 changes, the customer's voice can be reliably delivered to the operator U10 and the sound can be collected for the operator U10.

[0040] [Correspondence to claims] The contact center server 300 is an example of the "contact center device" in the claims. The call control processor 302 is an example of the "automatic call distribution processor" in the claims. The operator communication device 200 is an example of a "processing device" in the claims. The customer communication device 400 is an example of a "customer device" in the claims. The voice signal from the customer is an example of the "first voice information" in the claims. Microphone array unit 121 is an example of the "microphone" in the claims. The audio output unit 201 is an example of the "first processing unit" in the claims. The voice signal from the operator U10 is an example of the "second voice information" in the claims. The voice input unit 202 is an example of the "second processing unit" in the claims.

[0041] [Variations] (a): The customer's voice may be output from the speaker 111 without going through the voice output unit 201 of the operator communication device 200, or the voice of the operator U10 input from the microphone array unit 121 may be sent to the customer communication device 400 without going through the voice input unit of the operator communication device 200. In other words, the contact center server 300 may be communicably connected to the voice input / output device 100, and the contact center server 300 may control the speaker 111 and the microphone array unit 121. (b) The voice input / output device 100 may operate the microphone array unit 121 only while the operator U10 is speaking, and may process the microphone array unit 121 not to operate or the sound collection unit 122 not to collect sound when the operator U10 is not speaking. By performing such processing, it is possible to avoid a situation in which the microphone array unit 121 collects the customer's voice output from the speaker 111 and transmits it to the customer communication device 400. (c) An imaging device 131 may be provided for each operator U10, and each imaging device 131 may capture an image of the face of the operator U10 to acquire information about the mouth position and the ear position. Alternatively, the imaging device 131 may be mounted on the operator communication device 200.

[0042] (d): The above-mentioned invention-specific features can be combined as appropriate. (e): Means that can be realized in software can be realized in hardware, and means that can be realized in hardware can be realized in software. [Explanation of symbols]

[0043] U10 Operator 100 Audio input / output device 110 Directional speaker function 111 Speaker 112 Output control section 120 Area Sound Collection Device 121 Microphone array section 122 Sound pickup section 123 directional beam 130 Image processing device 131 Imaging device 132 Image processing recognition unit 200 Operator Communication Device 201 Audio output unit 202 Audio input section 300 Contact Center Servers 301 IVR Processing Department 302 Call control processing unit 303 Communication Processing Unit 400 Customer Communication Device 500 In-house communication network 600 Public Network

Claims

1. A voice input / output system including a processing device selected from a plurality of processing devices used by a plurality of operators based on an automatic call distribution processing unit of a contact center device, and a voice input / output device having a plurality of speakers and a plurality of microphones, Each of the processing devices in the plurality of processing devices comprises: a voice output unit that transmits first voice information received from the customer device to the voice input / output device; a voice input unit that transmits second voice information to the customer device; The audio input / output device is outputting the first audio information received from each of said processing devices from each of said speakers to each operator using each of said processing devices; an output control unit that adjusts the angle of directivity of the audio output from the plurality of speakers based on the position information of the ears of the plurality of operators acquired from an imaging device, thereby correcting the sound transmission area of ​​the audio signal and directing each of the speakers to the ears of each of the operators in the target sound transmission area; to transmit the second voice information input by each of the operators to each of the microphones to each of the processing devices used by each of the operators, a sound collection unit that adjusts the angle of directivity of the sound input of the plurality of microphones based on the position information of the mouths of the plurality of operators acquired from the imaging device, thereby correcting the sound collection area of ​​the sound signal and directing each of the microphones to the mouth of each of the operators in the target sound collection area, The output control unit When the distance between adjacent operators is equal to or less than a predetermined distance, a first speaker of the plurality of speakers is directed toward the ear of a target operator among the adjacent operators, and the first audio information is output, and a second speaker of the plurality of speakers is directed toward the ear of the adjacent operator, and an audio whose phase is shifted by half a wavelength from the wavelength of the output first audio information is output; The sound collection unit is Two or more of the microphones having different sound pickup directions are directed toward the mouth of the target operator, and a sound component included in each of the sounds picked up by the two or more microphones is subtracted using a spectral subtraction method to extract the sound component of the second sound information. Audio input / output system.

2. A voice input / output method in a voice input / output system including a processing device selected from a plurality of processing devices used by a plurality of operators, and a voice input / output device including a plurality of speakers and a plurality of microphones, based on an automatic call distribution processing unit of a contact center device, Each of the processing devices in the plurality of processing devices a voice output step of transmitting first voice information received from the customer device to the voice input / output device; a voice input step of transmitting second voice information to the customer device; The audio input / output device outputting the first audio information received from each of said processing devices from each of said speakers to each operator using each of said processing devices; an output control step of correcting a sound transmission area of ​​an audio signal by adjusting an angle of directivity of audio output from the plurality of speakers based on position information of the ears of the plurality of operators acquired from an imaging device, and directing each of the speakers to the ears of each of the operators in the target sound transmission area; to transmit the second voice information input by each of the operators to each of the microphones to each of the processing devices used by each of the operators, a sound collection step of adjusting the angle of directivity of the sound input of the plurality of microphones based on the position information of the mouths of the plurality of operators acquired from the imaging device, thereby correcting the sound collection area of ​​the sound signal and directing each of the microphones to the mouths of the respective operators in the target sound collection area; In the output control step, When the distance between adjacent operators is equal to or less than a predetermined distance, a first speaker of the plurality of speakers is directed toward the ear of a target operator among the adjacent operators, and the first audio information is output, and a second speaker of the plurality of speakers is directed toward the ear of the adjacent operator, and an audio whose phase is shifted by half a wavelength from the wavelength of the output first audio information is output; In the sound collection step, Two or more of the microphones having different sound pickup directions are directed toward the mouth of the target operator, and a sound component included in each of the sounds picked up by the two or more microphones is subtracted using a spectral subtraction method to extract the sound component of the second sound information. Audio input / output method.

Images