Information processing device, information processing method, and program

Abstract

To support grasping of an interaction content.SOLUTION: An information processing device includes: a specification unit for specifying a scene representing a scene of interaction when utterance is made based on utterance in prescribed interaction; and a creation unit for creating visualization information for visualizing a time series of utterance in interaction, a time series of the scenes and a relation between the scenes based on relationship definition information defining the scenes having relationship. The relationship definition information includes at least hierarchy structure definition in which the scenes with a parent-child relation are defined and relation definition defining the scenes in a depending relation. The creation unit creates visualization information for giving a component which hides a second scene and displays the second scene according to selection operation of a user to a first scene to be displayed when one or more second scenes having the first scene as a parent continue immediately after the first scene in the time series of the scenes.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program.

Background Art

[0002] In a contact center (also referred to as a call center), generally, an operation called after call work (ACW) is performed. ACW is work performed after the end of a telephone response to a customer, and is post-processing such as, for example, creation of a response record, ordering work for products or services, and the like.

[0003] While ACW is an important operation, since it is a time when customer response is not possible, it is required to improve the efficiency of ACW. On the other hand, as a technique for improving the efficiency of creating a response record, a technique is known in which speech in a call with a customer is converted into text by a speech recognition technique, and the scene of the speech is specified from the text (for example, Patent Document 1, etc.). Here, a scene is a scene of a dialogue being conducted between an operator and a customer, and examples include "opening" representing a scene such as an initial greeting, "inquiry grasping" representing a scene for grasping the content of an inquiry from a customer, "response" representing a case where a response and countermeasures are taken for the content of the inquiry, "closing" representing a scene such as a final greeting, and the like.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, if various topics are discussed during a call, multiple scenes are identified, making it difficult to grasp the relationships between scenes (e.g., structural relationships between scenes, connections between scenes, etc.). This makes it difficult to understand the overall content of the call, and for example, it may take a long time to create a call record.

[0006] One embodiment of the present invention has been made in view of the above points and aims to support the understanding of the content of a dialogue. [Means for solving the problem]

[0007] To achieve the above objective, an information processing device according to one embodiment includes: an identification unit that identifies a scene representing the scene of a dialogue at the time an utterance was made, based on an utterance in a predetermined dialogue; and a creation unit that creates visualization information for visualizing the time series of the utterance in the dialogue, the time series of the scenes, and the relationships between the scenes, based on relationship definition information that defines related scenes, wherein the relationship definition information includes at least a hierarchical structure definition that defines scenes that have a parent-child relationship and an association definition that defines scenes that have a dependency relationship, and the creation unit creates visualization information for displaying the second scenes by adding a component to the first scene that hides the second scenes and displays the second scenes in response to a user selection operation, if in the time series of the scenes one or more second scenes that have the first scene as their parent immediately follow the first scene. [Effects of the Invention]

[0008] It can help in understanding the content of the conversation. [Brief explanation of the drawing]

[0009] [Figure 1] This figure shows an example of the overall configuration of a contact center system according to the first embodiment. [Figure 2] This figure shows an example of the functional configuration of a contact center system according to the first embodiment. [Figure 3]This diagram illustrates an example of a hierarchical structure definition included in the definition information. [Figure 4] This is a diagram illustrating an example of related definitions included in the definition information. [Figure 5] This is a flowchart showing an example of call visualization processing according to the first embodiment. [Figure 6] This is a diagram (part 1) illustrating an example of an operator screen that visualizes the hierarchical structure of a scene. [Figure 7] This is a diagram (part 2) illustrating an example of an operator screen that visualizes the hierarchical structure of a scene. [Figure 8] This diagram illustrates an example of an operator screen that visualizes the relationships between scenes. [Figure 9] This diagram illustrates an example of an operator screen that visualizes the relationships between scenes and supplementary information. [Figure 10] This diagram schematically shows a specific example of one variation. [Figure 11] This figure shows an example of the functional configuration of a contact center system according to the second embodiment. [Figure 12] This is a flowchart showing an example of call visualization processing according to the second embodiment. [Figure 13] This diagram illustrates an example of an operator screen that visualizes scene-specific talk scripts as support information. [Figure 14] This diagram illustrates an example of an operator screen that visualizes the next scene candidate and its transition probability as supporting information. [Modes for carrying out the invention]

[0010] Hereinafter, as an embodiment of the present invention, the first embodiment and the second embodiment will be described. Hereinafter, in each embodiment, a contact center system 1 that can assist in grasping the conversation content (that is, the call content) in a call between an operator of a contact center and a customer will be described. However, the contact center is an example, and it can be similarly applied to cases where, for example, when grasping the call content of a person in charge working in an office or the like. Also, the call is not limited to between two parties and may be made among three or more parties.

[0011] Furthermore, hereinafter, it is assumed that the operator of the contact center makes a voice call with the customer, but it is not limited to this. For example, it can be similarly applied when text chats (including those that can send and receive stamps, attached files, etc. in addition to text), video calls, etc. are made.

[0012] [First Embodiment] First, the first embodiment will be described. In this embodiment, mainly for the purpose of improving the efficiency of ACW such as creating response records, the case where the relationship between scenes is also visualized to facilitate grasping the call content will be described. Here, ACW is generally work related to post-processing performed after the end of a call with a customer (that is, offline), and in addition to creating response records, it also includes, for example, ordering work for products and services.

[0013] [Overall Configuration] Fig. 1 shows an example of the overall configuration of the contact center system 1 according to this embodiment. As shown in Fig. 1, the contact center system 1 according to this embodiment includes a call visualization device 10, a voice recognition system 20, an operator terminal 30, a manager terminal 40, a PBX (Private Branch Exchange) 50, and a customer terminal 60. Here, the call visualization device 10, the voice recognition system 20, the operator terminal 30, the manager terminal 40, and the PBX 50 are installed in a contact center environment E which is a system environment of the contact center. Note that the contact center environment E is not limited to a system environment within the same building, and may be, for example, a system environment within a plurality of geographically separated buildings.

[0014] The call visualization device 10 creates information (hereinafter also referred to as visualization information) for visualizing the utterances in a call between a customer and an operator, the scene of the utterance, and the relationship between scenes, and transmits the visualization information to the operator terminal 30 (or the manager terminal 40 may also be used). The visualization information is information for displaying an operator screen or the like, which will be described later, on the display of the operator terminal 30, and is, for example, screen information defined by HTML (Hypertext Markup Language), CSS (Cascading Style Sheets), or the like.

[0015] The voice recognition system 20 performs voice recognition on a call between a customer and an operator, and converts the utterances during the call into text (character strings). Here, hereinafter, it is assumed that voice recognition is performed on both the customer's utterance and the operator's utterance, but it is not limited to this, and for example, voice recognition may be performed only on either one of the utterances.

[0016] The operator terminal 30 is various terminals such as a PC (Personal Computer) used by an operator who responds to inquiries from customers, and functions as an IP (Internet Protocol) telephone.

[0017] Administrator terminals 40 are various terminals such as PCs used by administrators (such administrators are also called supervisors) who manage the operators.

[0018] PBX50 is a telephone exchange (IP-PBX) connected to a communications network 70, which includes VoIP (Voice over Internet Protocol) networks and PSTN (Public Switched Telephone Network).

[0019] Customer terminals 60 are various devices used by customers, such as smartphones, mobile phones, and landline phones.

[0020] Note that the overall configuration of the contact center system 1 shown in Figure 1 is just one example, and other configurations are possible. For example, in the example shown in Figure 1, the call visualization device 10 is included in the contact center environment E (i.e., the call visualization device 10 is on-premise), but all or part of the functions of the call visualization device 10 may be implemented by cloud services or the like. Similarly, in the example shown in Figure 1, the speech recognition system 20 is on-premise, but all or part of the functions of the speech recognition system 20 may be implemented by cloud services or the like. Also similarly, in the example shown in Figure 1, the PBX 50 is an on-premise telephone exchange, but it may be implemented by cloud services.

[0021] Furthermore, although the operator terminal 30 is said to function as an IP phone, for example, a separate telephone may be included in the contact center system 1 in addition to the operator terminal 30.

[0022] <Functional Configuration> Figure 2 shows the functional configuration of the call visualization device 10, operator terminal 30, and administrator terminal 40 included in the contact center system 1 according to this embodiment.

[0023] ≪Call visualization device 10≫ As shown in Figure 2, the call visualization device 10 according to this embodiment includes a scene identification unit 101 and a visualization information creation unit 102. Each of these units is realized, for example, by processing that one or more programs installed in the call visualization device 10 cause a processor such as a CPU (Central Processing Unit) to execute. The call visualization device 10 according to this embodiment also includes a relationship definition information storage unit 110 and a call history information storage unit 120. Each of these units can be realized, for example, by a storage device such as an HDD (Hard Disk Drive), SSD (Solid State Drive), or flash memory.

[0024] The scene identification unit 101 identifies the scene for each utterance in a call based on the speech recognition results for the call between the customer and the operator (i.e., the text representing the customer's utterances and the text representing the operator's utterances). Known scene identification or scene classification techniques can be used to identify the scene of each utterance. For example, the technique described in Patent Document 1 can be used to identify the scene for each utterance.

[0025] Furthermore, the scene identification unit 101 stores time-series information, which associates the speaker (customer or operator), the text representing the utterance, and the scene in which the utterance occurred, as call history information 121 in the call history information storage unit 120. The call history information 121 also includes information such as a call ID to identify the call, the operator ID of the operator who handled the call, and the date and time of the call.

[0026] Here, a "scene" refers to a conversation between an operator and a customer, and the types of scenes that exist are predetermined. Typical scenes include, for example, "opening," which represents the initial greeting; "understanding the inquiry," which represents the understanding of the customer's inquiry; "responding," which represents the answering and handling of the inquiry; and "closing," which represents the final greeting.

[0027] Furthermore, an utterance is a single segment of speech (or text representing the result of speech recognition of that speech). The range of a segment can be arbitrarily set, but for example, it is possible to define a segment as the end unit of speech as described in Patent Document 1. An end unit of speech is a unit of speech that the speaker wants to speak, and for example, it is the range that is separated by a period "." or a question mark "?" when the speech is converted into text by speech recognition.

[0028] The visualization information creation unit 102 creates visualization information to visualize the text representing each speaker's utterance, the scene in which that utterance occurred, and the relationships between those scenes, based on the relationship definition information 111 stored in the relationship definition information storage unit 110. The visualization information creation unit 102 then transmits this visualization information to the operator terminal 30 (or administrator terminal 40). The relationship definition information 111 is information that defines the relationships between scenes (for example, structural relationships between scenes, relationships between scenes, etc.). Specific examples of the relationship definition information 111 will be described later.

[0029] The relationship definition information storage unit 110 stores the relationship definition information 111. The call history information storage unit 120 stores the call history information 121. The relationship definition information 111 is created in advance and stored in the relationship definition information storage unit 110.

[0030] Here, the relationship definition information 111 includes at least a hierarchical structure definition that defines the hierarchical structural relationships (parent-child relationships) between scenes, and an association definition that defines the relationships between scenes.

[0031] Figure 3 shows an example of a hierarchical structure definition. The example in Figure 3 defines the structural relationships between three scenes: "accident situation assessment," "injury status assessment," and "self-driving capability assessment." Specifically, "accident situation assessment" is defined as the parent, and "injury status assessment" and "self-driving capability assessment" are defined as children. Such parent-child relationships are defined based on, for example, semantic inclusion relationships or conceptual hierarchies between scenes.

[0032] An example of a relationship definition is shown in Figure 4. The example in Figure 4 defines the relationship between two scenes, "Option Cancellation" and "Billing Notice." Specifically, it defines a relationship that indicates that if "Option Cancellation" appears during a call, then "Billing Notice" must also appear. Such relationships are defined based on, for example, dependencies between scenes. Dependencies between scenes refer to relationships where, if one scene appears during a call, then another scene must also appear.

[0033] In addition to hierarchical structure definitions and association definitions, various other relationships may be defined in the relationship definition information 111. For example, in order to link identical scenes when they appear multiple times, an association definition may be given stating that identical scenes are related to each other. Also, for example, parallel relationships and opposite relationships that indicate semantic opposites may be defined. Furthermore, in addition to these, for example, a relationship may be defined that connects a scene that was interrupted midway with a scene that resumed it. The relationship that connects a scene that was interrupted midway with a scene that resumed it will be explained in more detail in the modified examples described later.

[0034] ≪Operator Terminal 30≫ As shown in Figure 2, the operator terminal 30 according to this embodiment has a UI control unit 301. The UI control unit 301 is realized, for example, by a process in which one or more programs installed on the operator terminal 30 cause a processor such as a CPU to execute.

[0035] The UI control unit 301 displays the operator screen, described later, on the display based on the visualization information received from the call visualization device 10.

[0036] ≪Administrator Terminal 40≫ As shown in Figure 2, the administrator terminal 40 according to this embodiment has a UI control unit 401. The UI control unit 401 is realized, for example, by a process that one or more programs installed on the administrator terminal 40 cause a processor such as a CPU to execute.

[0037] Based on the visualization information received from the call visualization device 10, the UI control unit 401 displays a screen on the display similar to the operator screen described later (this screen may also be called the supervisor screen or administrator screen, etc.).

[0038] <Call visualization processing> The following describes, with reference to Figure 5, the process of displaying the operator screen on the operator terminal 30's display after a call with a customer has ended, in order to improve ACW efficiency, and to visualize the content of the call. In the following, it is assumed that the voice recognition system 20 performs voice recognition on the call between the customer and the operator, and that the voice recognition result is transmitted to the call visualization device 10.

[0039] First, the scene identification unit 101 identifies the scene for each utterance in the call based on the speech recognition results received from the speech recognition system 20 (step S101). This creates call history information 121, which associates the speaker (customer or operator), the text representing the utterance, and the scene of that utterance, and stores it in the call history information storage unit 120. As mentioned above, the scene identification unit 101 can identify the scene for each utterance using known scene identification techniques or scene classification techniques, such as the technique described in Patent Document 1.

[0040] Next, the visualization information creation unit 102 creates visualization information to visualize the text representing each speaker's utterance, the scene of that utterance, and the relationships between scenes, based on the relationship definition information 111 stored in the relationship definition information storage unit 110 (step S102).

[0041] Then, the visualization information creation unit 102 transmits the visualization information created in step S102 to the operator terminal 30 of the operator who answered the call (step S103). As a result, the UI control unit 301 displays the operator screen, which will be described later, on the display of the operator terminal 30 based on the visualization information received from the call visualization device 10.

[0042] In step S103 described above, visualization information was sent to the operator terminal 30. However, visualization information may also be sent to the administrator terminal 40 in response to a request from the administrator terminal 40, for example. In this case, the UI control unit 401 will display the supervisor screen on the administrator terminal 40's display based on the visualization information received from the call visualization device 10.

[0043] <Operator screen> The following describes, as an example, the operator screen displayed on the operator terminal 30's screen.

[0044] • Operator screen example 1-1 Based on the relationship definition information 111 stored in the relationship definition information storage unit 110, visualization information is created that visualizes the hierarchical structure of the relationships between scenes. At this time, an example of an operator screen displayed by the UI control unit 301 based on this visualization information will be explained with reference to Figures 6 and 7.

[0045] The operator screen 1000 shown in Figure 6 includes a scene display area 1100 and a speech display area 1200. The scene display area 1100 displays scene buttons corresponding to the scenes identified during the call in chronological order. The speech display area 1200 displays the speech from the scene corresponding to the scene button selected from the scene display area 1100 in chronological order.

[0046] In the example shown in Figure 6, the scene display area 1100 shows a scene button 1110 corresponding to "Opening," a scene button 1120 corresponding to "Understanding the Request," a scene button 1130 corresponding to "Assessing the Accident Situation," a scene button 1140 corresponding to "Insurance Handling," and a scene button 1150 corresponding to "Closing."

[0047] Furthermore, the scene button 1130 corresponding to "Accident Situation Assessment" is provided with an expand button 1131 to display scene buttons corresponding to scenes that have "Accident Situation Assessment" as their parent (in other words, child scenes of "Accident Situation Assessment"). Similarly, the scene button 1140 corresponding to "Insurance Handling" is provided with an expand button 1141 to display scene buttons corresponding to scenes that have "Insurance Handling" as their parent (in other words, child scenes of "Insurance Handling").

[0048] For example, if the expand button 1131 is selected by the operator, as shown in Figure 7, the scene button 1160 corresponding to the scene "Injury Status Assessment" which has "Accident Situation Assessment" as its parent, and the scene button 1170 corresponding to the scene "Self-Driving Capability Assessment" which has "Accident Situation Assessment" as its parent, will be displayed. This indicates that "Accident Situation Assessment" appeared as the scene during the call, followed by "Injury Status," and then "Self-Driving Capability Assessment." At this time, the expand button 1131 changes to the compress button 132, which hides the scene buttons 1160 and 1170 and returns to the state shown in Figure 6.

[0049] As described above, in the operator screen according to this embodiment, if there are parent-child scenes among the scenes identified during a call, and one or more child scenes appear immediately after the parent scene, the child scenes are hidden, and an expand button to display these child scenes is added to the scene button corresponding to the parent scene. As a result, even if the scene structure during a call has a complex hierarchical structure, only the scene at the highest level of the hierarchy is displayed, making it easy for the operator to grasp the overall scene structure of the call.

[0050] Furthermore, if the operator wants to see a more detailed scene structure (i.e., the scene structure including lower-level scenes), they can press the expand button to display the subscenes of the scene to which that expand button is assigned. This allows the operator to easily grasp the overall scene structure of the call and, when necessary, to check a more detailed scene structure.

[0051] In the examples shown in Figures 6 and 7, "Accident Situation Assessment" has "Injury Status Assessment" and "Self-Driving Ability Assessment" as subscenes, and the hierarchical structure between scenes is two levels high. However, this is just one example, and the hierarchical structure between scenes can have three or more levels. For example, in a three-level structure where a subscene of one scene has further subscenes (grandchild scenes), the scene button corresponding to the subscene of that scene will be given an expand button to display the grandchild scenes. The same applies to structures with four or more levels.

[0052] • Operator screen example 1-2 Based on the relationship definition information 111 stored in the relationship definition information storage unit 110, visualization information is created that visualizes the relationships between scenes. At this time, an example of an operator screen displayed by the UI control unit 301 based on this visualization information will be explained with reference to Figure 8.

[0053] The operator screen 2000 shown in Figure 8 includes a scene display area 2100 and a speech display area 2200. The scene display area 2100 displays scene buttons corresponding to the scenes identified during the call in chronological order. The speech display area 2200 displays the speech from the scene corresponding to the scene button selected from the scene display area 2100 in chronological order.

[0054] In the example shown in Figure 8, the scene display area 2100 displays the following scene buttons: scene button 2110 corresponding to "Opening", scene button 2120 corresponding to "Understanding the Request", scene button 2130 corresponding to "Identity Verification", scene button 2140 corresponding to "Change Shipping Address", scene button 2150 corresponding to "Understanding the Request", scene button 2160 corresponding to "Cancellation of Option", scene button 2170 corresponding to "Billing Information", and scene button 2180 corresponding to "Closing".

[0055] Furthermore, scene button 2160, which corresponds to "Cancel Option," and scene button 2170, which corresponds to "Billing Notice," are connected by a connection line 2310. This indicates that "Cancel Option" and "Billing Notice" are related scenes.

[0056] Furthermore, scene button 2120, which corresponds to "Understanding the Request," and scene button 2150, which also corresponds to "Understanding the Request," are connected by a connection line 2320. This indicates that "Understanding the Request" appears multiple times.

[0057] Thus, in the operator screen according to this embodiment, if there are related scenes among the scenes identified during a call (including cases where there are multiple identical scenes), those scenes are connected by connection lines. This makes it possible for the operator to easily understand related scenes within the overall call scene.

[0058] Furthermore, if there are scenes with parent-child relationships among the scenes identified during a call, and one or more child scenes appear immediately after a parent scene, the child scenes will be hidden, and an expand button to display these child scenes will be added to the scene button corresponding to the parent scene. For example, in the example shown in Figure 8, an expand button has been added to the scene button 2160 corresponding to "Option Cancellation".

[0059] Furthermore, while the example shown in Figure 8 illustrates the relationship between two scenes, similarly, scenes with three or more related elements are also connected by connection lines.

[0060] • Operator screen example 1-3 Here, the relationship definition information 111 may define not only the relationships between scenes, but also conditions such as visualizing some kind of information (which we will call supplementary information) when a certain relationship is not met. In the following, we will assume that visualization information has been created that visualizes supplementary information when there are multiple identical scenes, or when one of two dependent scenes does not exist, in addition to the relationships between scenes. At this time, an example of an operator screen displayed by the UI control unit 301 based on this visualization information will be explained with reference to Figure 9.

[0061] The operator screen 3000 shown in Figure 9 includes a scene display area 3100, an utterance display area 3200, and a supplementary information display area 3300. The scene display area 3100 displays scene buttons corresponding to the scenes identified during the call in chronological order. The utterance display area 3200 displays utterances from the scenes corresponding to the scene buttons selected from the scene display area 3100 in chronological order. The supplementary information display area 3300 displays supplementary information.

[0062] In the example shown in Figure 9, the scene display area 3100 displays the following scene buttons: scene button 3110 corresponding to "Opening", scene button 3120 corresponding to "Understanding the Request", scene button 3130 corresponding to "Identity Verification", scene button 3140 corresponding to "Change Shipping Address", scene button 3150 corresponding to "Understanding the Request", scene button 3160 corresponding to "Cancel Option", and scene button 3170 corresponding to "Closing".

[0063] Furthermore, scene button 3120, which corresponds to "Understanding the Request," and scene button 3150, which also corresponds to "Understanding the Request," are connected by a connection line 3310. On the other hand, although scene button 3160, which corresponds to "Cancellation of Option," exists, the scene "Billing Notice," which is related to "Cancellation of Option," does not appear.

[0064] Therefore, the supplementary information display field 3300 shows the supplementary information "Multiple requests have been identified" and the supplementary information "The necessary scene (invoice guidance) could not be identified." In the example shown in Figure 9, the supplementary information is further categorized according to its importance, with "Multiple requests have been identified" being at the INFO level and "The necessary scene (invoice guidance) could not be identified" being at the WARN level.

[0065] As described above, the operator screen according to this embodiment displays supplementary information when the conditions defined in the relationship definition information 111 are not met (for example, when a certain relationship is not met). This makes it easy for the operator to understand, for example, when there are multiple identical scenes or when a certain scene among multiple dependent scenes is missing.

[0066] <Summary of the First Embodiment> As described above, the contact center system 1 according to this embodiment primarily aims to improve the efficiency of ACW by displaying an operator screen on the operator terminal 30 after the end of a call with a customer, which makes it easy for the operator to understand the content of the call. This operator screen visualizes not only the timeline of the utterances during the call and the timeline of the scenes in which those utterances occurred, but also the relationships between scenes (e.g., hierarchical structure between scenes, relationships between scenes, etc.) and supplementary information. As a result, the operator can easily understand the content of the call necessary for ACW.

[0067] In this embodiment, the focus was primarily on the offline display of the operator screen for the purpose of improving ACW efficiency, but it can also be applied to online situations (i.e., during a call with a customer). That is, even on the operator screen displayed during a call with a customer, the relationships between scenes (e.g., hierarchical structure between scenes, relationships between scenes, etc.) and supplementary information described in this embodiment may be visualized.

[0068] · Modification In a customer call, it is possible to transition to a certain scene A, then to one or more other scenes, and then back to the original scene A. In this case, scene A is interrupted, one or more other scenes are transitioned to, and then scene A is resumed. In this modified example, it is assumed that at least the relationship definition information 111 defines a relationship that connects the interrupted scene with the scene from which it was resumed.

[0069] A concrete example is when the scenes transition in the following order: (1) to (8). In this case, it means that (2), "Understanding the purpose (requesting a change of address)," was interrupted and resumed at (4).

[0070] (1) Opening (2) Understanding the purpose of the request (request for change of address) (3) Identity verification (4) Understanding the purpose of the request (request for address change) (5) Support (Address change support) (6) Understanding the purpose (confirming the contents of the invoice) (7) Action (Confirmation of billing details) (8) Closing At this time, in the scene display section of the operator screen, the scene button corresponding to (2) "Understanding the matter (request for address change)" and the scene button corresponding to (4) "Understanding the matter (request for address change)" are connected by a relationship line.

[0071] Furthermore, (6) "Understanding the Request (Confirming Invoice Details)" and (2) and (3) "Understanding the Request (Requesting Address Change)" can be considered, for example, scenes that have the same scene "Understanding the Request" as their parent. For this reason, the scene button corresponding to (6) "Understanding the Request (Confirming Invoice Details)", the scene button corresponding to (2) "Understanding the Request (Requesting Address Change)", and the scene button corresponding to (3) "Understanding the Request (Requesting Address Change)" may be connected by a connection line. In this case, these three scene buttons may be connected in parallel by a connection line, or the scene buttons for (2) and (3) "Understanding the Request (Requesting Address Change)" may be connected to each other by a connection line, and then the two scenes connected by this connection line may be connected to the scene button for (6) "Understanding the Request (Confirming Invoice Details)" by a connection line.

[0072] Specifically, if the scene button corresponding to (2) "Understanding the Request (Request for Address Change)" is A1, the scene button corresponding to (3) "Understanding the Request (Request for Address Change)" is A2, the scene button corresponding to (4) "Understanding the Request (Request for Address Change)" is A', and the connecting line is denoted as "-", then the connection can be made as "A1-A2-A'" or as "(A1-A2)-A'". A concrete example of connecting the scene buttons with a connecting line as "(A1-A2)-A'" is shown in Figure 10.

[0073] [Second Embodiment] Next, a second embodiment will be described. In this embodiment, the main purpose is to support operators online, and the description will focus on displaying an operator screen to help them understand the content of calls and assist them in responding to customers. In the following, the differences from the first embodiment will be mainly described, and the same or similar components as in the first embodiment will not be described.

[0074] <Functional Configuration> Figure 11 shows the functional configuration of the call visualization device 10, operator terminal 30, and administrator terminal 40 included in the contact center system 1 according to this embodiment.

[0075] ≪Call visualization device 10≫ As shown in Figure 11, the call visualization device 10 according to this embodiment includes a scene identification unit 101, a visualization information creation unit 102, and a support information acquisition / creation unit 103. Each of these units is realized, for example, by a process executed by a processor such as a CPU, which is run by one or more programs installed in the call visualization device 10. The call visualization device 10 according to this embodiment also includes a relationship definition information storage unit 110, a call history information storage unit 120, and a talk script storage unit 130. Each of these units can be realized, for example, by a storage device such as an HDD, SSD, or flash memory.

[0076] The visualization information creation unit 102 creates visualization information based on the relationship definition information 111 stored in the relationship definition information storage unit 110 and the support information acquisition / creation unit 103 that has acquired or created it. The visualization information creation unit 102 then transmits the visualization information to the operator terminal 30 (or administrator terminal 40). Support information refers to information that supports customer service. In this embodiment, the support information is assumed to be the talk script 131, which will be described later, and transition probability information to the next scene candidate.

[0077] The support information acquisition / creation unit 103 acquires or creates support information based on the current scene identified by the scene identification unit 101. That is, for example, if a talk script 131 is assumed as support information, the support information acquisition / creation unit 103 acquires the talk script 131 for the current scene from the talk script storage unit 130. On the other hand, for example, if transition probability information to the next scene candidate is assumed as support information, the support information acquisition / creation unit 103 creates transition probability information to the next scene candidate for the current scene based on the call history information 121. Here, since the call history information 121 is time-series information that associates the speaker, the text representing their utterance, and the scene of that utterance, the support information acquisition / creation unit 103 can statistically calculate the transition probability from the current scene to the next scene candidate using multiple call history information 121 stored in the call history information storage unit 120. For this reason, the support information acquisition / creation unit 103 creates the next scene candidate and its transition probability as support information.

[0078] The talk script memory unit 130 stores the talk script 131. The talk script 131 is a set of exemplary utterances (so-called scripts) for the operator, determined for each scene.

[0079] <Call visualization processing> The following describes, with reference to Figure 12, a process for displaying an operator screen on the operator terminal 30's display during a call with a customer, and visualizing information to support customer interaction along with the content of the call, for the purpose of providing online operator support. In the following, it is assumed that the speech recognition system 20 performs speech recognition in real time (for example, after each utterance) for calls between the customer and the operator, and that the speech recognition results are also transmitted to the call visualization device 10 in real time.

[0080] First, the scene identification unit 101 identifies the scene of the utterance represented by the speech recognition result received from the speech recognition system 20 (step S201).

[0081] Next, the support information acquisition / creation unit 103 takes the scene identified in step S201 as the current scene and acquires or creates support information based on the current scene (step S202).

[0082] Next, the visualization information creation unit 102 creates visualization information based on the relationship definition information 111 stored in the relationship definition information storage unit 110 and the support information acquired or created in step S202 (step S203).

[0083] Then, the visualization information creation unit 102 transmits the visualization information created in step S203 to the operator terminal 30 of the operator handling the call (step S204). As a result, the UI control unit 301 displays the operator screen, which will be described later, on the display of the operator terminal 30 based on the visualization information received from the call visualization device 10.

[0084] In step S204 described above, visualization information was sent to the operator terminal 30. In addition, visualization information may also be sent to the administrator terminal 40 in response to a request from the administrator terminal 40. In this case, the UI control unit 401 will display the supervisor screen on the administrator terminal 40's display based on the visualization information received from the call visualization device 10.

[0085] <Operator screen> The following describes, as an example, the operator screen displayed on the operator terminal 30's screen.

[0086] • Operator screen example #2-1 It is assumed that visualization information has been created based on the relationship definition information 111 stored in the relationship definition information storage unit 110 and the talk script 131 (support information) of the scene (current scene) identified in step S201 above. At this time, an example of an operator screen displayed by the UI control unit 301 based on this visualization information will be explained with reference to Figure 13.

[0087] The operator screen 4000 shown in Figure 13 includes a scene display area 4100, an utterance display area 4200, and a talk script display area 4300. In the scene display area 4100, scene buttons corresponding to each scene are displayed in chronological order each time a scene is identified during the call. In the utterance display area 4200, the speech recognition results (text) from the speech recognition system 20 are displayed in chronological order for each utterance. The speech recognition results from the speech recognition system 20 are transmitted to the operator terminal 30 in real time. In the talk script display area 4300, the talk script 131 for the current scene (in the example shown in Figure 13, "Accident Situation Assessment") is displayed.

[0088] As described above, the operator terminal screen according to this embodiment displays the talk script 131 for the current scene during a call with a customer. This allows the operator to know what to say or confirm with the customer in the current scene.

[0089] In the example shown in Figure 13, each scene button displayed in the scene display area 4100 is not assigned an expand button or connected by a linking line. This is for simplicity only; as explained in the first embodiment, an expand button (or compress button) may be assigned or connected by a linking line.

[0090] Operator screen example #2-2 It is assumed that visualization information has been created based on the relationship definition information 111 stored in the relationship definition information storage unit 110 and the transition probability information (support information) of the next scene candidate for the scene (current scene) identified in step S201 above. At this time, an example of an operator screen displayed by the UI control unit 301 based on this visualization information will be explained with reference to Figure 14.

[0091] The operator screen 5000 shown in Figure 14 includes a scene display area 5100 and an utterance display area 5200. In the scene display area 5100, each time a scene is identified during the call, the scene buttons corresponding to that scene are displayed in chronological order, along with the next scene candidate and its transition probability. In the utterance display area 5200, the speech recognition results (text) from the speech recognition system 20 are displayed in chronological order for each utterance. The speech recognition results from the speech recognition system 20 are transmitted to the operator terminal 30 in real time.

[0092] In the example shown in Figure 14, the scene display area 5100 shows the scene button 5110 corresponding to "Opening," the scene button 5120 corresponding to "Understanding the Request," and the scene button 5130 corresponding to the current scene, "Understanding the Accident Situation." It also displays the next scene candidates "Insurance Handling" with a transition probability of "60%", the next scene candidate "Confirming Contact Information" with a transition probability of "15%", and the next scene candidate "Repair Shop" with a transition probability of "5%". In this example, the top three scene candidates with the highest transition probabilities and their transition probabilities are displayed.

[0093] As described above, the operator screen in this embodiment displays the next scene candidates for the current scene during a call with a customer, along with their transition probabilities. This allows the operator to know which scene to transition to next from the current scene.

[0094] In the example shown in Figure 14, the scene buttons displayed in the scene display area 5100 do not have an expand button or are connected by a linking line. This is for simplicity only, and as explained in the first embodiment, they may also have an expand button (or compress button) or be connected by a linking line.

[0095] Furthermore, the operator screen 5000 shown in Figure 14 may include an auxiliary information display area 5300 that displays auxiliary information such as points to note when transitioning to the scene candidate with the highest transition probability among the next scene candidates (for example, what needs to be explained to the customer). This allows the operator to know the points to note when transitioning to the next scene.

[0096] Furthermore, while the operator screen 5000 shown in Figure 14 displays the next scene candidate and its transition probability, the operator screen may be displayed in various ways using the transition probability information of the next scene candidate. For example, the size and color of the next scene candidate may be displayed differently depending on the transition probability of the next scene candidate or the importance of each scene as defined separately. Alternatively, instead of displaying the transition probability as a value, the next scene candidate may be classified into categories such as large, medium, and small, and the next scene candidate and its transition probability category may be displayed. Alternatively, only the next scene candidate whose transition probability is above a certain predetermined threshold (e.g., 0.3) may be displayed.

[0097] Furthermore, in addition to these, you could also set whether or not to display the next scene suggestions depending on the current scene. This is because the next scene after a particular scene is often almost predetermined. For example, since the scene after the "Opening" is almost always "Understanding the Objective," you could set it so that the next scene suggestions are not displayed when the current scene is the "Opening."

[0098] Furthermore, instead of immediately displaying the next scene options upon transitioning to the current scene, it might be better to display them after a certain amount of time has elapsed. This is because there is often no need to think about the next scene immediately after transitioning to the current scene.

[0099] <Summary of the second embodiment> As described above, the contact center system 1 according to this embodiment is primarily intended to support operators online by displaying an operator screen on the operator terminal 30 that allows operators to easily grasp the content of a call during a customer call and to assist in responding to the customer. On this operator screen, the talk script 131 is visualized in real time, as well as the next scene candidates and their transition probabilities. This makes it easier for operators to determine what to say online and what kind of customer response to provide. Needless to say, both the talk script 131 for the current scene and the next scene candidates and their transition probabilities for the current scene may also be visualized on the operator screen.

[0100] The present invention is not limited to the embodiments specifically disclosed above, and various modifications, changes, and combinations with known technologies are possible without departing from the scope of the claims.

[0101] This application is based on the basic application No. 2021-166652 filed in Japan on 11 October 2021, the contents of which are incorporated herein by reference in their entirety. [Explanation of symbols]

[0102] 1. Contact Center System 10 Call visualization device 20. Voice Recognition System 30 Operator terminals 40 Administrator terminals 50 PBX 60 Customer terminals 70 Communication Networks 101 Scene Identification Section 102 Visualization information creation department 103 Support Information Acquisition and Creation Department 110 Relationship definition information storage unit 111 Relationship Definition Information 120 Call history information storage unit 121 Call history information 130 Talk Script Storage Unit 131 Talk Scripts 301 UI Control Unit 401 UI Control Unit E Contact Center Environment

Claims

1. A identifying unit that identifies a scene representing the context of the dialogue at the time the utterance was made, based on the utterance in a predetermined dialogue, A creation unit creates visualization information to visualize the time series of the utterances and scenes in the dialogue, and the relationships between the scenes, based on relationship definition information that defines related scenes. It has, The aforementioned relationship definition information includes at least a hierarchical structure definition that defines parent-child relationships between scenes, and a relationship definition that defines dependency relationships between scenes. The aforementioned creation unit, An information processing device that, in the time series of the aforementioned scenes, if one or more second scenes that have the first scene as their parent immediately follow the first scene, hides the second scenes and creates visualization information for displaying the second scenes by adding a component to the first scene that displays the second scenes in response to a user selection operation.

2. A identifying unit that identifies a scene representing the context of the dialogue at the time the utterance was made, based on the utterance in a predetermined dialogue, A creation unit that creates visualization information for visualizing the time series of utterances in the dialogue, the time series of scenes, and the relationships between scenes. It has, The aforementioned creation unit, An information processing device that creates visualization information to display a predetermined warning when, in the time series of the aforementioned scenes, only some of the scenes among multiple dependent scenes exist.

3. A identifying unit that identifies a scene representing the context of the dialogue at the time the utterance was made, based on the utterance in a predetermined dialogue, A creation unit that creates visualization information for visualizing the time series of utterances in the dialogue, the time series of scenes, and the relationships between scenes. It has, The aforementioned creation unit, An information processing device that creates visualization information to display candidate scenes to transition from the current scene and the probability of transitions, based on the current scene in the time series of the aforementioned scenes and past dialogue history information.

4. A procedure for identifying a scene representing the context of a given dialogue, based on utterances in that dialogue, A creation procedure for creating visualization information to visualize the time series of utterances and scenes in the dialogue, and the relationships between scenes, based on relationship definition information that defines related scenes, The computer executes this, The aforementioned relationship definition information includes at least a hierarchical structure definition that defines parent-child relationships between scenes, and a relationship definition that defines dependency relationships between scenes. The above creation procedure is, An information processing method that, in the time series of the aforementioned scenes, if one or more second scenes that have the first scene as their parent immediately follow the first scene, hides the second scenes and creates visualization information for displaying the second scenes by adding a component to the first scene that displays the second scenes in response to a user selection operation.

5. A procedure for identifying a scene representing the context of a given dialogue, based on utterances in that dialogue, A procedure for creating visualization information to visualize the time series of utterances in the aforementioned dialogue, the time series of the aforementioned scenes, and the relationships between the aforementioned scenes, The computer executes this, The above creation procedure is, An information processing method for creating visualization information to display a predetermined warning when, in the time series of the aforementioned scenes, only some of the scenes among multiple dependent scenes exist.

6. A procedure for identifying a scene representing the context of a given dialogue, based on utterances in that dialogue, A procedure for creating visualization information to visualize the time series of utterances in the aforementioned dialogue, the time series of the aforementioned scenes, and the relationships between the aforementioned scenes, The computer executes this, The above creation procedure is, An information processing method for creating visualization information to display candidate scenes to transition from the current scene and the probability of transitions, based on the current scene in the time series of the aforementioned scenes and past dialogue history information.

7. A program that causes a computer to function as an information processing device according to any one of claims 1 to 3.

Images