Information processing device, information processing method, and program

Abstract

An information processing device according to one embodiment comprises: a specification unit for specifying, in accordance with the content of conversation of a plurality of persons, the whole or part of conversation data that represents the conversation, as a time authentication set range; and a setting unit for setting time authentication to the specified set range.

Description

[Technical Field]

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

[0002] A mechanism called time authentication has been known for some time (for example, Non-Patent Document 1). By performing time authentication on electronic data, it becomes possible to prove that the electronic data existed at a certain time and that the electronic data has not been tampered with since that time.

[0003] However, in recent years, processing of voice data has become easier, and there is a need to perform time authentication on voice data of calls generated in contact centers (also called call centers). [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Time Business Certification Center, Internet<URL:https: / / www.dekyo.or.jp / tb / index.html> Summary of the Invention [Problem to be solved by the invention]

[0005] However, the volume of calls made at contact centers is generally huge, and time-stamping all voice data would require more access to the Time Stamping Authority (TSA), increasing costs. Furthermore, in recent years, there has been a need to time-stamp not only voice data but also text data converted from that voice data using speech recognition, further increasing costs.

[0006] In contrast, by subjecting only the necessary range of call data, including voice data and text data, to time authentication, it is possible to reduce the cost required for time authentication.

[0007] An embodiment of the present invention has been made in view of the above points, and aims to reduce the cost required for time authentication. [Means for solving the problem]

[0008] In order to achieve the above-mentioned objective, an information processing device according to one embodiment has an identification unit that identifies part or all of the conversation data representing a conversation between multiple people as a setting range for time authentication depending on the content of the conversation, and a setting unit that sets time authentication for the identified setting range. [Effects of the Invention]

[0009] The cost required for time authentication can be reduced. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating an example of the overall configuration of a contact center system according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram for schematically explaining an example of time authentication settings. [Figure 3] FIG. 2 is a diagram illustrating an example of a functional configuration of an authentication management apparatus according to the present embodiment. [Figure 4] FIG. 10 is a diagram illustrating an example of a call reasons DB. [Figure 5] FIG. 10 is a diagram illustrating an example of a setting condition DB. [Figure 6] 10 is a flowchart showing an example of the flow of a time authentication process. [Figure 7] FIG. 10 is a diagram illustrating an example of a time authentication management DB. DETAILED DESCRIPTION OF THE INVENTION

[0011] An embodiment of the present invention will be described below. In this embodiment, a contact center system 1 will be described that targets a contact center and can reduce the cost required for time authentication by subjecting only the necessary range of call data at the contact center to time authentication. Here, the call data may include not only voice data in a call between a contact center operator and a customer, but also text data converted from this voice data using voice recognition.

[0012] However, the contact center is just one example, and the present invention can be similarly applied to other locations, such as when time authentication is performed on conversation data including voice data and text data of conversations that occur during sales or counter work, targeting sales representatives of products or services, store front desk personnel, etc. In the following description, the contact center operator will be described as performing tasks such as responding to inquiries with customers via voice calls, but this is not limited to this. The present invention can also be similarly applied to cases where tasks are performed via text chat (including those that can send and receive stamps, attachments, etc. in addition to text), video calls (including those that can share displays of electronic data), etc. (i.e., when the call data does not include voice data and only includes text data). More generally, the present invention can be similarly applied to cases where time authentication is performed on conversation data related to a conversation between multiple people.

[0013] Time authentication refers to the process whereby a time authentication authority of a trusted business (a business certified for time authentication services) assigns time information to the hash value (or message digest) of electronic data. The hash value to which time information has been assigned is called a time stamp token (or simply a time stamp), and this time stamp token proves that the electronic data existed at the time indicated by the time information and that the electronic data has not been tampered with since that time. Hereinafter, performing time authentication on certain electronic data will also be referred to as "setting time authentication on electronic data," etc.

[0014] <Overall structure> First, the overall configuration of a contact center system 1 according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of the overall configuration of the contact center system 1 according to this embodiment.

[0015] 1, the contact center system 1 according to this embodiment includes an authentication management device 10, an operator terminal 20, an administrator terminal 30, a PBX 40, a customer terminal 50, and a time-stamping authority 60. Here, the authentication management device 10, the operator terminal 20, the administrator terminal 30, and the PBX 40 are installed in a contact center environment E, which is a system environment of a 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 multiple geographically separated buildings.

[0016] The authentication management device 10 extracts a required range (hereinafter also referred to as a set range) from at least one of the voice data in a call between a customer and an operator and the text data converted from that voice using voice recognition, and then sets time authentication for this set range. That is, the authentication management device 10 calculates a hash value of the data in the set range, and then requests a time stamp token of that hash value from the TSA 60 via a communication network 80 including the Internet. The authentication management device 10 then stores the time stamp token returned from the TSA 60 together with the call data. This allows call data to be obtained in which time authentication has been performed only on the required range of the required data (voice data, text data, or both).

[0017] Here, whether to subject voice data, text data, or both to time authentication, and what range to set, are determined according to the content of the call represented by the call data. More specifically, they are determined according to the call reason of the call represented by the call data and the terms spoken during the call. The call reason is the reason why the customer called the call center (reason for receiving the call), and includes the content of the customer's inquiry, etc. For simplicity, the following explanation will be given assuming that one call has only one call reason.

[0018] The operator terminal 20 is a terminal of various types, such as a PC, used by an operator who responds to inquiries from customers, and functions as an IP (Internet Protocol) telephone.

[0019] The manager terminal 30 is a terminal of any type, such as a PC (personal computer), used by a manager who manages operators (such a manager is also called a supervisor).

[0020] The PBX 40 is a telephone exchange (IP-PBX) and is connected to a communication network 70 including a Voice over Internet Protocol (VoIP) network and a Public Switched Telephone Network (PSTN).

[0021] The customer terminal 50 is a variety of terminals used by customers, such as a smartphone, a mobile phone, or a landline phone.

[0022] The time-stamp authority 60 is a time-stamp authority operated or managed by a certified time-stamp service provider. In response to a request from the authentication management device 10, the time-stamp authority 60 replies with a time-stamp token to which time information has been added for the hash value included in the request.

[0023] The overall configuration of the contact center system 1 shown in FIG. 1 is merely an example, and other configurations may be used. For example, in the example shown in FIG. 1, the authentication management device 10 is included in the contact center environment E (i.e., the authentication management device 10 is an on-premise type), but all or part of the functions of the authentication management device 10 may be implemented by a cloud service or the like. Similarly, in the example shown in FIG. 1, the PBX 40 is an on-premise type telephone exchange, but it may be implemented by a cloud service. Furthermore, although the operator terminal 20 has been described as functioning as an IP telephone, for example, a telephone separate from the operator terminal 20 may be included in the contact center system 1. Furthermore, for example, a device (e.g., a gateway device) connecting the contact center environment E to an external network may function as the authentication management device 10.

[0024] <Time authentication settings> In this embodiment, time authentication is set for a required range (set range) of required data (hereinafter also referred to as target data). Here, the target data refers to the data among the voice data and text data included in the call data that is the target for setting time authentication. Also, the set range refers to the range within the temporal range of the target data that is the target for setting time authentication.

[0025] An example of the time authentication settings in this embodiment will now be described with reference to Fig. 2. Fig. 2 is a diagram for schematically explaining an example of the time authentication settings.

[0026] The example shown in the left diagram of Figure 2 is a case where a certain important term W1 is uttered at time T1 and a certain important term W2 is uttered at time T2 in a call regarding a call reason "product information." In this case, time authentication is set, for example, with voice data and text data as the target data and a time range from time T1-ΔT1 to time T1+ΔT1 (i.e., a range of ΔT1 before and after the time when word W1 was uttered). Also, time authentication is set, for example, with text data as the target data and a time range from time T2-ΔT2 to time T2+ΔT2 (i.e., a range of ΔT2 before and after the time when word W2 was uttered). This is because the content of the utterance before and after the utterance of an important term can be important matters, and it is appropriate to perform time authentication on the content of the utterance.

[0027] Important terms are words and phrases that are important to the call reason. For example, when responding to an inquiry about the call reason "product information," the operator needs to fully explain to the customer the price and warranty details of the product. Therefore, in this case, words such as "price" and "warranty" fall under important terms. Hereinafter, "important terms" will also be simply referred to as "important terms."

[0028] The example shown on the right side of Figure 2 is a case where a call was made regarding the call reason "contract conclusion." In this case, time authentication is set, for example, with voice data and text data as target data and the entire time range from the start to the end of the call as the set range. This is because, for example, in a call regarding "contract conclusion," all of the spoken content can be important matters, so it is appropriate to perform time authentication on the entire text data and voice data from start to end as target data.

[0029] In this way, in this embodiment, time authentication is set for a certain time range (including all) using either or both of voice data and text data as target data depending on the call reason and key terms uttered during the call. This makes it possible to set time authentication for an appropriate range of appropriate data depending on the importance of the call reason and key terms uttered during the call, and can reduce the cost required for time authentication compared to, for example, setting time authentication unconditionally for all call data.

[0030] As will be described later, the data to be subject to time authentication and the time range for which time authentication is to be set are set in advance using call reasons and key terms as conditions (rules).

[0031] <Functional configuration> Next, the functional configuration of the authentication management device 10 according to this embodiment will be described with reference to Fig. 3. Fig. 3 is a diagram showing an example of the functional configuration of the authentication management device 10 according to this embodiment.

[0032] 3, the authentication management device 10 according to this embodiment includes a speech recognition-to-text conversion unit 101, a call reason detection unit 102, a setting category identification unit 103, a term determination unit 104, and a time authentication setting unit 105. Each of these units is realized by, for example, processing executed by a processor such as a CPU (Central Processing Unit) by one or more programs installed in the authentication management device 10.

[0033] The authentication management device 10 according to this embodiment also includes a call data DB 106, a call reason DB 107, a setting condition DB 108, and a time authentication management DB 109. Each of these DBs (databases) is implemented, for example, by a storage device such as an auxiliary storage device. Note that at least one of these DBs may be implemented, for example, by a database server connected to the authentication management device 10 via a communication network.

[0034] The speech recognition and text conversion unit 101 converts the voice call between the operator terminal 20 and the customer terminal 50 into text by speech recognition. At this time, the speech recognition and text conversion unit 101 also performs speech recognition for each speaker and converts it into text. As a result, the operator's voice and the customer's voice are each converted into text.

[0035] This voice recognition is performed repeatedly while a voice call is being made between the operator terminal 20 and the customer terminal 50 (for example, it may be performed every few seconds, or it may be performed for each word or phrase spoken by the customer or operator).

[0036] Here, the voice data representing the voice call between the operator terminal 20 and the customer terminal 50 and the text data representing the text converted from the voice call by the voice recognition and text conversion unit 101 are stored as call data in the call data DB 106, for example, after the voice call is completed. That is, the call data DB 106 stores call data including voice data and text data for each call. The voice data includes voice data of the operator (hereinafter also referred to as operator voice data) and voice data of the customer (hereinafter also referred to as customer voice data). Similarly, the text data includes text data of the operator (hereinafter also referred to as operator text data) and text data of the customer (hereinafter also referred to as customer text data). The call data also includes information such as a call ID that identifies the call.

[0037] The call reason detection unit 102 refers to the call reasons DB 107 and detects the call reasons from the text converted by the speech recognition text conversion unit 101. Here, the call reasons DB 107 is a database that stores information for detecting the call reasons from the content of utterances by customers and operators. Details of the call reasons DB 107 will be described later.

[0038] The setting category identification unit 103 refers to the setting condition DB 108 and identifies the setting category from the call reasons detected by the call reason detection unit 102. The setting category is a category that indicates whether the setting range is the entire time range from the start to the end of the call or a partial time range. Hereinafter, the setting category will be either "all," which indicates that the setting range is the entire time range from the start to the end of the call, or "part," which indicates that the setting range is a partial time range of the entire time range. Here, the setting condition DB 108 is a database that stores information for determining target data and setting ranges according to call reasons and key terms. Details of the setting condition DB 108 will be described later.

[0039] When the setting category identification unit 103 identifies the setting category as "part," the term determination unit 104 refers to the setting condition DB 108 and determines whether or not important terms appear in the text converted by the speech recognition text conversion unit 101.

[0040] Time authentication setting unit 105 sets time authentication for all or part of the target data by referring to setting condition DB 108. That is, when setting category identification unit 103 identifies the setting category as "all," time authentication setting unit 105 refers to setting condition DB 108 and sets time authentication for all of the target data corresponding to the call reasons detected by call reason detection unit 102 as the setting range. On the other hand, when setting category identification unit 103 identifies the setting category as "part," time authentication setting unit 105 refers to setting condition DB 108 and sets time authentication for the target data and setting range that correspond to the call reasons detected by call reason detection unit 102 and key terms that appear in the text converted by speech recognition text conversion unit 101.

[0041] Specifically, the time authentication is set as follows. That is, the time authentication setting unit 105 calculates a hash value of the set range of the target data, and then requests a timestamp token of this hash value from the timestamp authority 60. Then, when the timestamp token is returned from the timestamp authority 60, the time authentication setting unit 105 creates time authentication information including a time authentication ID, a call ID, a call reason ID, the timestamp token, etc., and stores this time authentication information in the time authentication management DB 109. As a result, the time authentication information associated with the call data by the call ID is stored in the time authentication management DB 109, and time authentication is set for the set range of the target data, which is at least a part of the call data. The time authentication ID is an ID that identifies the time authentication information. Details of the time authentication management DB 109 will be described later.

[0042] <<Cole Reason DB107>> Next, details of the call reasons DB 107 will be described with reference to Fig. 4. Fig. 4 is a diagram showing an example of the call reasons DB 107.

[0043] As shown in FIG. 4, call reason DB 107 stores information (data or records) that associates call reason IDs, call reason names, and detection conditions.

[0044] The call reason ID is an ID that identifies the call reason. The call reason name is the name of the call reason. The detection condition is a condition that the call reason detection unit 102 uses to detect the call reason.

[0045] For example, the information in the first row of the call reason DB 107 shown in Fig. 4 is a call reason ID "a-1", a call reason name "product information", and a detection condition "both: insurance". Similarly, for example, the information in the second row of the call reason DB 107 shown in Fig. 4 is a call reason ID "a-2", a call reason name "contract procedure", and a detection condition "both: contract".

[0046] Here, in this embodiment, as an example, the detection condition is expressed in the format of "speaker: word". This means that the detection condition is met when the speaker utters the word. For example, the detection condition "both: insurance" means that the detection condition is met when both the operator and the customer utter the word "insurance". Similarly, for example, the detection condition "customer: address change" means that the detection condition is met when the customer utters the word "address change". Similarly, although not illustrated in FIG. 4, for example, the detection condition "operator: A" means that the detection condition is met when the operator utters the word "A".

[0047] However, the above format of the detection condition is just an example, and the detection condition may be expressed in various formats (for example, a format such as a logical expression).

[0048] <<Setting Condition DB108>> Next, the setting condition DB 108 will be described in detail with reference to Fig. 5. Fig. 5 is a diagram showing an example of the setting condition DB 108.

[0049] As shown in FIG. 5, the setting condition DB 108 stores information (data or records) that associates call reason IDs, setting categories, key terms, target data, and setting ranges.

[0050] The call reason ID is an ID that identifies the call reason. The setting category is a category that indicates whether the setting range is the entire time range from the start to the end of the call, or a partial time range. Key terms are words, phrases, etc. that are important to the call reason. The target data is the data that is the target of time authentication settings. The setting range is the time range that is the setting range for time authentication in the target data. Note that information with a setting category of "all" does not include key terms (or includes NULL, empty values, etc. as key terms).

[0051] For example, the information in the first row of the setting condition DB 108 shown in Fig. 5 is call reason ID "a-1", setting category "part", ((key term "fee", target data "text, voice", setting range "3 minutes before and after key term"), (key term "guarantee", target data "text", setting range "5 minutes before and after key term"), ...). This means that the setting category of the call data for call reason ID "a-1" is "part", and when the key term "fee" is uttered, the text data and voice data are the target data and the setting range for time authentication is 3 minutes before and after the utterance, and when the key term "guarantee" is uttered, the text data is the target data and the setting range for time authentication is 5 minutes before and after the utterance.

[0052] 5, the information in the second row of the setting condition DB 108 is, for example, call reason ID "a-2," setting category "all," target data "text, voice," and setting range "all period." This means that the setting category of the call data for call reason ID "a-2" is "all," and the target data is text data and voice data, and the setting range for time authentication is the entire period (i.e., the entire time range).

[0053] However, the above-mentioned set range is merely an example, and various time ranges can be defined. For example, it is possible to define a time range before or after the utterance of a key term, such as "3 minutes before the utterance of the key term" or "5 minutes after the utterance of the key term." Another possible definition is "5 minutes after 1 minute has passed since the utterance of the key term." In addition to time ranges, for example, a set range may be defined as a phrase or a sentence boundary containing a key term. Alternatively, a set range may be defined as a range beginning and ending at a conjunction that marks a turning point in the conversation, or a specific range with the conjunction as its reference point. Alternatively, a range of a series of conversations between an operator and a customer, such as for contract verification or address verification, may be extracted using natural language analysis, and a range representing a specific scene from the extracted range may be defined as the set range.

[0054] Furthermore, in this embodiment, a case will be described in which the target data is any one of "text data," "voice data," or "text data and voice data," but this is not limiting, and for example, it is also possible to consider whether the data was spoken by a customer or an operator. That is, the target data may be any one of "customer text data," "operator text data," "customer voice data," "operator voice data," "customer text data and operator text data," "customer text data and customer voice data," "customer text data and operator voice data," "operator text data and customer voice data," "operator text data and operator voice data," "customer voice data and operator voice data," "customer text data and operator text data and customer voice data," "customer text data and operator text data and operator voice data," "customer text data and operator voice data," "customer text data and operator voice data," "operator text data and customer ... voice data and operator voice data."

[0055] <Time authentication process flow> The following describes a case where time authentication is set for certain call data, with reference to Fig. 6, on the assumption that call data DB 106 stores voice data representing a voice call between operator terminal 20 and customer terminal 50, text data representing text converted from this voice call by voice recognition text conversion unit 101, and call data including the call ID of the voice call. Fig. 6 is a flowchart showing an example of the flow of the time authentication process.

[0056] Call reason detection unit 102 references call reasons DB 107 and detects a call reason from the text represented by the text data included in the call data (step S101). That is, call reason detection unit 102 references each record stored in call reasons DB 107 and determines whether or not there is a record whose text satisfies the detection condition. Then, call reason detection unit 102 identifies the call reason ID and call reason name of the record determined to satisfy the detection condition. As a result, the call reason for this call reason ID and call reason name has been detected.

[0057] Next, the setting category identifying unit 103 refers to the setting condition DB 108 and identifies the setting category from the call reason detected in the above step S101 (step S102). That is, the setting category identifying unit 103 refers to the record of the call reason ID detected in the above step S101 among the records stored in the setting condition DB 108 and identifies whether the setting category included in this record is "all" or "part."

[0058] If the setting category is determined to be "partial" in step S102, the term determination unit 104 refers to the setting condition DB 108 and determines whether or not a key term appears in the text (step S103). That is, the term determination unit 104 refers to the record of the call reason ID detected in step S101 among the records stored in the setting condition DB 108 and determines whether or not a key term included in this record appears in the text (step S103).

[0059] If it is determined in step S103 that a key term appears, time authentication setting unit 105 refers to setting condition DB 108 and identifies, for each key term appearing in the text, the target data and setting range corresponding to the key term (step S104). That is, time authentication setting unit 105 refers to the record of the call reason ID detected in step S101 among the records stored in setting condition DB 108 and identifies, for each key term appearing in the text, the target data and setting range corresponding to the key term.

[0060] For example, if the call reason ID detected in step S101 above is "a-1" and the key terms determined to appear in the text in step S103 above are "fee" and "guarantee," the time authentication setting unit 105 identifies the target data as "text, audio" and the setting range as "3 minutes before and after the key term," and the target data as "text" and the setting range as "5 minutes before and after the key term."

[0061] If it is determined in step S103 that a key term does not appear, the authentication management device 10 ends the time authentication process. In this case, time authentication is not set.

[0062] Next, the time-stamp setting unit 105 sets time authentication for the target data and set range corresponding to each key term that appears in the text (step S105). That is, for each key term that appears in the text, the time-stamp setting unit 105 extracts a set range from the target data corresponding to the key term, calculates a hash value, and then requests a time-stamp token of this hash value from the time-stamp authority 60. Then, when the time-stamp token for the key term is returned from the time-stamp authority 60, the time-stamp setting unit 105 creates a time-stamp ID and then creates time-stamp information that includes this time-stamp ID, the call ID of the call data, the call reason ID, the key term, the target data and set range corresponding to the key term, and the time-stamp token.

[0063] As a result, time authentication information is created for each key term that appears in the text. For example, if the key terms "fee" and "guarantee" appear in the text, time authentication information corresponding to "fee" and time authentication information corresponding to "guarantee" are created.

[0064] Note that if the same key term appears multiple times in a text, they are treated separately. For example, if the key term "fee" appears twice in the text, time authentication information corresponding to the first occurrence of "fee" and time authentication information corresponding to the second occurrence of "fee" are generated. However, if the same key term appears multiple times within a short period of time, it is not necessary to distinguish between them. For example, this is the case when the key term "fee" appears multiple times within a three-minute period. This is to avoid setting duplicate time authentication due to partial overlap of the setting ranges. Alternatively, for example, if multiple setting ranges at least partially overlap, or if the time interval between multiple setting ranges is equal to or less than a predetermined threshold even if they do not overlap (i.e., if the setting ranges are close to each other), the key terms corresponding to each of the multiple setting ranges may not be distinguished, or the multiple setting ranges may be combined into one.

[0065] If the setting category is identified as "all" in step S102, time authentication setting unit 105 refers to setting condition DB 108 and identifies the target data and setting range corresponding to the call reason detected in step S101 (step S106). That is, time authentication setting unit 105 refers to the record of the call reason ID detected in step S101 among the records stored in setting condition DB 108, and identifies the target data and setting range included in this record.

[0066] For example, if the call reason ID detected in step S101 above is "a-2," the time-of-day authentication setting unit 105 specifies the target data as "text, voice" and the setting range as "entire period."

[0067] However, since the setting category is specified as "all" in step S102 above, the setting range does not necessarily have to be specified. This is because the setting category "all" means that time authentication is set with the entire period of the target data as the setting range.

[0068] Next, the time-stamp setting unit 105 sets time authentication for the entire target data (step S107). That is, the time-stamp setting unit 105 calculates a hash value for the set range of the target data identified in step S106 above (i.e., the entire period of the target data), and then requests a time-stamp token of this hash value from the time-stamp authority 60. Then, when the time-stamp token is returned from the time-stamp authority 60, the time-stamp setting unit 105 creates a time-stamp ID and then creates time authentication information that includes this time-stamp ID, the call ID of the call data, the call reason ID, the target data and set range corresponding to the call reason ID, and the time-stamp token. In this way, time authentication information for the entire target data is created.

[0069] Following step S105 or step S107, time-authentication setting unit 105 stores the created time-authentication information in time-authentication management DB 109 (step S108). This means that time-authentication has been set for the set range of target data, which is at least a part of the call data (including the case where the set range is the entire period).

[0070] An example of time authentication information stored in the time authentication management DB 109 is shown in Fig. 7. The time authentication information in the first row of the time authentication management DB 109 shown in Fig. 7 is a time authentication ID "T001," a call ID "C001," a call reason ID "a-1," a key term "fee," target data "text, voice," a set range "three minutes before and after the key term," and a time stamp token "...". This means that the text data and voice data included in the call data of call ID "C001" are set as target data, and the time stamp token in the set range of three minutes before and after the utterance of the key term "fee" is set as time authentication.

[0071] 7 includes the time authentication information in the third row of the time authentication management DB 109 as follows: time authentication ID "T003," call ID "C002," call reason ID "a-2," key term "-," target data "text, voice," setting range "entire period," and time stamp token "...." This means that the text data and voice data included in the call data of call ID "C002" are set as target data, and the time stamp token for the entire period is set as time authentication.

[0072] As described above, the authentication management device 10 according to this embodiment sets time authentication for all or part of a time range of target data, with some or all of the data included in the call data of a voice call between an operator and a customer being targeted depending on the call reason of the voice call and the key terms spoken during the call. This makes it possible to set time authentication for an appropriate range of appropriate data depending on the importance of the call reason of the call and the importance of the terms spoken during the call. Therefore, for example, the cost required for time authentication can be reduced compared to when time authentication is unconditionally set for all call data.

[0073] In the present embodiment, as an example, a case where time authentication is performed on certain call data stored in the call data DB 106 has been described. However, the present invention is not limited to this. Time authentication may be performed in real time while a call is being made between an operator and a customer. In this case, for example, the start point of a set range may be identified from the exchange between the operator and the customer, and the time authentication information of step S105 or step S107 may be created at the end point of the set range, the end point of the call, or when a termination condition is satisfied by natural language processing. Furthermore, when creating the time authentication information, a hash value may be calculated that includes not only the data in the set range but also the current time. Here, examples of termination conditions based on natural language processing include the appearance of a break in a sentence or phrase, the appearance of a conjunction that marks a turning point in the conversation, etc.

[0074] <Modification> Below, several variations of this embodiment will be described.

[0075] <<Variation 1>> In this embodiment, the time range is set to the time before, after, around, and around the utterance of a key term as the starting point, but the time range may also be set to the time range performed by an operator. For example, if the call reason is "contract procedures," the operator may open an electronic file of the terms and conditions and read out the contents of the terms and conditions. Therefore, for example, the time range may be set to the time range before, after, around, and around the action of "opening the electronic file of the terms and conditions."

[0076] The above-mentioned action of "opening the electronic terms and conditions file" can be detected, for example, by monitoring the folder in which the electronic terms and conditions file is stored. Therefore, this modified example can be realized by, for example, including information indicating the access date and time of the electronic file in the call data. Other examples of such actions include scrolling the terms and conditions file to the end, or, in the case of a text chat, sending the terms and conditions file to the customer as an attachment or the customer opening the terms and conditions file. Furthermore, in the case of a video call, the operator sharing and displaying the terms and conditions file on the customer's screen or displaying it for a certain period of time can also be detected.

[0077] In addition, utterances and actions related to important matters may be collectively referred to as "actions."

[0078] <<Variation 2>> 6, the call reason is detected from the text represented by the text data, but the call reason may also be identified and detected from the number entered by the customer into the customer terminal 50 in response to guidance from an IVR (Interactive Voice Response) system. In this case, this modified example can be realized by including in the call data information indicating the guidance from the IVR system and information indicating the number entered by the customer into the customer terminal 50 in response to this guidance.

[0079] <<Variation 3>> In this embodiment, for simplicity, it is assumed that one call has only one call reason, but in general, one call may have multiple call reasons, such as when a customer calls the call center for two reasons: to change their address and to change their contract plan.

[0080] In such a case (i.e., when multiple call reasons are detected in step S101 of FIG. 6), the call data can be divided for each call reason, and step S102 and subsequent steps can be performed on each piece of divided call data. There are various possible methods for dividing the call data, but one possible method is to divide the call data at time T2-ΔT (where ΔT is a predetermined time interval) if call reason A is detected at time T1 and call reason B is detected at time T2. This is because it is likely that a conversation regarding the same call reason was taking place a little before the time a certain call reason was detected.

[0081] <<Variation 4>> In this embodiment, the target of time authentication is the set range of text data, voice data, or both included in the call data. However, data related to the set range may also be time-authenticated. For example, data related to the voice recognition function used to convert voice data within the set range into text data (e.g., a voice recognition model, etc.), software or programs included in the authentication management device 10 that performed some processing on data within the set range, and libraries used by such software or programs may also be time-authenticated. This makes it possible to verify, for example, whether appropriate processing was performed on data for which time authentication was set, and who is responsible for defects (liability for non-conformity) in the event of a malfunction or inconvenience.

[0082] The present invention is not limited to the above-described specifically disclosed embodiments, and various modifications, changes, and combinations with known technologies are possible without departing from the scope of the claims.

[0083] This application is based on basic application No. 2021-059678 filed in Japan on March 31, 2021, the entire contents of which are incorporated herein by reference. [Explanation of symbols]

[0084] 1. Contact Center System 10 Authentication management device 20 Operator terminal 30 Administrator terminal 40 PBX 50 customer terminals 60 Time Stamp Authority 70, 80 Communication Network 101 Speech recognition and text conversion unit 102 Call Reason Detection Unit 103 Setting category identification part 104 Terminology judgment section 105 Time authentication setting section 106 Call Data DB 107 Call Reason DB 108 Setting Condition DB 109 Time Authentication Management DB

Claims

1. an identification unit that identifies, as a setting range for time authentication, a portion of conversation data, which is voice data representing voices of a conversation between multiple people, and text data, which is obtained by converting the voices represented by the conversation data into text for each speaker using a voice recognition technology; a setting unit that sets time authentication for the specified setting range; and The identification unit When the text data includes a predetermined character string, the information processing device uses a predetermined time width for the character string, sets the time when the voice corresponding to the character string is spoken as a reference time, and identifies the range of the time width before and after the reference time as the set range.

2. The identification unit Detecting an inquiry from a first speaker to a second speaker based on the text data and a predetermined condition; If a character string corresponding to the detected inquiry content is included in the text data, a predetermined time width is used to identify, as the set range, a range of the time width before and after a time when a voice corresponding to the character string is uttered, The information processing apparatus according to claim 1 , wherein the condition is a condition indicating that a predetermined word uttered by a predetermined speaker is included in the text data.

3. The identification unit specifying a part of a time range of the target as the set range, for the conversation data or the text data, or for both the conversation data and the text data; The setting unit The information processing device according to claim 1 , wherein the time authentication is set for the set range of the target.

4. a step of identifying a part of conversation data, which is voice data representing voices of a conversation between multiple people, and text data, which is obtained by converting the voices represented by the conversation data into text for each speaker using a voice recognition technology, as a setting range for time authentication; a setting procedure for setting time authentication for the specified setting range; The computer executes The identification procedure includes: An information processing method in which, when the text data includes a predetermined character string, a predetermined time width is used for the character string, and the time when the voice corresponding to the character string is spoken is set as a reference time, and the range of the time width before and after the reference time is identified as the set range.

5. a step of identifying a part of conversation data, which is voice data representing voices of a conversation between multiple people, and text data, which is obtained by converting the voices represented by the conversation data into text for each speaker using a voice recognition technology, as a setting range for time authentication; a setting procedure for setting time authentication for the specified setting range; on the computer, The identification procedure includes: When the text data includes a predetermined character string, a predetermined time width is used for the character string, and the time when the voice corresponding to the character string is spoken is set as a reference time, and the range of the time width before and after the reference time is identified as the set range.

Images