Classification device, classification method, and classification program

Abstract

A classification device according to the present invention comprises: a first identification unit that, with speech data including the speech of a first speaker and the speech of a second speaker in a dialogue serving as input, uses a first identification model / rule so as to identify the speech type of each instance of speech included in the speech data; a second identification unit that, with the speech data and the speech type of each of the speeches serving as input, uses a second identification model / rule, which is predetermined according to the speech type, so as to identify in the speech data first identification speech indicative of a question and second identification speech indicative of a response to the first identification speech and that outputs speech pair data indicative of the first identification speech and the second identification speech; and a result classification unit that, with the output speech pair data serving as input, uses a result classification model / rule so as to classify the type of a response result with respect to the dialogue included in the speech data.

Description

[Technical Field]

[0001] The disclosed technology relates to a classification device, a classification method, and a classification program. [Background technology]

[0002] Dialogues in customer interactions have traditionally been recorded and analyzed. An example of a dialogue that is desirable to record and analyze is a dialogue between a customer and a customer service representative at a contact center. An example of a dialogue pattern for such a dialogue is a dialogue pattern in which a customer speaks in response to a question from the customer service representative. The question may be a question, a request, a demand, or the like. In this dialogue pattern, the customer's utterance is an answer (how the answer was given). Another example of a dialogue pattern is a dialogue pattern in which a customer service representative speaks in response to a customer question. In this dialogue pattern, the customer service representative's utterance is an explanation (how the explanation was given). The type of utterance that provides an answer or explanation as described above can be treated as an "explanatory answer." For this example dialogue pattern, the response result for the dialogue can be classified into types of utterances. If the response result indicates whether the customer was interested in the customer's question, the type of response result would be "interested / not interested." Furthermore, when a respondent answers a questionnaire, multiple types of response results are expected, such as "applicable," "not applicable," and "don't know."

[0003] For the above-mentioned dialogues, by classifying the types of response results in advance, it becomes possible to efficiently record and analyze the response results. One method for classifying the types of response results is a technology for estimating the type of response result (Non-Patent Document 1). In this technology, learning data is prepared in which the type of response result is assigned in advance as training data to the response result utterances, and a model is created and used to estimate the type of response result from the learning data. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] R.-E. Fan, K.-W. Chang, C.-J. Hsieh, X.-R. Wang, and C.-J. Lin. LIBLINEAR: A library for large linear classification Journal of Machine Learning Research 9(2008), 1871-1874. Summary of the Invention [Problem to be solved by the invention]

[0005] Even if the content of a second speaker's response to a question from a first speaker is the same, the meaning of the utterance may differ depending on the content of the question, resulting in different response results. For example, if the question is an affirmative question and a negative question, even if the utterance is "Yes, that's right" in both cases, the former will indicate a positive response and the latter a negative response. However, in conventional methods, if the response result types are divided into "affirmative responses" and "negative responses," and training data for "Yes, that's right" is used, one set is assigned "affirmative responses" and the other set is assigned "negative responses," the same utterance content will be assigned different response result types. This poses a problem of poor accuracy in estimating the response result types.

[0006] In addition, in the past, a pair of a question utterance and its corresponding response utterance was used as a unit of training data, but the question utterance of the first speaker is not necessarily immediately followed by an utterance that is a response to that utterance (i.e., an utterance that falls into the "explanatory response" utterance type in the above dialogue pattern). Therefore, a pair of a question utterance of the first speaker and the immediately following utterance of the second speaker may not constitute correct training data.

[0007] The disclosed technology has been made in consideration of the above points, and aims to provide a classification device, classification method, and classification program that can accurately classify the type of response result according to the content of the dialogue. [Means for solving the problem]

[0008] A first aspect of the present disclosure is a classification device including: a first identification unit that takes as input utterance data including an utterance from a first speaker and an utterance from a second speaker in a dialogue, and identifies the utterance type of each of the utterances included in the utterance data using a first identification model / rule for estimating an utterance type indicating a type of utterance in the dialogue; a second identification unit that takes as input the utterance data and the utterance type of each of the utterances, and identifies a first specified utterance indicating a question and a second specified utterance indicating a response to the first specified utterance in the utterance data using a second identification model / rule predetermined according to the utterance type, and outputs group data of utterances indicating the first specified utterance and the second specified utterance; and a result classification unit that takes as input the output group data of utterances, and classifies the type of response result for the dialogue included in the utterance data using a result classification model / rule for classifying the type of response result in the dialogue.

[0009] A second aspect of the present disclosure is a classification method in which a computer is caused to execute a process in which, as input, utterance data including utterances by a first speaker and an utterance by a second speaker in a dialogue is used to identify the utterance type of each of the utterances included in the utterance data using a first identification model / rule for estimating an utterance type indicating the type of utterance in the dialogue, the utterance data and the utterance type of each of the utterances are used as input to identify a first identification utterance indicating a question and a second identification utterance indicating a response to the first identification utterance in the utterance data using a second identification model / rule predetermined according to the utterance type, outputting a set of utterance data indicating the first identification utterance and the second identification utterance, and, as input, using a result classification model / rule for classifying types of response results in a dialogue, classifying the type of response result for the dialogue included in the utterance data.

[0010] A third aspect of the present disclosure is a classification program that causes a computer to execute the following process: input utterance data including utterances from a first speaker and a second speaker in a dialogue, identify the utterance type of each of the utterances included in the utterance data using a first identification model / rule for estimating an utterance type indicating the type of utterance in the dialogue, input the utterance data and the utterance type of each of the utterances, identify a first specific utterance indicating a question and a second specific utterance indicating a response to the first specific utterance in the utterance data using a second identification model / rule predetermined according to the utterance type, output group data of utterances indicating the first specific utterance and the second specific utterance, and input the output group data of utterances, classify the type of response result for the dialogue included in the utterance data using a result classification model / rule for classifying types of response results in a dialogue. [Effects of the Invention]

[0011] According to the disclosed technology, it is possible to accurately classify the types of response results according to the content of the dialogue. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 2 is a block diagram showing the hardware configuration of the classification device according to the present embodiment. [Figure 2] FIG. 1 is a block diagram showing a functional configuration of a classification device according to an embodiment of the present invention. [Figure 3] FIG. 2 is a diagram illustrating a configuration related to input and output of an input unit and a first specifying unit. [Figure 4] FIG. 10 is a diagram showing a configuration relating to input and output of a second identification unit and a result classification unit. [Figure 5] 10 is a flowchart showing the flow of classification processing by the classification device. [Figure 6] 10 is a flowchart showing the flow of a process for identifying a question utterance. [Figure 7] 10 is a flowchart showing the flow of a process for identifying a corresponding utterance. DETAILED DESCRIPTION OF THE INVENTION

[0013] An example of an embodiment of the disclosed technology will be described below with reference to the drawings. Note that the same reference numerals are used to designate identical or equivalent components and parts in each drawing. Also, the dimensional proportions in the drawings are exaggerated for the sake of explanation and may differ from the actual proportions.

[0014] The classification device according to the present embodiment provides certain improvements over conventional approaches to classifying types of interaction outcomes, and represents an advancement in the field of interaction outcome type classification.

[0015] First, the above-mentioned problems will be described using specific examples, and an overview of the technology of the present disclosure will be described.

[0016] In the following dialogue example, the first speaker is a customer (service user) and the second speaker is a customer (operator). However, the speakers are not limited to this, and other users and bots can also be considered as speakers.

[0017] First, we will show conversation examples 1 and 2 below, which show cases where the meaning of an utterance differs depending on the content of the question, even if the content of the utterance is the same.

[0018] [Dialogue example 1] First speaker: "There is a service to check the safety of people in the event of a disaster, but you're not interested, are you?" Second speaker: "Yes, that's right." [Dialogue example 2] First speaker: "There is a service to check the safety of people in the event of a disaster. Would you be interested in it?" Second speaker: "Yes, that's right."

[0019] The corresponding result for dialogue example 1 indicates a lack of interest, while the corresponding result for dialogue example 2 indicates an interest.

[0020] Next, dialogue example 3 and dialogue example 4 are shown below for cases where a question utterance by the first speaker is not immediately followed by an utterance of the "explanatory response" type in response to that utterance.

[0021] [Dialogue example 3] First speaker: "We have a service to check the safety of people in the event of a disaster. Would you be interested?" Second speaker: "What is a safety service?" First speaker: "It's a service that sends safety confirmation emails to employees and asks them to reply." Second Speaker: "Then it's unnecessary." [Dialogue example 4] First speaker: "We have a service to check the safety of people in the event of a disaster. Would you be interested?" Second speaker: "What is a safety service?" First speaker: "It's a service that sends safety confirmation emails to employees and asks them to reply." Second Speaker: "That's the one that gets sent automatically, right?"

[0022] In Dialogue Example 3, the second speaker first asks a question in response to the first speaker's question, and after receiving the answer, makes an utterance that serves as an "explanatory response" to the first question. In Dialogue Example 4, just like Dialogue Example 3, the second speaker first asks a question in response to the first speaker's question and receives the answer, but then continues to ask questions.

[0023] However, unlike Dialogue Example 3 and Dialogue Example 4, the first speaker's question is not necessarily immediately followed by an utterance that responds to that utterance. In Dialogue Example 3, the second speaker first asks a question in response to the first speaker's question, and after receiving the answer, an utterance is made in response to the first question. In Dialogue Example 4, as in Dialogue Example 3, the second speaker first asks a question in response to the first speaker's question, receives an answer to the question, but then continues to ask questions.

[0024] As described above, in the dialogue handled in this embodiment, it is assumed that a first speaker makes an utterance (question utterance) to ask a second speaker a question, and the second speaker makes an utterance (response utterance) in response to the question from the first speaker. In the technology disclosed herein, the question utterance is an example of a first specific utterance, and the response utterance is an example of a second specific utterance.

[0025] In the example of a dialogue between a customer and a customer service representative at a contact center, the first speaker is the customer service representative, and the second speaker is the customer. Therefore, the speaker's role in the dialogue determines whether to make a question or a response. Note that while the dialogue has two speakers, the first speaker and the second speaker, this example may have three or more speakers, as long as a question and a response are made in the dialogue.

[0026] Generally, depending on the content of a question utterance, a response utterance to the question utterance may be assigned a different response result even if the response utterance has the same content. In the prior art, it has been difficult to accurately classify the response results in such cases. It has also been difficult to identify the question utterance and the response result corresponding to the question utterance from the utterance content.

[0027] In this embodiment, first, an utterance type is identified, and then the identified utterance type is used to identify a set of utterance data (a question utterance and a response utterance to the question utterance). Then, in this embodiment, the type of response result is classified by using the identified set of utterance data. By using the utterance type, it is possible to identify a set of utterances even if there is no response utterance immediately after the question utterance. Furthermore, by using the set of utterances, it is possible to accurately classify the response result by the question utterance corresponding to the response utterance, even if the content of the response utterance is the same.

[0028] The configuration of this embodiment will be described below.

[0029] FIG. 1 is a block diagram showing the hardware configuration of a classification device 100. As shown in FIG.

[0030] 1, the classification device 100 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, and a communication interface (I / F) 17. Each component is connected to each other via a bus 19 so as to be able to communicate with each other.

[0031] The CPU 11 is a central processing unit that executes various programs and controls each component. That is, the CPU 11 reads programs from the ROM 12 or storage 14 and executes the programs using the RAM 13 as a work area. The CPU 11 controls the above components and performs various arithmetic processing in accordance with the programs stored in the ROM 12 or storage 14. In this embodiment, a classification program is stored in the ROM 12 or storage 14.

[0032] The ROM 12 stores various programs and various data. The RAM 13 temporarily stores programs or data as a working area. The storage 14 is configured with a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive) and stores various programs including the operating system and various data.

[0033] The input unit 15 includes a pointing device such as a mouse and a keyboard, and is used to perform various inputs.

[0034] The display unit 16 is, for example, a liquid crystal display, and displays various information. The display unit 16 may function as the input unit 15 by adopting a touch panel system.

[0035] The communication interface 17 is an interface for communicating with other devices such as terminals, etc. For this communication, for example, a wired communication standard such as Ethernet (registered trademark) or FDDI, or a wireless communication standard such as 4G, 5G, or Wi-Fi (registered trademark) is used.

[0036] Next, a description will be given of each functional configuration of the classification device 100. Fig. 2 is a block diagram showing the configuration of the classification device of this embodiment. Each functional configuration is realized by the CPU 11 reading out a classification program stored in the ROM 12 or storage 14, expanding it in the RAM 13, and executing it.

[0037] 2, the classification device 100 includes an input unit 110, a first identification unit 112, a second identification unit 114, a result classification unit 116, an output unit 118, an utterance data storage unit 120, an utterance type storage unit 122, a group data storage unit 124, and a model / rule storage unit 126. The model / rule storage unit 126 stores a first identified model / rule 130, a second identified model / rule 132, and a result classification model / rule 134. The models and rules will be described in detail later.

[0038] FIG. 3 shows a configuration relating to input and output of the input unit 110 and the first identification unit 112.

[0039] As shown in FIG. 3, the input unit 110 accepts input utterance data and stores the utterance data converted into utterance text in the utterance data storage unit 120. The input utterance data accepted by the input unit 110 is voice or character string data. The input utterance data includes, for example, an utterance by a first speaker and an utterance by a second speaker for each dialogue, such as dialogue examples 1 to 4 above. If the input utterance data is voice, the input unit 110 converts the input utterance data into text using speech recognition, and stores the utterance text as utterance data in the utterance data storage unit 120. Furthermore, the input unit 110 identifies and labels the first speaker and the second speaker for each utterance using speech recognition. If the input utterance data is a character string, the input unit 110 stores the character string representing the utterance as utterance text in the utterance data storage unit 120 as is as utterance data. In the processes using utterance data in the following sections, the utterance data is used as utterance text. Note that the utterance data is not limited to utterance text, and may be an audio signal.

[0040] 3, the first identification unit 112 receives utterance data from the utterance data storage unit 120 as input and outputs an utterance type for each utterance. The first identification unit 112 receives utterance data as input and uses the first identification model / rule 130 from the model / rule storage unit 126 to identify the utterance type for each utterance included in the utterance data, and stores the identified utterance type in the utterance type storage unit 122. The utterance type storage unit 122 stores each utterance in the utterance data and the utterance type identified for that utterance. In other words, the data stored in the utterance type storage unit 122 is utterance data whose utterance type has already been identified.

[0041] The first identification model / rule 130 is a model or rule for estimating / identifying an utterance type indicating the type of utterance in a dialogue. When using a model, for example, the model is created in advance by machine learning using training data in which utterances are assigned labels indicating the utterance type, and stored in the model / rule storage unit 126. Any multi-class classification method may be used as the machine learning method. Furthermore, a case where the first identification model / rule 130 is identified using a rule will be described. When identifying using a rule, for example, a list of endings indicating questions, such as "ka" (or "ka") or "yonegai" ("right"), is created. If the ending matches this list, the utterance type is determined to be "question." A list of expressions for asking needs is created. If the utterance type is initially determined to be "question" and contains an expression from the list, the utterance type is determined to be "needs hearing." A list of endings indicating requests or requests, such as "kudasai" (please) or "wo onegaishimasu" (please give me onegai) is created. If the ending matches this list, the utterance type is determined to be "request." A list of utterances for greetings and backchannels is created. If the ending matches this list, the utterance type is determined to be "greeting backchannel." If the answer is not one of the above, the utterance type is an "explanatory response."

[0042] Examples of utterance types are shown below. Labels representing utterance types include "needs hearing," "question," and "explanatory response." The label "other" may also be used when none of these apply. Examples of "other" include backchannels, greetings, requests, and demands. "Needs hearing" is a utterance type defined as an utterance inquiring about needs. "Question" is a utterance type defined as an utterance asking a question. "Explanatory response" is a utterance type defined as an utterance providing an explanation or answer. For example, when using a model, the first identification model / rule 130 is trained in advance to estimate three types of utterance types by machine learning from training data labeled with three types of labels. The first identification unit 112 uses the first identification model / rule 130 to identify which utterance type each input utterance belongs to: "needs hearing," "question," or "explanatory response." In the technology of the present disclosure, "needs hearing" is an example of a first utterance type, "question" is an example of a second utterance type, and "explanatory response" is an example of a third utterance type.

[0043] Examples of utterance types for the above dialogue examples 1 to 4 are given below. The input is the utterance of the utterance text included in the utterance data, and the output is the utterance type. Note that the speaker is labeled in the utterance. [Dialogue example 1] (Input) First speaker: Speech text "There is a service to check the safety of people in the event of a disaster, but you're not interested, are you?" / (Output) Speech type "Needs hearing" (Input) Second speaker: Speech text "Yes, that's right" / (Output) Speech type "Explanatory response" [Dialogue example 2] (Input) First speaker: Speech text "There is a service to check the safety of people in the event of a disaster. Would you be interested?" / (Output) Speech type "Needs hearing" (Input) Second speaker: Speech text "Yes, that's right" / (Output) Speech type "Explanatory response" [Dialogue example 3] (Input) First speaker: Speech text "There is a service to check the safety of people in the event of a disaster. Would you be interested?" / (Output) Speech type "Needs hearing" (Input) Second speaker: Speech text "What is the safety confirmation service?" / (Output) Speech type "Question" (Input) First speaker: Speech text "This is a service that sends safety confirmation emails to employees and asks them to reply to them." / (Output) Speech type "Explanatory response" (Input) Second speaker: Speech text "In that case, it's not necessary" / (Output) Speech type "Explanatory response" [Dialogue example 4] (Input) First speaker: Speech text "There is a service to check the safety of people in the event of a disaster. Would you be interested?" / (Output) Speech type "Needs hearing" (Input) Second speaker: Speech text "What is the safety confirmation service?" / (Output) Speech type "Question" (Input) First speaker: Speech text "This is a service that sends safety confirmation emails to employees and asks them to reply to them." / (Output) Speech type "Explanatory response" (Input) Second speaker: Speech text "That's the one that's automatically sent, right?" / (Output) Speech type "Question"

[0044] Note that if an utterance cannot be classified into any of the utterance types, the utterance type is set to "NULL" or "Other." Even if an "Other" label is prepared, it is assigned using the first identification model / rule 130. When a model is used for the first identification model / rule 130, the "Other" label is assigned if the estimation result of any label has low likelihood. When rules are used for the first identification model / rule 130, the "Other" label is assigned if none of the rules for determining each label applies. Furthermore, the number of utterance types is not limited to three, and the first identification model / rule 130 may be prepared in advance and used for three or more utterance types. For example, the label "Explanation Answer" includes either of the two cases, explanation and answer, but it may also be divided into four types, the label "Explanation" and the label "Answer."

[0045] FIG. 4 shows a configuration relating to input and output of the second identification unit 114 and the result classification unit 116.

[0046] 4, the second identification unit 114 receives as input the utterance data of each utterance and the utterance type stored in the utterance type storage unit 122, and outputs utterance group data. The utterance group data is data indicating a question utterance and a corresponding utterance to the question utterance. The second identification unit 114 receives as input the utterance type of each utterance, and identifies the question utterance and the corresponding utterance in the utterance data using the second identification model / rule 132 stored in the model / rule storage unit 126, and stores the utterance group data in the group data storage unit 124.

[0047] The second identification model / rule 132 is determined according to the utterance type, and is a model / rule for identifying a question utterance and a corresponding utterance, respectively. A case where a model is used will be described. A pair of an utterance of the utterance type "needs hearing" and an utterance of the utterance type "answer utterance" is assigned a label indicating whether or not they are a pair of corresponding utterances to create training data, and a model for estimating whether or not they are a pair of corresponding utterances is created. A pair of an utterance of the utterance type "needs hearing" and a subsequent utterance of the utterance type "answer utterance" is input, and if it is estimated to be a pair of corresponding utterances, the pair of utterances is stored as utterance pair data. An utterance of the utterance type "needs hearing" and a subsequent utterance of the utterance type "answer utterance" are considered to be an input utterance pair until the next utterance of the utterance type "needs hearing" is received, or until it is determined to be a pair of corresponding utterances, or until the call ends. The same applies to training data when training a model. Next, we will explain the case where rules are used. (1) The rule for identifying a question utterance is that the speaker is the first speaker and the utterance type is "needs hearing." Then, after identifying the question utterance, (2) the rule for identifying a corresponding utterance is applied. (2) The rule for identifying a corresponding utterance applies to cases where the dialogue consists of two utterances, such as dialogue examples 1 and 2 (2A), and to cases where the dialogue consists of four utterances, such as dialogue examples 3 and 4 (2B). For ease of explanation, in the explanation of the dialogue examples below, the question utterance and the corresponding utterance in the utterance pair data are distinguished from the utterance type by placing <> in parentheses.

[0048] Rule (2A) requires that the sequence of utterances in a dialogue satisfy the following conditions for two utterances: (2A-1) is the condition for the first utterance in the sequence of utterances, and (2A-2) is the condition for the second utterance in the sequence of utterances. (2A-1) The first speaker's utterance is a question. (2A-2) The type of speech of the second speaker is "explanatory response" In the above case, (2A-2) is identified as the corresponding utterance.

[0049] Rule (2B) requires that the sequence of utterances in a dialogue satisfy the following conditions for four utterances: (2B-1) is the condition for the first utterance in the sequence, (2B-2) is the condition for the second utterance in the sequence, (2B-3) is the condition for the third utterance in the sequence, and (2B-4) is the condition for the fourth utterance in the sequence. (2B-1) The first speaker's utterance is a question. (2B-2) The type of speech of the second speaker is a "question" (2B-3) The type of speech of the first speaker is "explanatory response" (2B-4) The type of speech of the second speaker is "explanatory response" In the above case, (2B-4) is identified as the corresponding utterance.

[0050] In this way, (2) the rules for identifying corresponding utterances define conditions for the combination of the speaker and utterance type of each utterance in the sequence of utterances for the dialogue included in the utterance data. The utterance types "question" and "explanatory answer" are used to identify corresponding utterances.

[0051] Furthermore, as a prerequisite for identifying the corresponding utterance, the question utterance of the first speaker must be identified first. Therefore, the second identification unit 114 identifies the question utterance, and performs a determination based on the condition (2) on the utterance following the question utterance to identify the corresponding utterance. The second identification unit 114 outputs the pair of the question utterance and the corresponding utterance identified in the utterance data as utterance pair data.

[0052] Examples of utterance group data for the above dialogue examples 1 to 4 are given below. The input is the utterance type, and the output is a question utterance and a response utterance, which are utterance group data. Note that the utterance text of the dialogue example is omitted for the sake of convenience. [Dialogue example 1] (Input) First speaker: Speech type "Needs hearing" / (Output) <Question utterance> (Input) Second speaker: Utterance type "Explanatory response" / (Output) <Corresponding utterance> In the case of Dialogue Example 1, the utterance text of the first speaker, "There is a safety confirmation service for when a disaster occurs, but you're not interested, are you?", is a <question utterance>, and the utterance type of the utterance text of the second speaker, "Yes, that's right," is an "explanatory response." In this case, the utterance text of the second speaker is a <response utterance>. [Dialogue example 2] (Input) First speaker: Speech type "Needs hearing" / (Output) <Question utterance> (Input) Second speaker: Utterance type "Explanatory response" / (Output) <Corresponding utterance> In the case of dialogue example 2, the utterance text of the first speaker, "There is a safety confirmation service for times of disaster. Are you interested in it?" is a <question utterance>, and the utterance type of the utterance text of the second speaker, "Yes, that's right," is an "explanatory response." In this case, the utterance text of the second speaker is a <response utterance>. [Dialogue example 3] (Input) First speaker: Speech type "Needs hearing" / (Output) <Question utterance> (Input) Second speaker: Utterance type "Question" (Input) First speaker: Speech type "Explanatory response" (Input) Second speaker: Utterance type "Explanatory response" / (Output) <Corresponding utterance> In dialogue example 3, the utterance text from the first speaker, "There is a safety confirmation service for times of disaster, would you be interested?" is a <question utterance>, followed by the utterance text from the second speaker, "What is a safety confirmation service?", whose utterance type is "question". Next, the utterance text from the first speaker, "It is a service that sends safety confirmation emails to employees and asks them to reply to those emails", whose utterance type is "explanatory answer". Next, the utterance text from the second speaker, "In that case, it's not necessary", whose utterance type is "explanatory answer". In this case, the utterance text from the last second speaker, "explanatory answer", is a <corresponding utterance>. [Dialogue example 4] (Input) First speaker: Speech type "Needs hearing" / (Output) <Question utterance> (Input) Second speaker: Utterance type "Question" (Input) First speaker: Speech type "Explanatory response" (Input) Second speaker: Utterance type "Question" / (Output) <null> Dialogue example 4 satisfies the conditions for a question utterance, but does not fit the rules for identifying a corresponding utterance, so there is no corresponding utterance. <null>This becomes:

[0053] As described above, when the fourth utterance is of the second speaker's utterance type "question," the range of rules for identifying the corresponding utterance may be expanded. In this case, for example, a condition is whether the fifth, sixth, and subsequent utterances by the second speaker are of the utterance type "explanatory response." If the condition is met, the last utterance is set as the <corresponding utterance> and becomes the <corresponding utterance> in the utterance pair data. The range of conditions can also be expanded in a similar manner when the utterance type of the fourth utterance is "NULL." By expanding the range of rules for identifying the corresponding utterance in this way, it is possible to identify the corresponding utterance and output utterance pair data even when the dialogue is not progressing as expected.

[0054] 4, the result classification unit 116 receives the utterance pair data from the pair data storage unit 124 and outputs a classification result of the type of response result in the dialogue. The result classification unit 116 receives the utterance pair data as input and uses the result classification model / rule 134 from the model / rule storage unit 126 to classify the type of response result for the dialogue included in the utterance data into "need present" or "no need present" indicating whether there is a need.

[0055] The result classification model / rule 134 is a model / rule for classifying the types of response results in a dialogue. When a model is used for the result classification model / rule 134, for example, the model is created in advance by machine learning using training data to which classification labels for the types of response results are assigned, and stored in the model / rule storage unit 126. The classification labels for the response results are defined as, for example, "needs" and "no needs." A model for classifying these two types of response results is trained in advance by machine learning from training data with these two types of labels. The result classification unit 116 uses the result classification model / rule 134 to classify the types of response results for utterance pair data into "needs" and "no needs." Furthermore, when rules are used for the result classification model / rule 134, a list of negative expressions such as "not" and "not" is created in advance. According to the rule, if a question utterance in the utterance pair data contains a negative expression from the negative expression list, it is considered "negative," and if not, it is considered "positive." The rule states that if the corresponding utterance in the paired utterance data contains a negative expression from the negative expression list, it is determined to be "negative", and if not, it is determined to be "positive". The presence or absence of a need is determined using the following negative / positive judgment combination rules for question utterances and corresponding utterances. Combinations are represented in brackets <>, and the rule for that combination is indicated by a right arrow →. First, let's say the combination is <question utterance, corresponding utterance> → judgement of need. <positive, positive> → judgement that there is a need. <positive, negative> → judgement that there is no need. <negative, positive> → judgement that there is no need. <negative, negative> → judgement that there is a need.

[0056] For dialogue examples 1 to 4, the types of response results are classified as follows: dialogue example 1 is "needs", dialogue example 2 is "no needs", dialogue example 3 is "no needs", and dialogue example 4 is "no classification". "No classification" means that there is no corresponding utterance in the utterance pair data, so it is not input to the result classification model / rule 134 and the type of response result is not classified.

[0057] Note that the classification of the types of response results by the result classification model / rules 134 is not limited to "need" and "no need." For example, the result classification model / rules 134 may classify the types of response results by dividing the labels into degrees of need. When there is a need, the labels are classified as "somewhat need," "need," or "very need," and when there is no need, the labels are classified as "somewhat no need," "no need," or "no need at all." In this way, the result classification unit 116 may use the result classification model / rules 134 to classify the types of response results into degrees of need.

[0058] The output unit 118 outputs the classification result of the type of the interaction response result obtained by the result classification unit 116.

[0059] Next, the operation of the classification device 100 will be described.

[0060] 5 is a flowchart showing the flow of the classification process by the classification device 100. The CPU 11 reads out the classification program from the ROM 12 or the storage 14, loads it into the RAM 13, and executes it, thereby performing the classification process.

[0061] In step S100, the CPU 11 functions as the input unit 110 to accept input utterance data, and stores the utterance data as utterance text in the utterance data storage unit 120.

[0062] In step S102, the CPU 11, functioning as the first identification unit 112, receives the utterance data as input, and uses the first identification model / rule 130 to identify the utterance type of each utterance included in the utterance data, and stores the identified utterance type in the utterance type storage unit 122.

[0063] In step S104, the CPU 11, as the second identification unit 114, uses the utterance type of each utterance as input and identifies the question utterance and the corresponding utterance in the utterance data using the second identification model / rule 132, and stores them in the group data storage unit 124 as utterance group data.

[0064] In step S106, the CPU 11, as the result classification unit 116, receives the utterance set data as input and classifies the type of response result for the dialogue included in the utterance data using the result classification model / rules 134. For example, if "need exists" and "no need exists" are defined as the types of response results, the result is classified as "need exists" or "no need exists" indicating whether or not there is a need. Note that in this step, if there is no corresponding utterance in the utterance set data, the classification result is output as "no classification."

[0065] In step S108, the CPU 11 functions as the output unit 118 to output the classification result regarding the type of the response result of the dialogue.

[0066] Next, the process of step S104 will be described with reference to the flowcharts of Figures 6 and 7. In the process of step S104, a process of identifying a question utterance and a process of identifying a response utterance are executed.

[0067] The flow of the process for identifying a question utterance is shown in Fig. 6. In this process, each utterance with an utterance type in the utterance type storage unit 122 is treated as a processing target, and processing is performed for each utterance.

[0068] In step S200, the CPU 11 receives, from the utterance type storage unit 122, an utterance to be processed and the utterance type of the utterance as input.

[0069] In step S202, CPU 11 determines whether the speaker of the utterance to be processed is the first speaker. If it is determined that the speaker is the first speaker, the process proceeds to step S204, and if it is determined that the speaker is not the first speaker (i.e., if it is determined that the speaker is the second speaker), the process ends.

[0070] In step S204, CPU 11 determines whether or not a question utterance has been identified using the utterance type of the utterance to be processed. If it is determined that the utterance has been identified, the process proceeds to step S206; if it is determined that the utterance has not been identified, the process ends. A condition for determining whether an utterance is a question utterance is, for example, that the utterance type of the utterance is "needs hearing." In the process performed by second identification unit 114, as described above, a question utterance (first identified utterance) is identified according to the utterance type of the first speaker's utterance, and in the following steps, a corresponding utterance (second identified utterance) is identified based on the utterance type of the second speaker's utterance after the question utterance.

[0071] In step S206, the CPU 11 identifies the utterance to be processed as a question utterance of the first speaker in the utterance set data.

[0072] In step S208, the CPU 11 executes a process of identifying a response utterance using the identified question utterance of the first speaker.

[0073] FIG. 7 shows the flow of the process of identifying a corresponding utterance. The process of identifying a corresponding utterance is executed as a subroutine of step S208. The inputs for the process of identifying a dialogue are the identified question utterance of the first speaker and the utterance type of the utterance corresponding to the question utterance. For utterances corresponding to question utterances, it is sufficient to determine up to which utterance number to use in the rules for identifying the corresponding utterance.

[0074] In step S300, the CPU 11 determines whether the sequence of utterances in the dialogue satisfies the above condition (2A) for two utterances. If it is determined that the condition (2A) is satisfied, the process proceeds to step S304, and if it is determined that the condition (2A) is not satisfied, the process proceeds to step S302.

[0075] In step S302, the CPU 11 determines whether the sequence of utterances in the dialogue satisfies the above condition (2B) for the four utterances. If it is determined that the condition (2B) is satisfied, the process proceeds to step S304, and if it is determined that the condition (2B) is not satisfied, the process proceeds to step S306.

[0076] In step S304, the CPU 11 identifies the utterance determined to satisfy the condition in step S300 or S302 as the corresponding utterance of the second speaker in the utterance set data. The utterance determined to satisfy the condition is the last utterance in the utterance sequence among the utterances corresponding to the question utterance.

[0077] In step S306, the CPU 11 outputs that there is no corresponding utterance (NULL) in the utterance set data.

[0078] As described above, the classification device 100 of this embodiment can accurately classify the type of response result according to the content of the utterance in the dialogue.

[0079] The classification process executed by the CPU after reading the software (program) in the above embodiment may be executed by various processors other than the CPU. Examples of such processors include dedicated electrical circuits, such as programmable logic devices (PLDs) (such as field-programmable gate arrays (FPGAs)) whose circuit configuration can be changed after manufacture, graphics processing units (GPUs), and application-specific integrated circuits (ASICs) that are processors with circuit configurations specifically designed to execute specific processes. The classification process may be executed by one of these various processors, or by a combination of two or more processors of the same or different types (e.g., multiple FPGAs, or a combination of a CPU and an FPGA). The hardware structure of these various processors is, more specifically, an electrical circuit that combines circuit elements such as semiconductor devices.

[0080] In the above embodiment, the classification program is pre-stored (installed) in the storage 14, but the present invention is not limited to this. The program may be provided in a form stored in a non-transitory storage medium such as a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versatile Disk Read Only Memory), or a USB (Universal Serial Bus) memory. The program may also be downloaded from an external device via a network.

[0081] The following additional notes are provided regarding the above-described embodiments.

[0082] (Additional note 1) Memory and at least one processor coupled to said memory; Including, The processor: receiving, as input, utterance data including utterances from a first speaker and a second speaker in a dialogue, identifying the utterance type of each of the utterances included in the utterance data using a first identification model / rule for estimating an utterance type indicating a type of utterance in the dialogue; using utterance data of each of the utterances and the utterance type as input, and using a second identification model / rule predetermined according to the utterance type, identifying a first specific utterance indicating a question and a second specific utterance indicating a response to the first specific utterance in the utterance data, and outputting utterance pair data indicating the first specific utterance and the second specific utterance; Using the output utterance set data as input, classifying the type of response result for the dialogue included in the utterance data using a result classification model / rule for classifying the type of response result in the dialogue; The classification device is configured as follows.

[0083] (Additional note 2) A non-transitory storage medium storing a program executable by a computer to perform a classification process, receiving, as input, utterance data including utterances from a first speaker and a second speaker in a dialogue, identifying the utterance type of each of the utterances included in the utterance data using a first identification model / rule for estimating an utterance type indicating a type of utterance in the dialogue; using utterance data of each of the utterances and the utterance type as input, and using a second identification model / rule predetermined according to the utterance type, identifying a first specific utterance indicating a question and a second specific utterance indicating a response to the first specific utterance in the utterance data, and outputting utterance pair data indicating the first specific utterance and the second specific utterance; Using the output utterance set data as input, classifying the type of response result for the dialogue included in the utterance data using a result classification model / rule for classifying the type of response result in the dialogue; Non-transitory storage medium. [Explanation of symbols]

[0084] 100 Classifier 110 Input section 112 1st Specific Part 114 Second Specific Part 116 Result classification section 118 Output section 120 Speech data storage unit 122 Speech type memory unit 124 sets of data storage 126 Model / Rule Memory 130 First Specific Model / Rules 132 Second Specific Model / Rules 134 Outcome Classification Models / Rules< / null> < / null>

Claims

1. a first identification unit that receives, as input, utterance data including utterances of a first speaker and an utterance of a second speaker in a dialogue, and identifies the utterance type of each of the utterances included in the utterance data using a first identification model / rule for estimating an utterance type indicating a type of the utterance in the dialogue; a second identification unit that receives utterance data of each of the utterances and the utterance type as input, identifies a first specific utterance indicating a question and a second specific utterance indicating a response to the question of the first specific utterance in the utterance data by using a second identification model / rule predetermined according to the utterance type, and outputs pair data of utterances indicating the first specific utterance and the second specific utterance; a result classification unit that uses the output utterance set data as input and classifies the type of response result for the dialogue included in the utterance data using a result classification model / rule for classifying the type of response result in the dialogue; the second identification unit determines that the second specific utterance does not exist if the second specific model / rule does not apply; The result classification unit sets the type of the response result to "no classification" when the second specific utterance is not present.

2. The classification device of claim 1, wherein the second identification unit identifies the first identified utterance as a question utterance according to the type of utterance of the first speaker, and identifies the second identified utterance based on the type of utterance of the second speaker following the identified first identified utterance, even if the second identified utterance does not occur immediately after the first identified utterance and there is another utterance in between.

3. a model in the first specific model / rule is trained to output estimation results of a first utterance type indicating a need inquiry, a second utterance type indicating a question, and a third utterance type indicating an explanation or an answer, as the utterance types; 3. The classification device according to claim 1, wherein the first identification unit inputs the utterance data to the first identification model / rule, and identifies, for each of the utterances, the first utterance type, the second utterance type, or the third utterance type based on an estimation result output by the first identification model / rule.

4. The rules in the second specific model / rule include: a rule for identifying the first specific utterance is that the speaker of the utterance is the first speaker, and the utterance of the first speaker is the first utterance type; The classification device described in claim 3, wherein the rule for identifying the second specific utterance defines conditions for the speaker of each utterance in a sequence of utterances and for the combination of the second utterance type and the third utterance type for the dialogue included in the utterance data.

5. The model in the result classification model / rule is trained to classify the presence or absence of a need for the type of response result, The classification device according to any one of claims 1 to 4, wherein the result classification unit inputs the utterance group data into the result classification model / rule and classifies the type of the response result for the dialogue included in the utterance data into whether or not there is a need.

6. a model in the outcome classification model / rules trained to classify the degree of need; The classification device according to claim 5 , wherein the result classification unit inputs the utterance set data into the result classification model / rule and classifies the types of the response results for the dialogue into degrees of needs.

7. On the computer, receiving, as input, utterance data including utterances of a first speaker and an utterance of a second speaker in a dialogue, and identifying the utterance type of each of the utterances included in the utterance data using a first identification model / rule for estimating an utterance type indicating a type of utterance in the dialogue; using utterance data of each of the utterances and the utterance type as input, and using a second specification model / rule predetermined according to the utterance type, specifying a first specification utterance indicating a question and a second specification utterance indicating a response to the question of the first specification utterance in the utterance data, and outputting utterance pair data indicating the first specification utterance and the second specification utterance; Using the output utterance set data as input, classifying the type of response result for the dialogue included in the utterance data using a result classification model / rule for classifying the type of response result in the dialogue; When the second specific utterance is identified and does not fit the second specific model / rule, it is determined that the second specific utterance does not exist; In the classification of the type of the response result, if the second specific utterance is not present, the type of the response result is set to “no classification.” The classification method that causes the processing to occur.

8. On the computer, receiving, as input, utterance data including utterances of a first speaker and an utterance of a second speaker in a dialogue, and identifying the utterance type of each of the utterances included in the utterance data using a first identification model / rule for estimating an utterance type indicating a type of utterance in the dialogue; using utterance data of each of the utterances and the utterance type as input, and using a second specification model / rule predetermined according to the utterance type, specifying a first specification utterance indicating a question and a second specification utterance indicating a response to the question of the first specification utterance in the utterance data, and outputting utterance pair data indicating the first specification utterance and the second specification utterance; Using the output utterance set data as input, classifying the type of response result for the dialogue included in the utterance data using a result classification model / rule for classifying the type of response result in the dialogue; When the second specific utterance is identified and does not fit the second specific model / rule, it is determined that the second specific utterance does not exist; In the classification of the type of the response result, if the second specific utterance is not present, the type of the response result is set to “no classification.” A classification program that performs the processing.

Images