Dialogue system and center device

Abstract

To provide a dialogue system that allows dialogue with a certain quality guaranteed while reducing the human burden.SOLUTION: A center device includes: receiving means for receiving a user message from a terminal device; extraction means for extracting one or more characteristic words included in the user message; response candidate generation means for generating one or more response candidates, which are response message candidates for the user message; word display means for displaying one or more selectable words from the characteristic words included in the user message so that the operator can select them; response candidate display means for selectably displaying the one or more response candidates based on the selected selectable word when the selectable word is selected by the operator from the one or more selectable words; and response transmitting means, when a response candidate is selected by the operator from the one or more response candidates, for transmitting the selected response candidate as a response message to the terminal device.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to an interactive system and a center device that use terminal devices having a voice input / output function. [Background technology]

[0002] Loneliness is a factor that leads to serious health risks such as dementia, and the increase in single-person households (especially elderly people living alone) coupled with the recent increased risk of infectious diseases makes loneliness more likely to occur, making alleviating loneliness an important social issue. Casual conversation and other everyday interactions are thought to be effective in alleviating loneliness.

[0003] In recent years, communication services have been proposed that aim to alleviate feelings of loneliness by installing devices with voice input and output functions (e.g., interactive robots) in the homes of elderly people, especially those living alone, and engaging in everyday conversations such as casual voice chat through these devices.

[0004] In this system, when a user speaks to a device at home, the content is converted into a string of characters using voice recognition technology and sent as a message to the management center of the service provider. At the management center, an operator who notices the receipt of the message types a reply into the center device, which is then sent back to the user. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2021-157419 [Patent Document 2] Japanese Patent Application Laid-Open No. 2007-286376 Summary of the Invention [Problem to be solved by the invention]

[0006] There are currently dialogue services involving humans and automated dialogue services using communication robots. However, the labor shortage caused by the declining labor force is also a social issue, and there are limits to how much human operators can provide. On the other hand, communication robots that do not require human intervention may be put off by users due to the low quality of the dialogue (inappropriate responses, standard responses, etc.).

[0007] Patent Document 1 proposes a UI that supports the operator's responses by displaying candidate responses for each category (affirmative, counterargument, topic change) along with an evaluation value. However, when an operator responds to a message from a user, it is a heavy burden for the operator to think of a response message for each message, and there is a risk that the quality of the response messages will vary depending on the operator. Although Patent Document 1 displays candidate responses to the operator, the operator must check each candidate and decide on a response message, which still places a burden on the operator.

[0008] Patent Document 2 proposes a technology that switches between automatic responses and responses via an operator by requesting a remote support device for response assistance when a robot cannot decide on a response message. However, in conversations where it is not easy for a robot to determine the appropriateness of a response message, such as everyday conversations such as casual conversations, switching to an operator response each time may place an excessive burden on the operator.

[0009] In view of the above circumstances, an object of the present invention is to provide a dialogue system that can reduce the burden on humans while ensuring a certain level of quality in dialogue. [Means for solving the problem]

[0010] A dialogue system according to one aspect of the present invention comprises: A terminal device; a center device capable of communicating with the terminal device; A dialogue system comprising: The terminal device an input means for inputting a user message that is a message from a user; a message sending means for sending the input user message; a response output means for receiving and outputting a response message to the user message; and The center device a receiving means for receiving the user message from the terminal device; an extraction means for extracting one or more characteristic phrases included in the user message; a reply candidate generating means for generating one or more reply candidates which are candidates for the reply message in response to the user message; a phrase display means for displaying one or more selectable phrases from the characteristic phrases included in the user message so that the operator can select them; a response candidate display means for displaying, when a selectable phrase is selected from the one or more selectable phrases by the operator, one or more response candidates based on the selected selectable phrase in a selectable manner; a response sending means for, when a candidate answer is selected by the operator from the one or more candidate answers, sending the selected candidate answer to the terminal device as the response message; It has.

[0011] According to this embodiment, the center device displays appropriate selectable phrases to the operator based on feature phrases extracted from the user message, and displays and allows the operator to select from reply candidates generated based on the selectable phrases selected by the operator. This minimizes the burden on the operator of selection operations while displaying appropriate reply candidates that match the user message and outputting a reply message selected from the reply candidates to the user, thereby ensuring the quality of the dialogue.

[0012] The phrase display means may display the one or more selectable phrases on the user message in a manner that makes them distinguishable from phrases other than the one or more selectable phrases and that makes them selectable.

[0013] Distinguishable manner means, for example, bolding, different colors, boxed text, etc. By prominently displaying the selectable words, the operator can easily select the selectable words.

[0014] The phrase display means may determine the selectable phrases based on individual characteristic phrases set for each user from among the one or more characteristic phrases.

[0015] This allows selectable phrases to be set for each user, allowing for natural conversation that is tailored to the user.

[0016] The phrase display means may highlight an individual characteristic phrase set for each user from among the one or more selectable phrases.

[0017] By prominently displaying the individual characteristic phrases, the operator can easily select the individual characteristic phrases that are suitable for each user.

[0018] The individual characteristic phrases for each user may be set based on a dialogue history including the user message and a response message.

[0019] This allows selectable phrases to be set for each user, allowing for natural conversation that is tailored to the user.

[0020] The phrase display means may display some of the one or more characteristic phrases as the one or more selectable phrases based on the number of times the one or more characteristic phrases appear and / or their position within the user message.

[0021] This allows the operator to naturally select a response candidate that matches the user's wishes.

[0022] The phrase display means may highlight some of the one or more selectable phrases based on the number of times the one or more characteristic phrases appear and / or their positions in the user message.

[0023] This allows words with higher importance (contents that the user wants to talk about) to stand out among the selectable words.

[0024] the phrase display means performs display based on the number of times the one or more characteristic phrases appear, The number of times the one or more characteristic phrases appeared is the number of times it appears in said user message; and / or It may also be the number of times the message appears in a dialogue history including the user message and the response message.

[0025] This allows for natural conversation that is tailored to the user.

[0026] the answer candidate display means displays the one or more answer candidates in a selectable and editable manner; When the selected candidate reply is edited, the response sending means may send the edited candidate reply as the response message to the terminal device.

[0027] This allows for editing of candidate responses individually for each user, enabling a more user-friendly dialogue.

[0028] the answer candidate generation means generates the one or more answer candidates based on the one or more feature expressions extracted by the extraction means; When a selectable phrase is selected from the one or more selectable phrases displayed selectably by the phrase display means, The answer candidate display means may selectably display one or more answer candidates based on the selectable phrase selected from the one or more answer candidates generated by the answer candidate generation means.

[0029] In this way, if response candidates are generated before a selectable phrase is selected, the response candidates can be displayed in a selectable manner in a short time after the selectable phrase is selected, thereby making the dialogue with the user smoother in terms of time.

[0030] When a selectable phrase is selected from the one or more selectable phrases displayed selectably by the phrase display means, the answer candidate generation means generates the one or more answer candidates based on the selected selectable phrase; The answer candidate display means may selectably display the one or more answer candidates generated by the answer candidate generation means based on the selectable phrases.

[0031] This means that only response candidates based on the selected selectable phrase are generated, and there is no need to generate response candidates that are not displayed on the operator device, thereby reducing the processing load and processing time on the center device.

[0032] A center device according to one aspect of the present invention comprises: a receiving means for receiving a user message from a terminal device that exchanges messages with a user in a two-way manner; an extraction means for extracting one or more characteristic phrases included in the user message; a reply candidate generating means for generating one or more reply candidates which are candidates for a reply message to the user message; a phrase display means for displaying one or more selectable phrases from the characteristic phrases included in the user message so that the operator can select them; a response candidate display means for displaying, when a selectable phrase is selected from the one or more selectable phrases by the operator, one or more response candidates based on the selected selectable phrase in a selectable manner; a response sending means for, when an operator selects a candidate answer from the one or more candidate answers, sending the selected candidate answer to the terminal device as the response message; It is equipped with: [Effects of the Invention]

[0033] According to the present invention, it is possible to provide a dialogue system that can conduct dialogue with a certain level of quality guaranteed while reducing the burden on humans.

[0034] The effects described here are not necessarily limited to those described above, and may be any of the effects described in the present invention. [Brief explanation of the drawings]

[0035] [Figure 1] 1 is a diagram showing a configuration of a dialogue system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram showing a hardware configuration of a center device of the dialogue system. [Figure 3] The operation flow of the dialogue system is shown below. [Figure 4] 10 shows an example of a display mode of selectable phrases by the phrase display means. [Figure 5] 10 shows an example of a display mode of selectable phrases by the phrase display means. DETAILED DESCRIPTION OF THE INVENTION

[0036] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0037] 1. Dialogue system configuration

[0038] FIG. 1 is a diagram showing the configuration of a dialogue system according to this embodiment.

[0039] The dialogue system 10 realizes a communication service in which everyday conversations such as casual chatting are mainly conducted. The dialogue system 10 is difficult to standardize responses, anticipates a wide variety of responses, and realizes everyday conversations such as continuous casual chatting with multiple users. Unlike automatic response systems for product guidance, contract procedures, etc., which can be implemented without the human intervention of an operator, the dialogue system 10 requires the human intervention of an operator to ensure the quality of the dialogue, as emotions such as empathy and intentions are important.

[0040] The dialogue system 10 is composed of a center device 100 as a server device managed by a business operator that provides services through the dialogue system 10, an operator device 160 connected to the center device 100, and multiple terminal devices 200 connected to a network so as to be able to communicate with the center device 100.

[0041] The terminal device 200 is installed in the home of a user (e.g., an elderly person) of this system. The terminal device 200 has an input means 201, a message sending means 202, and a response output means 203. The input means 201 inputs a user message in the form of voice data, which is a message from the user, via a microphone M. The message sending means 202 sends the input user message to the center device 100 via the network. The response output means 203 receives a response message to the user message from the center device 100 via the network, and outputs it as voice from a speaker S to the user.

[0042] The terminal device 200 may have at least the above components, but may also be an interactive robot with a small doll-like appearance that makes the user feel like they are having a conversation with a person (especially everyday conversation such as casual chatting).

[0043] The center device 100 is managed by a business operator that provides services using the dialogue system 10, and includes at least a receiving means 101, an extracting means 102, a reply candidate generating means 103, a determining means 104, a phrase displaying means 105, a reply candidate displaying means 106, and a reply sending means 107. It also includes a control means for controlling these means. These means are realized by well-known hardware (such as a server computer or a personal computer) or software stored in a storage means 140, as appropriate.

[0044] The storage means 140 stores a dictionary 141 and a dialogue history 142. The dictionary 141 stores categories (places, foods, hobbies, preferences, etc.) and polarity values ​​(semantic flags such as positive or negative) in association with words. The dictionary 141 may be set for each user, may be user-independent, or may be both. The dialogue history 142 may be data accumulating past dialogues for each user, or may be user-independent data accumulating a combination of dialogue histories of multiple users (multiple terminal devices 200), or may be both.

[0045] The operator device 160 is managed by a business that provides services through the dialogue system 10, and displays, on a web browser to the operator, candidate responses that are response messages to user messages received by the center device 100. The center device 100 and the operator device 160 are in a web server-web client relationship.

[0046] The center device 100 and the operator device 160 may be integrated into one hardware unit.

[0047] 2. Hardware configuration of the center device

[0048] FIG. 2 is a diagram showing the hardware configuration of the center device 100. As shown in FIG.

[0049] As shown in the figure, the center device 100 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, an input / output interface 15, and a bus 14 connecting these together.

[0050] The CPU 11 accesses the RAM 13 and other memory as needed, and performs various arithmetic processing while controlling all of the blocks in the center device 100. The ROM 12 is a non-volatile memory that permanently stores firmware such as the OS, programs, and various evaluation indexes to be executed by the CPU 11. The RAM 13 is used as a working area for the CPU 11, and temporarily stores the OS, various applications currently being executed, and various data currently being processed.

[0051] The input / output interface 15 is connected to a display unit 16, an operation reception unit 17, a storage unit 18, a communication unit 19, and the like.

[0052] The display unit 16 is a display device that uses, for example, an LCD (Liquid Crystal Display), an OLED (Organic ElectroLuminescence Display), a CRT (Cathode Ray Tube), or the like.

[0053] The operation reception unit 17 is, for example, a pointing device such as a mouse, a keyboard, a touch panel, or other input device.

[0054] The storage unit 18 is a non-volatile memory such as a hard disk drive (HDD), a flash memory (solid state drive (SSD)), or other solid-state memory. In addition to the OS, the storage unit 18 stores schedule information, various flags, evaluation indicators, and software for implementing each means of the center device 100. The storage unit 18 also stores applications and other programs and databases for exchanging messages with the terminal device 200, including generating a response message.

[0055] The communication unit 19 is, for example, a NIC (Network Interface Card) for Ethernet (registered trademark) or one of various modules for wireless communication such as a wireless LAN, and is responsible for communication processing with the terminal device 200.

[0056] The hardware configuration of the operator device 160 and the terminal device 200 is basically the same as that of the center device 100, but the terminal device 200 has a microphone M and a speaker S on the front surface as described above.

[0057] 3. Operational flow of the dialogue system 10

[0058] Figure 3 shows the operation flow of the dialogue system.

[0059] The input means 201 of the terminal device 200 inputs a user message, which is a message from the user, through the microphone M (step S1, Yes). The user message is voice data. The message transmission means 202 of the terminal device 200 transmits the input user message to the center device 100 via the network. The message transmission means 202 may perform voice recognition on the user message as voice data, convert it into a user message as text data, and transmit it to the center device 100, or may transmit the user message as voice data to the center device 100.

[0060] The receiving means 101 of the center device 100 receives the user message from the terminal device 200 via the communication unit 19 (step S2). The receiving means 101 may receive the user message in the form of voice data and convert it into a user message in the form of text data, or may receive the user message in the form of text data converted from voice data by the terminal device 200.

[0061] The extraction means 102 of the center device 100 performs morphological analysis on the user message, which is text data, to divide it into a plurality of words and phrases, and extracts one or more characteristic words and phrases contained in the user message from the plurality of words and phrases (step S3). For example, the extraction means 102 registers a dictionary 141 in which a category is assigned to each word and phrase, and extracts words in a predetermined category (place, food, hobby, preference, etc.) as characteristic words and phrases. The extraction means 102 may also extract different characteristic words and phrases for each user. Specifically, words (words) related to the hobbies and preferences of each user are registered in the dictionary 141 as individual characteristic words and phrases for each user, and characteristic words and phrases are extracted based on the individual characteristic words and phrases related to the user with whom the user is interacting. In extracting characteristic words and phrases, only individual characteristic words and phrases may be extracted, or individual characteristic words and phrases may be preferentially extracted. Furthermore, for example, words related to a user's hobbies and preferences may be registered in advance in the dictionary 141 as characteristic words and phrases, or dialogues with users may be accumulated as dialogue history 142, and the hobbies and preferences of each user may be identified from the dialogue and registered as individual characteristic words and phrases. Furthermore, the user's hobbies and preferences may be learned from each user's past user messages using AI or machine learning, and words from the user messages that match the hobbies and preferences may be registered as individual characteristic words and phrases. Note that in this embodiment, "phrase" includes a single word such as a noun, and a phrase containing multiple words.

[0062] The reply candidate generation means 103 of the center device 100 generates one or more reply candidates that are candidates for reply messages that empathize with the user message, based on one or more characteristic phrases. The reply candidate generation means 103 may generate reply candidates by using AI or machine learning to learn various user messages and appropriate reply messages for them. Furthermore, the reply candidate generation means 103 generates rule-based standard reply candidates (e.g., "Welcome home") for pre-registered phrases such as greetings (e.g., "I'm home").

[0063] As a first example, the extraction means 102 of the center device 100 performs morphological analysis on the user message "I'm back. When I went to the park, beautiful camellias were blooming." to extract one or more characteristic phrases "park" and "camellia" (step S3). The answer candidate generation means 103 generates an answer candidate "Welcome back. Beautiful camellias were blooming. That's wonderful." based on the characteristic phrase "camellia," and further generates another answer candidate based on the characteristic phrase "park" (step S4).

[0064] As a second example, the extraction means 102 of the center device 100 performs morphological analysis on the user message "I'm back. Today's day service was fun." to extract the characteristic phrases "day service" and "fun" (step S3). The answer candidate generation means 103 generates an answer candidate "Welcome back. You had fun at the day service. That's great." based on the characteristic phrases "day service" and "fun" (step S4).

[0065] As a third example, the extraction means 102 of the center device 100 performs morphological analysis on the user message "I'm not feeling well today" to extract the characteristic phrases "health" and "bad" (step S3). The answer candidate generation means 103 generates an answer candidate "You're not feeling well. That's unfortunate" based on the characteristic phrases "health" and "bad" (step S4).

[0066] The determination means 104 of the center device 100 calculates a response evaluation value, which is an evaluation value that evaluates the appropriateness of the response candidate, and determines whether to respond to the user message automatically or manually based on the response evaluation value (step S5). The determination method used by the determination means 104 will be described in detail later.

[0067] (1) Automatic response

[0068] If it is determined that an automatic response should be made, the response sending means 107 of the center device 100 sends one of the one or more response candidates generated by the response candidate generating means 103 (step S4) as a response message to the terminal device 200 (step S9). That is, when an automatic response is made, the center device 100 does not need to display a user message or response candidates on the operator device 160. The response sending means 107 may convert the response message from text data into a response message from voice data and send it to the terminal device 200, or may send the response message from text data to the terminal device 200.

[0069] The response output means 203 of the terminal device 200 receives the response message from the center device 100 and outputs the response message from the speaker S at a predetermined timing (for example, immediately after receiving the response message, when a human presence sensor detects a person around the terminal device 200). The response output means 203 may receive and output a response message in the form of voice data from the center device 100, or may receive a response message in the form of text data from the center device 100 and convert it into a response message in voice data using voice synthesis technology and output the response message.

[0070] (2) Manual response

[0071] If it is determined that a manual response is to be made, the phrase display means 105 of the center device 100 displays the user message on the operator device 160 used by the operator, and also displays one or more selectable phrases in a selectable manner. The one or more selectable phrases are at least a part of the one or more characteristic phrases extracted from the user message (step S3). The phrase display means 105 may display all of the one or more characteristic phrases as one or more selectable phrases.

[0072] The phrase display means 105 may display some of the one or more characteristic phrases as one or more selectable phrases based on the number of times the one or more characteristic phrases appear and / or their positions in the user message. For example, the phrase display means 105 may display, as selectable phrases, only characteristic phrases that appear more frequently or that appear later in the user message. The number of times the one or more characteristic phrases appear may be the number of times they appear in the current user message, or the number of times they appear in the dialogue history 142 including past user messages and response messages.

[0073] Furthermore, the phrase display means 105 may preferentially determine, among one or more feature phrases, individual feature phrases related to the user with whom the user is interacting as selectable phrases. For example, only individual feature phrases may be displayed as selectable phrases, or individual feature phrases may be preferentially displayed as selectable phrases.

[0074] In a first example, the phrase display means 105 may display all of one or more characteristic phrases "park" and "camellia" as selectable phrases. "Displaying selectably" means that the center device 100, which is a web server, displays them with hyperlinks on the web browser of the operator device 160, which is a web client, so that they can be selected by manual operation by the operator (clicking, hovering the mouse over, tapping, etc.).

[0075] 4 and 5 show examples of display modes of selectable phrases by the phrase display means.

[0076] As shown in FIG. 4, the phrase display means 105 may display one or more selectable phrases 111 within the user message 112 in a manner that allows them to be distinguished from phrases other than the one or more selectable phrases. As shown in FIG. 5, the phrase display means 105 may display one or more selectable phrases 111 together outside the user message 112 in a manner that allows them to be distinguished from phrases other than the one or more selectable phrases. A distinguishable manner means, for example, bolding, a different color, underlining, boxed text, etc. As shown in FIG. 5, displaying the selectable phrases 111 together outside the user message 112 with a heading 115 that reads "Keywords" is also one type of distinguishable manner. By prominently displaying the selectable phrases 111, the operator can easily select the selectable phrases 111. The phrase display means 105 further displays a cursor 113 for selecting the selectable phrases 111.

[0077] The phrase display means 105 may highlight some of the one or more selectable phrases. Highlighting may, for example, be a process of further changing the color of a selectable phrase when it is displayed in bold, or further bolding a selectable phrase when it is displayed in a color different from the other phrases, or thickening or changing the color of a boxed frame when a selectable phrase is displayed in boxed text, but is not limited to these. This makes more important words (what the user wants to talk about) stand out among the selectable phrases, allowing the operator to naturally select a response candidate that suits the user's wishes.

[0078] For example, some of the one or more selectable phrases may be highlighted based on the number of times that one or more characteristic phrases appear and / or their position in the user message. Specifically, the phrase display means 105 may highlight selectable phrases that appear more frequently or that appear later in the user message, among the one or more selectable phrases, to make it easier for the operator to select them compared to other selectable phrases. The number of times that one or more characteristic phrases appear may be the number of times they appear in the current user message, or the number of times they appear in the dialogue history 142 including past user messages and response messages. In this way, by setting selectable phrases for each user, it becomes possible to naturally have a dialogue that is tailored to the user.

[0079] Furthermore, the phrase display means 105 may highlight individual feature phrases related to the user with whom the conversation is being held, among one or more selectable phrases. For example, only the individual feature phrases may be highlighted, or at least the individual feature phrases may be highlighted. In this way, by setting selectable phrases to be highlighted for each user, a conversation that is tailored to the user can be held naturally.

[0080] The operator refers to the user message and one or more selectable phrases displayed on the operator device 160, and selects one selectable phrase by operating (clicking, hovering the mouse, tapping, etc.). This allows the operator to intuitively decide on the content of the response that is in line with their intention.

[0081] The answer candidate display means 106 of the center device 100 determines one selectable phrase selected by the operator via the operator device 160 from one or more selectable phrases in the user message (step S6, Yes). The answer candidate display means 106 displays one or more answer candidates generated by the answer candidate generation means 103 based on the selected selectable phrase (step S4) on the operator device 160 in a selectable and editable manner (step S7). If there are multiple answer candidates generated based on the selected selectable phrase, the answer candidate display means 106 may display the multiple answer candidates in descending order of response evaluation value (described below). This allows the operator to select an answer candidate that suits their intentions.

[0082] In the first example, the answer candidate display means 106 determines one selectable phrase "camellia" selected from one or more selectable phrases "park" and "camellia" displayed on the operator device 160. The answer candidate display means 106 displays the answer candidate 114 "Welcome home. Beautiful camellias are blooming. That's wonderful." (see FIGS. 4 and 5) generated by the answer candidate generation means 103 based on the selected selectable phrase "camellia" (step S4) on the operator device 160 in a selectable and editable manner (step S7).

[0083] The operator uses the cursor 113 to select one of the one or more candidate responses displayed on the operator device 160. This allows the operator to edit the candidate responses as necessary and select the edited candidate response. This allows the operator to intuitively determine candidate responses, and furthermore, by editing candidate responses individually for each user, a more user-friendly dialogue can be naturally conducted.

[0084] The response sending means 107 of the center device 100 determines one candidate response selected via the operator device 160 from one or more candidate responses displayed on the operator device 160 (step S8). The response sending means 107 sends the selected candidate response to the terminal device 200 as a response message (step S9). When the selected candidate response is edited, the response sending means 107 sends the edited candidate response to the terminal device 200 as a response message. The response sending means 107 may convert the text data response message into a voice data response message using voice synthesis technology and send it to the terminal device 200, or may send the text data response message to the terminal device 200.

[0085] The response output means 203 of the terminal device 200 receives the response message from the center device 100 and outputs the response message from the speaker S at a predetermined timing (for example, immediately after receiving the response message, when a human presence sensor detects a person around the terminal device 200). The response output means 203 may receive and output a response message in the form of voice data from the center device 100, or may receive a response message in the form of text data from the center device 100 and convert it into a response message in voice data using voice synthesis technology and output the response message.

[0086] 4. Judgment means

[0087] The determination means 104 of the center device 100 determines whether to respond to the user message automatically or manually based on the evaluation of the response candidates (step S5). The evaluation of the response candidates involves evaluating the appropriateness of the response message to the user message, taking into consideration factors such as whether the response is considerate to the user and whether it will not cause discomfort. A specific example of the determination method used by the determination means 104 is described below.

[0088] (1) Example 1

[0089] For example, the evaluation of the reply candidate may be evaluated as a binary value, positive or negative, and the determination means 104 determines to provide an automatic response if the evaluation of the reply candidate is positive, and determines to provide a manual response if the evaluation of the reply candidate is negative. In other words, a positive response is expected to be a response that is considerate of the user. For this reason, the determination means 104 determines to provide an automatic response to facilitate smooth dialogue with the user and reduce the burden on the operator. On the other hand, a negative response may cause discomfort to the user. For this reason, the determination means 104 determines not to provide an automatic response but to provide a response through an operation by an operator to determine a response message, i.e., to provide a manual response.

[0090] For example, the determination means 104 extracts evaluative words from answer candidates. "Evaluative words" are words and individual characteristic words associated with polarity values ​​(semantic flags such as positive or negative) and categories (places, foods, hobbies, preferences, etc.) in the dictionary 141. The evaluation of an answer candidate may be determined to be negative or positive based on the polarity value that appears more frequently among evaluative words with positive polarity values ​​and evaluative words with negative polarity values ​​included in the answer candidate. The determination means 104 may determine the evaluation of an answer candidate to be negative or positive based on the polarity value of the evaluative word that appears last among multiple evaluative words included in the answer candidate. If the answer candidate includes an evaluative word with a negative polarity value, the determination means 104 may determine the evaluation of the answer candidate to be negative regardless of the number or position of positive evaluative words.

[0091] The determination means 104 may also determine whether the evaluation of a candidate answer is negative or positive based on the category of the evaluative words included in the candidate answer. For example, if the candidate answer contains words from a specific category (places, food, hobbies, etc.), it is determined to be positive. Furthermore, based on the individual characteristic words related to the user with whom the candidate answer is being spoken, the evaluation means 104 may determine whether the candidate answer is positive if it contains evaluative words related to the preferences of the user with whom the candidate answer is spoken, and determine whether the candidate answer is negative if it contains evaluative words related to things the user with whom the candidate answer is not good at.

[0092] (2) Example 2

[0093] Alternatively, the determination unit 104 may calculate a response evaluation value indicating the appropriateness of a response message to a user message for each candidate reply, and determine whether to automatically or manually respond to the user message based on the response evaluation value. The response evaluation value may be calculated to be higher for a candidate reply that is more likely to empathize with the user message. For example, the determination unit 104 may calculate a higher response evaluation value for a candidate reply that includes the user's hobbies and preferences, or includes a word or phrase with a positive polarity value, an emotional word or phrase, or a word or phrase from a specific category.

[0094] For example, the determination unit 104 extracts evaluative words from the response candidates. The determination unit 104 may determine the response evaluation value based on the polarity value (positive or negative), category (place, food, hobby or preference, individual characteristic word or phrase, etc.) and / or position (first half, second half, etc.) of the evaluative words in the user message.

[0095] Specifically, the determination means 104 calculates a response evaluation value based on a positive / negative index value obtained by determining whether the response candidate is positive or negative from the number and positions of positive or negative evaluative phrases in the response candidate. For example, the determination means 104 uses 0 as a reference point, adds or subtracts 1 point for each evaluative phrase having a positive or negative polarity value, and calculates the total score as the response evaluation value of the response candidate. Note that the determination means 104 may determine that the response candidate is positive if the total score is equal to or greater than a predetermined value (e.g., a positive value), or that the response candidate is negative if the total score is less than the predetermined value (e.g., a negative value).

[0096] (3) Example 3

[0097] The determination means 104 may weight (e.g., convert to 1.5 points) words in the latter half of a reply candidate (e.g., evaluative words used in the last part of a reply candidate, evaluative words used in the last sentence of a reply candidate). The determination means 104 may add an additional point (e.g., +1 point) when a positive evaluative word is a word in a category related to the user's hobbies and preferences. When a reply candidate is evaluated as including a negative evaluative word, the determination means 104 may set the response evaluation value of the reply candidate to the lowest point, regardless of whether or not the reply candidate includes a positive evaluative word.

[0098] The determination means 104 may determine to automatically respond if the calculated response evaluation value is equal to or greater than a threshold. If the response evaluation value is equal to or greater than the threshold, it is assumed that the degree of empathy for the user is high, and therefore an automatic response can be made to reduce the human load on the operator.

[0099] When the answer candidate generation means 103 generates multiple answer candidates, the determination means 104 may calculate a response evaluation value for each of the multiple answer candidates and determine to automatically respond if there is one answer candidate whose response evaluation value differs from the other response evaluation values ​​by a predetermined value or more. For example, if the answer candidate with the highest response evaluation value is greater than the evaluation values ​​of all other answer candidates by a predetermined value or more, the determination means 104 automatically responds with the highest response candidate as a response message. This allows the automatic response of an appropriate answer candidate, reducing the human workload of the operator.

[0100] The determination means 104 may determine whether to automatically or manually respond to a user message based on the evaluation of the user message in addition to the evaluation of the reply candidates. That is, the determination means 104 may evaluate the user message in the same manner as the evaluation of reply candidates, taking into account the polarity value and category of the user message, the position of the evaluative phrase in the user message, and the like. For example, if the user message contains negative content, responding with positive content may cause discomfort to the user, so an automatic response may not be performed, and the response may be performed after confirmation by an operator.

[0101] Specifically, the determination means 104 may further add or subtract points from the response evaluation value of the reply candidate, taking into account the positive and negative relationships between the user message and the reply candidate. For example, when the user message is positive and the reply candidate is positive, it is assumed that the degree of empathy is high, so the response evaluation value is added. When the user message is negative and the reply candidate is positive, no addition is made. When the user message is positive and the reply candidate is negative, it is likely to cause discomfort to the user, so confirmation by an operator is required and points are subtracted. When the user message is negative and the reply candidate is negative, it is assumed that the degree of empathy is not low, so no deduction is made.

[0102] In a second example, the determination means 104 determines that the evaluation of the user message is positive because the evaluative phrase "it was fun" contained in the user message "I'm back. I had fun at the day service today." has a positive polarity value. The determination means 104 determines that the response evaluation of the response candidate is positive because the evaluative phrases "it was fun" and "good" contained in the response candidate "Welcome back. I heard you had fun at the day service today. That's great." have positive polarity values. The determination means 104 may determine that the response candidate "Welcome back. I heard you had fun at the day service today. That's great" is to be the automatic response based on a combination of the evaluation of the user message (positive) and the evaluation of the response candidate (positive).

[0103] In a third example, the determination means 104 determines that the evaluation of the user message is negative because the evaluative phrase "bad" contained in the user message "I'm not feeling well today." has a negative polarity value. The determination means 104 determines that the evaluation of the reply candidate is negative because the evaluative phrases "bad" and "unfortunate" contained in the reply candidate "You're not feeling well, huh? That's unfortunate." have negative polarity values. The determination means 104 may determine that the reply candidate "You're not feeling well, huh? That's unfortunate" should be a manual response (i.e., a response after confirmation by an operator) based on a combination of the evaluation of the user message (negative) and the evaluation of the reply candidate (negative).

[0104] (4) Example 4

[0105] The determination means 104 may calculate the response evaluation value by taking into account not only the polarity value included in the evaluation phrase of the response candidate but also other evaluation indicators (a) to (c). The determination means 104 may add points when each evaluation indicator is satisfied. The determination means 104 may weight and add points to some of the evaluation indicators (for example, (a) the item of the response candidate's sensitivity index value is given 2 points and the other items are given 1 point). The determination means 104 may calculate each evaluation indicator as a numerical value between 0 and 1, for example, and take a weighted average. The determination means 104 may also use a deductive method instead of adding points to the evaluation value.

[0106] The determination means 104 may calculate the response evaluation value by adding or subtracting points based on the above-mentioned separate evaluation index: (a) the sensitive index value of the response candidate, (b) whether or not the candidate appears in the dialogue history 142 including the user message and the response message, and / or (c) the number of times the candidate appears in the dialogue history 142 for a predetermined period. The determination means 104 may determine that an automatic response should be made if the total score is a positive value, that the response evaluation value is positive, or that a manual response should be made if the total score is a negative value, or that an automatic response should be made if the response evaluation value is equal to or greater than a threshold value.

[0107] (a) When the sensitivity index value, which indicates whether the content of a candidate reply is sensitive and may cause discomfort to the user, indicates a non-sensitive value, the determination means 104 may add points to the response evaluation value of the candidate reply, making it more likely to be determined as an automatic response. For example, if the candidate reply is composed of a standard phrase (such as a greeting) or a phrase that is likely to co-occur (be used together) with a standard phrase, the candidate reply is not sensitive and therefore may be added points. This determination may be made by estimating the degree of co-occurrence from a dictionary in which co-occurring phrases are registered or from the dialogue history.

[0108] (b) When a candidate reply has a history of appearance in the dialogue history 142, the determination means 104 may add points to the response evaluation value of the candidate reply, making it more likely that the candidate reply is determined to be an automatic response. For example, the determination means 104 may compare the generated candidate reply with the user-specific or user-independent dialogue history 142, and if the candidate reply has been used as a response message, determine that the candidate reply has a history of appearance in the dialogue history 142 and add points to the candidate reply. This is because similar dialogues often generate similar response messages, and if a response has been used in the past, it is assumed that the risk of responding is low.

[0109] (c) When a phrase or expression used in a reply candidate appears infrequently in the dialogue history 142 over a predetermined period, the determination unit 104 may increase the response evaluation value of the reply candidate to a value that makes it more likely to be determined as an automatic response. That is, the determination unit 104 may compare the dialogue history 142 with the user over a certain period (e.g., one day or one week) with the generated reply candidate to determine whether equivalent phrases or expressions are used. If equivalent phrases or expressions are used, the determination unit 104 may deduct points, and if not, may increase points. The determination unit 104 may increase the response evaluation value of a reply candidate when it does not include an expression used in a dialogue with the user over the most recent predetermined period, making it more likely to be determined as an automatic response. This is because, while a reply candidate may include a convenient phrase such as "That's good" or "That's fun," repeated use of the same expression can become formulaic and undesirable. For this reason, the determination unit 104 may increase points for reply candidates that use different expressions and deduct points for reply candidates that use expressions used in dialogues over the most recent predetermined period.

[0110] 5. Conclusion

[0111] The dialogue system 10 provides a communication service in which everyday conversations such as casual chatting are primarily conducted. The dialogue system 10 is difficult to standardize responses to, anticipates a wide variety of responses, and realizes everyday conversations such as continuous casual chatting with multiple users. Unlike automated response systems for product information, contract procedures, etc., which can be implemented without the intervention of an operator, the dialogue system 10 requires human operators to ensure the quality of the dialogue, as emotions such as empathy and intentions are important. Meanwhile, labor shortages have become a social issue, and there are limits to what an operator can do manually.

[0112] Therefore, first, according to this embodiment, the center device 100 generates reply candidates (step S4) based on feature words and phrases extracted from the user message (step S3), and determines whether to automatically or manually respond to the user message based on the reply evaluation value, which is an evaluation value of the generated reply candidates (step S5). This increases the number of opportunities for automatic responses and reduces the operator's workload compared to, for example, a case in which a simple greeting is automatically responded to and other responses are manually responded to. Furthermore, because reply candidates with positive response evaluation values ​​are automatically used as response messages, the quality of the automatic responses can be ensured while reducing the operator's workload. Furthermore, in the case of manual responses, the operator does not need to manually generate a response message; instead, the operator simply selects from the displayed reply candidates after editing them as necessary. This ensures the quality of the automatically generated response messages for manual responses while reducing the operator's workload.

[0113] Second, according to this embodiment, even when a manual response is made via an operator, appropriate selectable phrases are displayed to the operator based on feature phrases extracted from the user message (step S3), and reply candidates generated (step S4) based on the selectable phrases selected by the operator (step S6) are displayed to the operator (step S7) and the operator is allowed to select one (step S8). This minimizes the burden of the selection operation on the operator while displaying appropriate reply candidates that match the user message and outputting a reply message selected from the reply candidates to the user, thereby ensuring the quality of the dialogue.

[0114] 6. Variations

[0115] In this embodiment, the first process is to determine whether to provide an automatic or manual response. The second process is to display selectable phrases in a selectable manner in the case of a manual response, and display selectable response candidates based on the selected selectable phrase in a selectable manner. Both the first process (determining whether to provide an automatic or manual response) (step S5) and the second process (displaying selectable phrases in a selectable manner) (steps S6 to S8) contribute to reducing the human workload of the operator and ensuring the quality of the response.

[0116] Therefore, as a modified example, omitting the first process (determining whether to provide an automatic or manual response) and simply performing the second process (displaying selectable phrases in a selectable manner) contributes to reducing the operator's workload and ensuring the quality of responses. In other words, the determination of whether to provide an automatic or manual response (step S5) may be omitted, and a manual response may be provided to almost all user messages except for standard phrases such as greetings. In this case, the manual response involves displaying selectable phrases in a selectable manner and selecting a response message from candidate responses based on the selected selectable phrase (steps S6 to S8). Even with this method, the operator's only processing involves selecting a selectable phrase and selecting a candidate response, thereby contributing to reducing the operator's workload and ensuring the quality of responses.

[0117] In addition, when the determination of whether to use the first automatic response or manual response (step S5) is omitted, the center device 100 may extract feature words from the user message (step S3), and after the operator selects selectable words that are at least part of the feature words (step S6, Yes), generate answer candidates based on the selected selectable words (step S4), and display the generated answer candidates in a selectable manner (step S7). In other words, the order may be reversed from that of the above embodiment. This results in generating only answer candidates based on the selected selectable words, and there is no need to generate answer candidates that are not displayed on the operator device 160, thereby reducing the processing load and processing time of the center device 100. On the other hand, if answer candidates are generated (step S4) before the operator selects a selectable word (step S6, Yes), as in the above embodiment, answer candidates can be displayed in a selectable manner (step S7) in a short time after the operator selects a selectable word (step S6, Yes), thereby facilitating a smooth dialogue with the user.

[0118] As another variation, omitting the second process (displaying selectable phrases in a selectable manner) and simply performing the first process (determining whether to use an automatic or manual response) can contribute to reducing the operator's workload and ensuring the quality of the response. That is, the display of response candidates based on selectable phrases selected by the operator (steps S6 to S8) can be omitted, and response candidates generated based on feature phrases can be displayed. In this case, in the case of a manual response, the operator does not generate a response message or select selectable phrases; instead, the operator edits the displayed response candidates as necessary and then selects one. This reduces the operator's workload while ensuring the quality of the response message based on the automatically generated response candidates. Furthermore, in the case of an automatic response, the automatic response uses response candidates with positive response evaluation values ​​as the response message, thereby ensuring the quality of the automatic response while reducing the operator's workload.

[0119] In addition, in the present embodiment, the determination of whether to provide an automatic or manual response is based on a response evaluation value, and the determination of whether to provide an automatic or manual response is based on a response evaluation value and a user evaluation value have been described. However, the determination of whether to provide an automatic or manual response may also be based on a user evaluation. The method of evaluating user messages is the same as that described in the embodiments, and therefore will not be described here. If the evaluation of the user message is positive or has a high evaluation value (e.g., above a threshold), it is assumed that the user message is about a fun, everyday event. In this case, an automatic response is determined to be appropriate because there is a low risk of the user feeling uncomfortable if the user responds with content consistent with the user message. On the other hand, if the evaluation of the user message is negative or has a low evaluation value, it is assumed that the user message is about a sensitive topic, such as a worry or health problem. In this case, a manual response is determined to be appropriate because there is a risk that an automatic response will cause discomfort to the user. This ensures the quality of the automatic response while reducing the human workload on the operator.

[0120] In the present embodiment, the description has been given of a system in which some feature words and phrases (for example, individual feature words and phrases) are preferentially extracted as feature words and phrases, displayed as selectable words and phrases, or highlighted. As a variation, information on recent topics such as news and disaster information is obtained from news sites on the Internet, SNS (Social Networking Service), etc., and registered as current affairs feature words and phrases in the dictionary 141. Then, similar to the description of individual feature words and phrases, current affairs feature words and phrases may be preferentially extracted as feature words and phrases, displayed as selectable words, or highlighted.

[0121] Although the embodiments and modified examples of the present technology have been described above, the present technology is not limited to the above-described embodiments, and it goes without saying that various modifications can be made within the scope of the gist of the present technology.

[0122] The dialogue system according to one embodiment of the present invention can contribute to solving social issues such as loneliness and isolation among the elderly, extending healthy life expectancy, improving the quality of life (QoL) of the elderly, and a declining labor force. [Explanation of symbols]

[0123] 10 Dialogue Systems 100 Center Device 101 Receiving means 102 Extraction means 103 Response candidate generation means 104 Judgment means 105 Word display means 106 Response candidate display means 107 Response sending means 111 selectable phrases 112 User Messages 113 Cursor 114 Response Candidates 140 Memory means 141 Dictionaries 142 Dialogue History 160 Operator Device 200 Terminal Device 201 Input Method 202 Message sending method 203 Response output means M microphone S Speaker

Claims

1. A terminal device; a center device capable of communicating with the terminal device; A dialogue system comprising: The terminal device an input means for inputting a user message that is a message from a user; a message sending means for sending the input user message; a response output means for receiving and outputting a response message to the user message; and The center device a receiving means for receiving the user message from the terminal device; an extraction means for extracting one or more characteristic phrases included in the user message; a reply candidate generating means for generating one or more reply candidates which are candidates for the reply message in response to the user message; a phrase display means for displaying one or more selectable phrases from the characteristic phrases included in the user message so that the operator can select them; a response candidate display means for displaying, when a selectable phrase is selected from the one or more selectable phrases by the operator, one or more response candidates based on the selected selectable phrase in a selectable manner; a response sending means for, when a candidate answer is selected by the operator from the one or more candidate answers, sending the selected candidate answer to the terminal device as the response message; have Dialogue system.

2. 2. The dialogue system according to claim 1, The phrase display means displays the one or more selectable phrases on the user message in a manner that makes the one or more selectable phrases distinguishable from phrases other than the one or more selectable phrases and that makes the one or more selectable phrases selectable. Dialogue system.

3. 3. A dialogue system according to claim 1 or 2, The phrase display means determines the selectable phrases based on individual characteristic phrases set for each user from among the one or more characteristic phrases. Dialogue system.

4. Dialogue system according to any one of claims 1 to 3, The phrase display means highlights an individual characteristic phrase set for each user from among the one or more selectable phrases. Dialogue system.

5. 5. A dialogue system according to claim 3 or 4, The individual characteristic phrases for each user are set based on a dialogue history including the user message and the response message. Dialogue system.

6. Dialogue system according to any one of claims 1 to 5, The phrase display means displays some of the one or more characteristic phrases as the one or more selectable phrases based on the number of times the one or more characteristic phrases appear and / or their positions in the user message. Dialogue system.

7. 7. A dialogue system according to any one of claims 1 to 6, The phrase display means highlights some of the one or more selectable phrases based on the number of times the one or more characteristic phrases appear and / or their positions in the user message. Dialogue system.

8. 8. A dialogue system according to claim 6 or 7, the phrase display means performs display based on the number of times the one or more characteristic phrases appear, The number of times the one or more characteristic phrases appear is the number of times it appears in the user message; and / or The number of times the message appears in the dialogue history including the user message and the response message. Dialogue system.

9. 9. A dialogue system according to any one of claims 1 to 8, the answer candidate display means displays the one or more answer candidates in a selectable and editable manner; When the selected candidate reply is edited, the response sending means sends the edited candidate reply as the response message to the terminal device. Dialogue system.

10. Dialogue system according to any one of claims 1 to 9, the answer candidate generation means generates the one or more answer candidates based on the one or more feature expressions extracted by the extraction means; When a selectable phrase is selected from the one or more selectable phrases selectably displayed by the phrase display means, The answer candidate display means displays one or more answer candidates based on the selectable phrase selected from the one or more answer candidates generated by the answer candidate generation means in a selectable manner. Dialogue system.

11. Dialogue system according to any one of claims 1 to 9, When a selectable phrase is selected from the one or more selectable phrases selectably displayed by the phrase display means, the answer candidate generation means generates the one or more answer candidates based on the selected selectable phrase; The answer candidate display means displays the one or more answer candidates generated by the answer candidate generation means based on the selectable phrases in a selectable manner. Dialogue system.

12. a receiving means for receiving a user message from a terminal device that exchanges messages with a user in a two-way manner; an extraction means for extracting one or more characteristic phrases included in the user message; a reply candidate generating means for generating one or more reply candidates which are candidates for a reply message to the user message; a phrase display means for displaying one or more selectable phrases from the characteristic phrases included in the user message so that the operator can select them; a response candidate display means for displaying, when a selectable phrase is selected from the one or more selectable phrases by the operator, one or more response candidates based on the selected selectable phrase in a selectable manner; a response sending means for, when an operator selects a candidate answer from the one or more candidate answers, sending the selected candidate answer to the terminal device as the response message; A center device comprising:

Images