JP7652309B2 - Information processing device, chat control method, and chat control program - Google Patents

Description

The present invention relates to an information processing device, a voice chat control method, and a voice chat control program.

Support work at support centers and the like often involves replying to customers by email or other means. When replying to emails or other such messages, staff members work in pairs to check the messages to prevent problems with the content, misdelivery, or mistakes such as attaching the wrong documents. In this type of checking process, there is a risk of checking errors due to reduced concentration or fatigue when working late at night on a two-shift system or working all night. Electronic devices have also been proposed that can prevent operational errors when drowsiness is detected.

JP 2016-6957 A

When replying to emails, etc., one idea is to have a pair of people check the response, but one of the pair could be replaced by a chatbot and check the response via voice chat. However, if the chatbot is sleepy, for example, and responds inappropriately to a question, it will treat the check as complete once there is a response. This can lead to the person in charge checking while still sleepy, resulting in careless mistakes.

One aspect is to provide an information processing device, a voice chat control method, and a voice chat control program that can reduce careless mistakes caused by drowsiness.

In one aspect, the information processing device has a determination unit and a change unit. The determination unit determines whether the user is sleepy for each question based on information about a chatbot that asks questions and receives answers according to a scenario of questions in a checklist via chat with the user when the user sends an email to a customer. The change unit, when it is determined that the user is sleepy, interferes with the check process by the user .

It can help prevent careless mistakes caused by drowsiness.

FIG. 1 is a block diagram showing an example of a configuration of an information processing apparatus according to an embodiment. FIG. 2 is a diagram illustrating an example of the check item storage unit. FIG. 3 is a diagram illustrating an example of the response time table storage unit. FIG. 4 is a flowchart illustrating an example of a voice chat control process according to the embodiment. FIG. 5 is a diagram illustrating an example of a computer that executes a voice chat control program.

Below, examples of the information processing device, voice chat control method, and voice chat control program disclosed in the present application are described in detail with reference to the drawings. Note that the disclosed technology is not limited to the following examples. In addition, the following examples may be combined as appropriate within the scope of no contradiction.

FIG. 1 is a block diagram showing an example of the configuration of an information processing device according to an embodiment. In the information processing device 100, a chatbot is operating that performs voice chat with a user, for example, a support center staff member, and the chatbot asks questions to the user through voice chat and receives answers. The information processing device 100 determines whether the user is drowsy based on information regarding the dialogue between the chatbot and the user. If the information processing device 100 determines that the user is drowsy, it changes the characteristics of the voice of the chatbot. This enables the information processing device 100 to suppress careless mistakes caused by drowsiness.

As shown in FIG. 1, the information processing device 100 has a communication unit 110, a display unit 111, an operation unit 112, an input/output unit 113, a storage unit 120, and a control unit 130. Note that the information processing device 100 may have various functional units that known computers have in addition to the functional units shown in FIG. 1, such as various input devices and audio output devices.

The communication unit 110 is realized, for example, by a NIC (Network Interface Card) or the like. The communication unit 110 is connected to other information processing devices via a network (not shown) in a wired or wireless manner, and is a communication interface that manages the communication of information between the other information processing devices. The communication unit 110 receives, for example, check item data from the other information processing devices. The communication unit 110 outputs the received check item data to the control unit 130.

The display unit 111 is a display device for displaying various types of information. The display unit 111 is realized, for example, by a liquid crystal display or the like as a display device. The display unit 111 displays various screens, such as a display screen input from the control unit 130.

The operation unit 112 is an input device that accepts various operations from the user of the information processing device 100. The operation unit 112 is realized, for example, by a keyboard, a mouse, or the like as an input device. The operation unit 112 outputs the operation input by the user to the control unit 130 as operation information. Note that the operation unit 112 may be realized by a touch panel or the like as an input device, and the display device of the display unit 111 and the input device of the operation unit 112 may be integrated.

The input/output unit 113 has a microphone terminal and a headphone terminal, and is connected to a headset worn by a user of the information processing device 100. The input/output unit 113 converts an audio signal input from the user's headset into audio data and outputs it to the control unit 130. The input/output unit 113 also converts audio data input from the control unit 130 into an audio signal and outputs it to the user's headset.

The storage unit 120 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory), a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 120 has a check item storage unit 121 and a response time table storage unit 122. The storage unit 120 also stores information used for processing by the control unit 130.

The check item storage unit 121 stores, for example, check items in a checklist when sending an email. FIG. 2 is a diagram showing an example of a check item storage unit. As shown in FIG. 2, the check item storage unit 121 has items such as "Question No.", "Check Item", and "Answer". Furthermore, "Answer" further has items such as "Yes" and "No".

"Question No." is an identifier that identifies a question in the checklist, that is, a check item. "Check item" is information that indicates the content of a question in the checklist. "Yes" in the "Answer" column is information that indicates the next question No. when the user answers Yes to the check item. "No" in the "Answer" column is information that indicates the next question No. when the user answers No to the check item. For example, if the user answers "Yes" to the chatbot's question of the check item "Does the domain and customer name match?" in question No. "2", the process moves to question No. "3". On the other hand, if the user answers "No" to the question, the process moves to question No. "2-1". In other words, "Yes" and "No" in the "Answer" column are information that indicates the next question to the user depending on the answer. The "Answer" column may also have an item for when the question is not applicable.

Returning to the explanation of FIG. 1, the response time table storage unit 122 stores the average response time for each question in association with a threshold value. FIG. 3 is a diagram showing an example of the response time table storage unit. As shown in FIG. 3, the response time table storage unit 122 has items such as "Question No.", "Average response time", and "Threshold value".

"Question No." is an identifier that identifies a question in the checklist, that is, a check item. "Average answer time" is information that indicates the average past answer time for each question in seconds. "Threshold" is a threshold for detecting cases where a user takes a long time to answer a question from a chatbot, and is information that indicates an offset value for the average answer time in seconds. For example, for question No. "1," the first time, which is the detection standard for cases where a user takes a long time to answer, is set to average answer time + threshold = 5 seconds + 1 second = 6 seconds. Note that the first time may be stored as the "threshold" threshold. The first time may also be corrected using a correction value α that indicates individual differences between users.

Returning to the explanation of FIG. 1, the control unit 130 is realized, for example, by a CPU (Central Processing Unit) or MPU (Micro Processing Unit) executing a program stored in an internal storage device using a RAM as a working area. The control unit 130 may also be realized, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

The control unit 130 has a chatbot 131, a calculation unit 132, a determination unit 133, a change unit 134, and a generation unit 135, and realizes or executes the functions and actions of the information processing described below. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 1, and may be other configurations as long as they perform the information processing described below.

The chatbot 131 performs voice chat with the user, and checks a checklist in the form of the user answering questions posed by the chatbot. The chatbot 131 refers to the check item storage unit 121, and asks questions in order starting from question No. "1" according to a question scenario, and moves on to the next question depending on the user's answer. The chatbot 131 outputs voice data of the voice chat to the calculation unit 132 and the generation unit 135. Furthermore, when an instruction to change the voice characteristics is input from the change unit 134, the chatbot 131 changes the voice characteristics according to the instruction.

The calculation unit 132 detects the end time of the question from the chatbot 131 and the end time of the user's answer based on the voice data input from the chatbot 131. The calculation unit 132 calculates the answer time by calculating the difference between the end time of the question from the chatbot 131 and the end time of the user's answer. The calculation unit 132 outputs the calculated answer time to the determination unit 133. The calculation unit 132 also stores the calculated answer time in the memory unit 120 as history.

The determination unit 133, for example, refers to the real time clock (RTC) or high precision event timer (HPET) of the information processing device 100 to acquire the current time. The determination unit 133 determines whether the acquired current time is in a preset late-night time period. The late-night time period can be, for example, from 10 p.m. to 5 a.m. the following day. If the determination unit 133 determines that the time period is in the late-night time period, it determines that the user is sleepy. If the determination unit 133 determines that the user is sleepy, it outputs a first change instruction to the change unit 134 and proceeds to determining the answer time. On the other hand, if the determination unit 133 determines that the time period is not in the late-night time period (however, since sleepiness may occur after lunch even during the day, the user can set, change, or add the time period during which sleepiness occurs), it determines that the user is not sleepy. If the determination unit 133 determines that the user is not sleepy, it does not output a first change instruction to the change unit 134 and proceeds to determining the answer time.

The determination unit 133 refers to the answer time table storage unit 122 for each question, and calculates a first time, which is a detection standard for when the answer time is long, based on the average answer time and a threshold value. When the answer time is input from the calculation unit 132, the determination unit 133 determines whether the input answer time, i.e., the current answer time, is longer than the first time. If the determination unit 133 determines that the current answer time is equal to or shorter than the first time, it refers to past answer times (not shown) stored in the storage unit 120 and calculates the average answer time for each question. The determination unit 133 updates the average answer time for the corresponding question number in the answer time table storage unit 122 with the calculated average, and outputs a non-change instruction to the change unit 134 to not change the voice characteristics. On the other hand, if the determination unit 133 determines that the current answer time is longer than the first time, it outputs a second change instruction to the change unit 134 to change the voice characteristics.

When the first change instruction is input from the determination unit 133, the change unit 134 instructs the chatbot 131 to change the pitch of the voice to be higher than that during normal time periods, i.e., time periods other than late night time periods. For example, the change unit 134 instructs the chatbot 131 to raise the pitch by a whole tone during late night time periods compared to normal time periods.

When the second change instruction is input from the determination unit 133, the change unit 134 instructs the chatbot 131 to change the characteristics of the voice. For example, the change unit 134 instructs the chatbot 131 to change one or more characteristics of the pitch, speed, and volume as the voice characteristics. For example, the change unit 134 instructs the chatbot 131 to change from a male voice to a female voice, to speed up the playback speed of the voice, or to increase the volume.

In addition, the modification unit 134 changes the voice characteristics from the next question, for example, as the timing for changing the voice characteristics for the chatbot 131. Note that the modification unit 134 may change the voice characteristics in the middle of a question. Furthermore, the modification unit 134 may insert other voices or music between multiple questions, that is, between questions. Examples of other voices or music include words urging the user to wake up, or playing a phrase of fast-paced music.

When the modification unit 134 receives a no-change instruction from the determination unit 133, it does not instruct the chatbot 131 to change the voice characteristics. When the modification unit 134 receives a second change instruction or a no-change instruction from the determination unit 133 and performs an action in response to these instructions, it determines whether or not checking of all questions has been completed. When the modification unit 134 determines that checking of all questions has not been completed, it instructs the determination unit 133 to continue to determine the late-night time period and the answer time. When the modification unit 134 determines that checking of all questions has been completed, it ends the voice chat control process.

When the generation unit 135 receives voice data from the chatbot 131, it transcribes the voice data and generates a checked checklist that reflects the check contents. The generation unit 135 stores the generated checked checklist in the storage unit 120. The generation unit 135 may obtain the contents of the check items by referring to the check item storage unit 121. The generation unit 135 may also print the checked checklist using a printing device (not shown).

Next, the operation of the information processing device of the embodiment will be described. Figure 4 is a flowchart showing an example of the voice chat control process of the embodiment.

When the chatbot 131 starts a check with the user via voice chat, the determination unit 133 refers to the RTC or HPET to obtain the current time (step S1).

The determination unit 133 determines whether the acquired current time is in the late night hours (step S2). If the determination unit 133 determines that it is in the late night hours (step S2: Yes), it determines that the user is drowsy and outputs a first change instruction to the change unit 134. When the change unit 134 receives the first change instruction from the determination unit 133, it instructs the chatbot 131 to change the pitch of the voice to a higher pitch than in normal hours (step S3) and proceeds to step S4. At this time, the chatbot 131 changes the voice characteristics in accordance with the instruction to change the voice characteristics received from the change unit 134. On the other hand, if the determination unit 133 determines that it is not in the late night hours (step S2: No), it proceeds to step S4 without outputting the first change instruction.

The chatbot 131 refers to the check item memory unit 121, asks a question through voice chat, and accepts an answer to the question from the user. The chatbot 131 outputs the voice data of the voice chat to the calculation unit 132. The calculation unit 132 detects the end time of the question from the chatbot 131 and the end time of the user's answer based on the voice data input from the chatbot 131 (step S4).

The calculation unit 132 calculates the answer time based on the question end time of the chatbot 131 and the answer end time of the user (step S5). The calculation unit 132 outputs the calculated answer time to the determination unit 133.

When the response time is input from the calculation unit 132, the determination unit 133 determines whether the input current response time is longer than the first time (step S6). When the determination unit 133 determines that the current response time is equal to or shorter than the first time (step S6: No), it calculates the average response time for each question and updates the corresponding average response time in the response time table storage unit 122. In addition, the determination unit 133 outputs a no-change instruction to the modification unit 134 to not change the voice characteristics. When the no-change instruction is input from the determination unit 133, the modification unit 134 does not instruct the chatbot 131 to change the voice characteristics, and proceeds to step S8.

On the other hand, if the determination unit 133 determines that the current response time is longer than the first time (step S6: Yes), it outputs a second change instruction to the modification unit 134 to change the voice characteristics. When the second change instruction is input from the determination unit 133, the modification unit 134 instructs the chatbot 131 to change the voice characteristics (step S7) and proceeds to step S8. At this time, the chatbot 131 changes the voice characteristics in accordance with the instruction to change the voice characteristics input from the modification unit 134.

The change unit 134 determines whether or not checking of all questions has been completed (step S8). If the change unit 134 determines that checking of all questions has not been completed (step S8: No), the process returns to step S1. If the change unit 134 determines that checking of all questions has been completed (step S8: Yes), the change unit 134 ends the checking process. This enables the information processing device 100 to suppress careless mistakes caused by drowsiness.

In this way, the information processing device 100 determines whether the user is drowsy based on information about the dialogue between the chatbot 131, which asks questions and receives answers via voice chat with the user, and the user. Furthermore, if the information processing device 100 determines that the user is drowsy, it changes the voice characteristics of the chatbot 131. As a result, it is possible to suppress careless mistakes caused by drowsiness. In other words, if the user is answering a question while in a sleepy state, it is possible to encourage the user to wake up.

The information processing device 100 also refers to the answer time table storage unit 122, which stores an average answer time calculated based on the question end time of the chatbot 131 and the answer end time of the user, in association with a threshold value for each question, and if the calculated answer time is longer than a first time based on the average value and the threshold value, it determines that the user is drowsy. As a result, if the user's reaction time to a question is long, it can be determined that the user is drowsy.

The characteristics of the audio may be one or more of pitch, speed, and volume. As a result, the user may be prompted to wake up.

Furthermore, if the information processing device 100 determines that the user is sleepy, it changes the characteristics of the voice from the next question or during the middle of a question. As a result, it is possible to urge the user to wake up.

Furthermore, if the information processing device 100 determines that the user is sleepy, it inserts other voices or music between the multiple questions. As a result, it is possible to urge the user to wake up.

In addition, the information processing device 100 acquires the current time as information related to the dialogue, and if the current time is within a preset late-night time period or a time period during which drowsiness occurs, it determines that the user is drowsy. In addition, if the information processing device 100 determines that the user is drowsy, it changes the pitch of the voice of the chatbot 131 to be higher than that during times other than the late-night time period.

In the above embodiment, the check process is performed using voice chat, but the present invention is not limited to this. For example, the present invention can be applied to a check process in which a question is displayed on a display screen and input such as "Y" or "N" from a keyboard or selection using a mouse or touch panel is accepted. In this case, it is possible to check that the user's five senses are working by interrupting the check process by displaying a cancel button on the display screen when the answer time is longer than the first time, and restarting the check process by accepting an operation on the cancel button.

In addition, in the above embodiment, the information processing device 100 performed the operation of the chatbot and the voice chat control processing, but this is not limited to the above. For example, the operation of the chatbot and the voice chat control processing may be performed on a server, and the user may make a call to the server to conduct a voice chat. Furthermore, the chatbot may be operated on another server.

Furthermore, each component of each unit shown in the figure does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution/integration of each unit is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed/integrated in any unit depending on various loads, usage conditions, etc. For example, the determination unit 133 and the change unit 134 of the information processing device 100 may be integrated. Furthermore, each process shown in the figure is not limited to the above order, and may be performed simultaneously or in a reverse order as long as the process contents are not contradictory.

Furthermore, the various processing functions performed by each device may be executed in whole or in part on a CPU (or a microcomputer such as an MPU or MCU (Micro Controller Unit)). Needless to say, the various processing functions may be executed in whole or in part on a program that is analyzed and executed by a CPU (or a microcomputer such as an MPU or MCU), or on hardware using wired logic.

The various processes described in each of the above embodiments can be realized by executing a program prepared in advance on a computer. Therefore, an example of a computer that executes a program having the same functions as the above embodiments will be described below. Figure 5 shows an example of a computer that executes a voice chat control program.

As shown in FIG. 5, computer 200 has a CPU 201 that executes various arithmetic processes, an input device 202 that accepts data input, and a monitor 203. Computer 200 also has a medium reading device 204 that reads programs and the like from a storage medium, an interface device 205 for connecting to various devices, and a communication device 206 for connecting to other information processing devices and the like by wire or wirelessly. Computer 200 also has a RAM 207 that temporarily stores various information, and a hard disk device 208. Each of devices 201 to 208 is connected to a bus 209.

The hard disk drive 208 stores a voice chat control program having the same functions as the chatbot 131, calculation unit 132, judgment unit 133, change unit 134, and generation unit 135 shown in FIG. 1. The hard disk drive 208 also stores a check item storage unit 121 and a response time table storage unit 122. The input device 202, for example, accepts input of various information such as operation information from the user of the computer 200. The monitor 203, for example, displays various screens such as a display screen to the user of the computer 200. The interface device 205 is connected to, for example, a headset or a printing device. The communication device 206, for example, has the same functions as the communication unit 110 shown in FIG. 1, is connected to a network (not shown), and exchanges various information with other information processing devices.

The CPU 201 reads out each program stored in the hard disk device 208, expands it in the RAM 207, and executes it to perform various processes. These programs can also cause the computer 200 to function as the chatbot 131, the calculation unit 132, the determination unit 133, the change unit 134, and the generation unit 135 shown in FIG. 1.

The voice chat control program does not necessarily have to be stored in the hard disk drive 208. For example, the computer 200 may read and execute a program stored in a storage medium that the computer 200 can read. Examples of storage media that the computer 200 can read include portable storage media such as CD-ROMs, DVDs (Digital Versatile Discs), and USB (Universal Serial Bus) memories, semiconductor memories such as flash memories, and hard disk drives. The voice chat control program may also be stored in a device connected to a public line, the Internet, a LAN, or the like, and the computer 200 may read and execute the voice chat control program from the device.

REFERENCE SIGNS LIST 100 Information processing device 110 Communication unit 111 Display unit 112 Operation unit 113 Input/output unit 120 Storage unit 121 Check item storage unit 122 Answer time table storage unit 130 Control unit 131 Chatbot 132 Calculation unit 133 Determination unit 134 Change unit 135 Generation unit

Claims (7)

a chatbot that, when a user sends an email to a customer, displays questions on a display screen according to a scenario of a checklist of questions through chat with the user and receives answers from the user inputted via one or more of a keyboard, a mouse and a touch panel ; and a determination unit that determines whether the user is feeling drowsy for each question based on information about the dialogue with the user.
a change unit that, when it is determined that the user is drowsy, blocks a check process by the user by displaying a cancel button on the display screen ;
13. An information processing device comprising: The determination unit refers to a storage unit that stores an average answer time calculated based on a question end time of the chatbot and a response end time of the user, and a threshold value in association with each question, and determines that the user is drowsy when the calculated response time is longer than a first time based on the average value and the threshold value.
2. The information processing apparatus according to claim 1 , When it is determined that the user is drowsy, the change unit executes the interference from the next question or in the middle of the question.
3. The information processing apparatus according to claim 1 , wherein the information processing apparatus is a computer . The change unit, when it is determined that the user is drowsy, inserts another voice or music between the plurality of questions.
4. The information processing device according to claim 1, wherein the information processing device is a computer. The determination unit acquires a current time as information related to the dialogue, and determines that the user is drowsy if the current time is within a preset late-night time period or a preset drowsiness time period.
2. The information processing apparatus according to claim 1 , When a user sends an email to a customer, a chatbot displays questions on a display screen according to a scenario of questions in a checklist through chat with the user , and receives answers from the user inputted through one or more of a keyboard, a mouse, and a touch panel , and determines whether the user is feeling drowsy for each question based on information about the dialogue between the chatbot and the user;
When it is determined that the user is drowsy, a check process by the user is hindered by displaying a cancel button on the display screen .
A chat control method, the processing of which is executed by a computer. When a user sends an email to a customer, a chatbot displays questions on a display screen according to a scenario of questions in a checklist through chat with the user , and receives answers from the user inputted through one or more of a keyboard, a mouse, and a touch panel , and determines whether the user is feeling drowsy for each question based on information about the dialogue between the chatbot and the user;
When it is determined that the user is drowsy, a check process by the user is hindered by displaying a cancel button on the display screen .
A chat control program that causes a computer to execute a process.

Images