Information processing system, information processing device, information processing method, and program

Abstract

Display analysis results based on predicted turnover rates for operators and other employees. [Solution] The information processing system comprises: an audio data acquisition unit that acquires audio data; an extraction unit that extracts a plurality of emotion values ​​representing the speaker's emotions and text data representing the content of the speaker's speech from the audio data for each speaker; a first prediction unit that predicts a predicted value of the probability of leaving the company by inputting the plurality of emotion values ​​and the text data into a first model based on the correspondence between the emotion values ​​and text data and the probability of leaving the company; a second prediction unit that predicts a predicted value of the customer service skills by inputting the plurality of emotion values ​​and the text data into a second model based on the correspondence between the emotion values ​​and text data and customer service skills; and an output unit that outputs the predicted value of the probability of leaving the company and the predicted value of the customer service skills.

Description

Technical Field

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

Background Art

[0002] There is a technology for predicting the probability of an operator's resignation in call center operations. For example, Patent Document 1 provides a computer implementation method relating to predicting the probability of employee turnover for agents employed in a contact center, the method comprising the steps of providing an employee turnover model, the employee turnover model comprising a machine learning model trained according to a training dataset of corresponding inputs and outputs, the training of the employee turnover model comprising learning patterns in the inputs that indicate values ​​for the outputs, the inputs comprising a plurality of data types, including one or more data types comprising agent employment data, one or more data types comprising agent interaction data, and one or more data types comprising agent adherence data, and the outputs comprising the probability of employee turnover for the agents, and measuring and recording agent journey data of a first agent currently employed in a contact center, the agent journey data describing aspects related to the employment of the first agent in the contact center, and including data types whose types correspond to the inputs of the employee turnover model, and the first agent A computer implementation method is disclosed, which includes the steps of: determining that it is necessary to predict the current turnover probability of a first agent; using a turnover model, which includes providing values ​​for input to the turnover model from applicable current values ​​taken from corresponding data types of agent journey data of a first agent; and, given the provided inputs, calculating the current turnover probability of the first agent as the output of the turnover model; determining whether the calculated current turnover probability of the first agent satisfies a threshold turnover probability, where satisfies the threshold turnover probability indicates that the first agent has a high turnover risk; and generating and transmitting an alert communication to a computing device associated with a second employee of a contact center in response to the determination that the first agent has a high turnover risk. [Prior art documents] [Patent Documents]

[0003] [Patent Document 1] Special Publication No. 2023-540970 [Overview of the Initiative] [Problems that the invention aims to solve]

[0004] However, the technology described in Patent Document 1 has the problem that, although it can extract operators with a high predicted probability of leaving the company and notify the manager, it cannot display the analysis results based on the predicted probability of leaving the company.

[0005] One aspect of the present invention has been made in view of the above points, and aims to provide an information processing system, information processing device, information processing method, and program that can display analysis results based on predicted values ​​of the probability of employees leaving the company, such as operators. [Means for solving the problem]

[0006] The present invention has been made to solve the above problems, and one aspect of the present invention is an information processing system comprising: an audio data acquisition unit that acquires audio data; an extraction unit that extracts a plurality of emotion values ​​representing the emotions of the speaker and text data representing the content of the speech made by the speaker from the audio data for each speaker; a first prediction unit that predicts a predicted value of the probability of leaving the company by inputting the plurality of emotion values ​​and the text data into a first model based on the correspondence between the emotion values ​​and the text data and the probability of leaving the company; a second prediction unit that predicts a predicted value of the customer service skills by inputting the plurality of emotion values ​​and the text data into a second model based on the correspondence between the emotion values ​​and the text data and customer service skills; and an output unit that outputs the predicted value of the probability of leaving the company and the predicted value of the customer service skills.

[0007] Furthermore, one aspect of the present invention is an information processing device comprising: an audio data acquisition unit that acquires audio data; an extraction unit that extracts a plurality of emotion values ​​representing the emotions of the speaker and text data representing the content of the speech made by the speaker from the audio data for each speaker; a first prediction unit that predicts a predicted value of the probability of leaving the company by inputting the plurality of emotion values ​​and the text data into a first model based on the correspondence between the emotion values ​​and the text data and the probability of leaving the company; a second prediction unit that predicts a predicted value of the customer service skills by inputting the plurality of emotion values ​​and the text data into a second model based on the correspondence between the emotion values ​​and the text data and customer service skills; and an output unit that outputs the predicted value of the probability of leaving the company and the predicted value of the customer service skills.

[0008] Furthermore, one aspect of the present invention is an information processing method executed by a computer, comprising: an audio data acquisition step of acquiring audio data; an extraction step of extracting a plurality of emotion values ​​representing the emotions of the speaker and text data representing the content of the speech made by the speaker from the audio data for each speaker; a first prediction step of predicting a predicted value of the probability of leaving the company by inputting the plurality of emotion values ​​and the text data into a first model based on the correspondence between the emotion values ​​and text data and the probability of leaving the company; a second prediction step of predicting a predicted value of the customer service skills by inputting the plurality of emotion values ​​and the text data into a second model based on the correspondence between the emotion values ​​and text data and customer service skills; and an output step of outputting the predicted value of the probability of leaving the company and the predicted value of the customer service skills.

[0009] Furthermore, one aspect of the present invention is a program that causes a computer to execute the following steps: an audio data acquisition step of acquiring audio data; an extraction step of extracting a plurality of emotion values ​​representing the emotions of the speaker and text data representing the content of the speech made by the speaker from the audio data for each speaker; a first prediction step of predicting a predicted value of the probability of leaving the company by inputting the plurality of emotion values ​​and the text data into a first model based on the correspondence between the emotion values ​​and text data and the probability of leaving the company; a second prediction step of predicting a predicted value of the customer service skills by inputting the plurality of emotion values ​​and the text data into a second model based on the correspondence between the emotion values ​​and text data and customer service skills; and an output step of outputting the predicted value of the probability of leaving the company and the predicted value of the customer service skills. [Effects of the Invention]

[0010] According to the present invention, it is possible to display analysis results based on predicted values ​​of the probability of operators and other employees leaving their jobs. [Brief explanation of the drawing]

[0011] [Figure 1] This is a system configuration diagram showing an example of the configuration of an information processing system according to the first embodiment of the present invention. [Figure 2] This is a block diagram showing an example of the hardware configuration of the information processing device according to this embodiment. [Figure 3] This is a block diagram showing an example of the functional configuration of the information processing device according to this embodiment. [Figure 4] This block diagram shows an example of the functional configuration of the call terminal device according to this embodiment. [Figure 5] This block diagram shows an example of the functional configuration of the operating terminal device according to this embodiment. [Figure 6] This flowchart shows an example of processing by the telephone terminal device and information processing device according to this embodiment. [Figure 7] This flowchart shows an example of processing in the information processing device according to this embodiment. [Figure 8]It is a flowchart showing an example of information processing in the information processing apparatus according to the present embodiment. [Figure 9] It is a flowchart showing an example of information processing in the information processing apparatus according to the present embodiment. [Figure 10] It is a flowchart showing an example of processing in the information processing apparatus and the operation terminal apparatus according to the present embodiment. [Figure 11] It is a flowchart showing an example of processing in the operation terminal apparatus according to the present embodiment. [Figure 12] It is a diagram showing an example of text data stored in the storage unit of the information processing apparatus according to the present embodiment. [Figure 13] It is a diagram showing an example of a resignation probability prediction value stored in the storage unit of the information processing apparatus according to the present embodiment. [Figure 14] It is a diagram showing an example of a response skill prediction value stored in the storage unit of the information processing apparatus according to the present embodiment. [Figure 15] It is a diagram showing an example of a display screen in the operation terminal apparatus according to the present embodiment. [Figure 16] It is a diagram showing another example of a display screen in the operation terminal apparatus according to the present embodiment. [Figure 17] It is a diagram showing another example of a display screen in the operation terminal apparatus according to the present embodiment.

Mode for Carrying Out the Invention

[0012] [First Embodiment] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0013] <Configuration of Information Processing System SYS> First, the configuration of the information processing system SYS will be described.

[0014] FIG. 1 is a system configuration diagram showing an example of the configuration of an information processing system SYS according to the first embodiment of the present invention. The information processing system SYS comprises an information processing device 100, a call terminal device 200, and an operation terminal device 300. The information processing device 100, the call terminal device 200, and the operation terminal device 300 are connected to each other via a network NW so that they can communicate with one another.

[0015] This section provides a more detailed explanation of the SYS information processing system.

[0016] The SYS information processing system extracts multiple emotion values ​​representing multiple emotions from multiple speakers (e.g., customer and operator) from audio data of customer-operator calls in a call center, and extracts text data representing the content of each speaker's utterance. The SYS system then inputs the text data and emotion values ​​into a first model to predict the probability of employee turnover. The SYS system also inputs the text data and emotion values ​​into a second model to predict customer service skills. Finally, the SYS system outputs predicted values ​​for the probability of employee turnover and predicted values ​​for customer service skills as analysis results based on the probability of employee turnover.

[0017] Here, employee turnover rate and customer service skills refer, for example, to the employee turnover rate and customer service skills of an operator.

[0018] This configuration allows the SYS information processing system to display analysis results based on predicted employee turnover rates for operators and other employees.

[0019] Let's explain the SYS information processing system in more detail.

[0020] The information processing device 100 is a server device that calculates predicted values ​​for employee turnover probability and predicted values ​​for customer service skills. The information processing device 100 acquires voice data of a call from the call terminal device 200. From the voice data, the information processing device 100 extracts multiple emotion values ​​that represent the emotions of each speaker (for example, operator and customer). Emotions are, for example, emotions that the customer or operator feels during a call, such as joy, anger, sadness, happiness, empathy, kindness, and sincerity. The emotion value is a numerical value that indicates the degree of the above emotions.

[0021] Furthermore, the emotions used are not limited to those mentioned above; any emotion may be used as long as it is an emotion felt by the customer or operator.

[0022] Furthermore, the information processing device 100 extracts text data representing the content of each utterance from the audio data using speech recognition. This text data corresponds to the audio data from which the emotion value has been extracted.

[0023] The telephone terminal device 200 is a terminal device used for making telephone calls. In this embodiment, the call terminal device 200 is a terminal device used by an operator to make calls with customers. The call terminal device 200 has the function of making calls with customers and the function of recording those calls.

[0024] The communication terminal device 200 may be a telephone, smartphone, communication application, chat tool, etc., or it may be a recorder or camera capable of recording voice data. Furthermore, while this embodiment uses a call center conversation as an example, it can be applied to any conversation involving multiple people, whether face-to-face or online.

[0025] The operating terminal device 300 is a terminal device used by operators, such as administrators. The operating terminal device 300 displays predicted values ​​for employee turnover probability and customer service skills, the temporal changes in predicted employee turnover probability, and predicted employee turnover probability for multiple people (e.g., comparative display).

[0026] Next, the hardware configuration of the information processing device 100 will be described.

[0027] <Hardware Configuration> Figure 2 is a block diagram showing an example of the hardware configuration of the information processing device according to this embodiment. The information processing device 100 comprises a CPU 101, a storage medium interface unit 102, a storage medium 103, an input device 104, an output device 105, a ROM 106 (Read Only Memory), a RAM 107 (Random Access Memory), an auxiliary storage unit 108, and a network interface unit 109. The CPU 101, the storage medium interface unit 102, the input device 104, the output device 105, the ROM 106, the RAM 107, the auxiliary storage unit 108, and the network interface unit 109 are interconnected via a bus.

[0028] In this context, "CPU101" refers to a processor in general, and includes not only devices commonly known as CPUs in the narrow sense, but also, for example, GPUs and DSPs. Furthermore, "CPU101" is not limited to being implemented by a single processor, but may be implemented by combining multiple processors of the same or different types.

[0029] <cpu101> The CPU 101 controls the information processing device 100 by reading and executing programs stored in the auxiliary storage unit 108, ROM 106, and RAM 107, and by reading various data stored in the auxiliary storage unit 108, ROM 106, and RAM 107 and writing various data to the auxiliary storage unit 108 and RAM 107. The CPU 101 also reads various data stored in the storage medium 103 via the storage medium interface unit 102 and writes various data to the storage medium 103.

[0030] <Storage medium 103> The storage medium 103 is a portable storage medium such as a magneto-optical disk, a flexible disk, or flash memory, and stores various types of data.

[0031] <Storage medium interface unit 102> The storage medium interface unit 102 is an interface for reading and writing to the storage medium 103.

[0032] <Input device 104> The input device 104 includes input devices such as a mouse, keyboard, touch panel, microphone, volume control buttons, power button, settings button, and infrared receiver.

[0033] <Output device 105> The output device 105 is an output device such as a display unit or a speaker.

[0034] <ROM106、RAM107> ROM 106 and RAM 107 store programs and various data necessary to operate each functional unit of the information processing device 100.

[0035] <Auxiliary storage section 108> The auxiliary storage unit 108 is a hard disk drive, flash memory, etc., and stores programs and various data for operating each functional unit of the information processing device 100.

[0036] <Network Interface Unit 109> The network interface unit 109 has a communication interface and is connected to the network NW via wireless communication.

[0037] For example, the CPU 101 of the information processing device 100 corresponds to the control unit 150 in the functional configuration shown in Figure 3. Also, the ROM 106, RAM 107, auxiliary storage unit 108, or any combination thereof of the information processing device 100 corresponds to the storage unit 120 in the functional configuration shown in Figure 3. Furthermore, the input device 104 and output device 105 of the information processing device 100 correspond to the input unit 130 and output unit 140 in the functional configuration shown in Figure 3.

[0038] Although the hardware configuration of the call terminal device 200 and the operation terminal device 300 are not shown in the diagrams and descriptions, they have the same hardware configuration as the information processing device 100 shown in Figure 2.

[0039] Next, the functional configuration of the information processing device 100 will be described.

[0040] <Functional configuration of the information processing device 100> Figure 3 is a block diagram showing an example of the functional configuration of the information processing device 100 according to this embodiment. The information processing device 100 includes a communication unit 110, a storage unit 120, an input unit 130, an output unit 140, and a control unit 150. The communication unit 110, the storage unit 120, the input unit 130, the output unit 140, and the control unit 150 are interconnected via a bus.

[0041] <Communications Department 110> The communication unit 110 has the function of communicating with the call terminal device 200. The communication unit 110 outputs various information received from the call terminal device 200 to the control unit 150. The communication unit 110 also outputs various information received from the operation terminal device 300 to the control unit 150. Furthermore, the communication unit 110 transmits information input from the control unit 150 to the operation terminal device 300.

[0042] <Storage section 120> The storage unit 120 is composed of a storage medium, such as an HDD (Hard Disk Drive), flash memory, EEPROM (Electrically Erasable Programmable Read Only Memory), RAM (Random Access read / write Memory), ROM (Read Only Memory), or any combination of these storage media. This storage unit 120 can, for example, use non-volatile memory.

[0043] The memory unit 120 stores text data 121, predicted employee turnover rate 122, predicted customer service skills 123, the first model 124, and the second model 125.

[0044] Text data 121 is text data representing the content of each speaker's utterance, generated by speech recognition on the audio data. Here, text data 121 associates information about the call with emotion values ​​extracted from the audio data.

[0045] The predicted turnover probability value 122 is the predicted turnover probability for each operator over a predetermined period. The predetermined period can be, for example, daily, every other day, weekly, monthly, or yearly.

[0046] The predicted response skill value of 123 is a predicted value for the response skill of each operator for each call.

[0047] The first model is based on the correspondence between text data and sentiment values ​​and the probability of employee turnover. More specifically, it is a learned model that has learned the correspondence between text data and sentiment values ​​and the probability of employee turnover. When text data and sentiment values ​​are input into this learned model, a predicted value of the probability of employee turnover is obtained as output. The first model may also be a generative AI (Artificial Intelligence). If the first model is a generative AI, it may reside outside the information processing device 100 and be configured to be called and used from the information processing device 100.

[0048] The second model is based on the correspondence between text data and sentiment values ​​and customer service skills. More specifically, it is a learning model that has learned the correspondence between text data and sentiment values ​​and evaluation values ​​of customer service skills. When text data and sentiment values ​​are input into this learning model, it outputs predicted values ​​for customer service skills. The second model may be a generative AI. If the second model is a generative AI, it may reside outside the information processing device 100 and be configured to be called and used by the information processing device 100.

[0049] <Input section 130> The input unit 130 is an input device such as a mouse, keyboard, or microphone connected to the information processing device 100. The input unit 130 receives operation input from an external source. The input unit 130 outputs an operation signal corresponding to the operation input to the control unit 150.

[0050] <Output section 140> The output unit 140 is an output device such as a display device. The output unit 140 outputs the presentation information output from the control unit 150 to an output device or another device such as an operation terminal device 300.

[0051] <Control Unit 150> The control unit 150 has the function of controlling the information processing device 100. The control unit 150 reads various data, applications, programs, etc. stored in the storage unit 120 and controls the information processing device 100.

[0052] The processing of the control unit 150 will be described in more detail. The control unit 150 includes an audio data acquisition unit 151, a text data extraction unit 152, a turnover probability prediction unit 153, a customer service skill prediction unit 154, and an output processing unit 155.

[0053] <Audio data acquisition unit 151> The voice data acquisition unit 151 acquires voice data from the call terminal device 200 via the network NW and the communication unit 110. The voice data acquisition unit 151 outputs the voice data to the text data extraction unit 152.

[0054] <Text data extraction unit 152> When audio data is input from the audio data acquisition unit 151, the text data extraction unit 152 extracts text data from the audio data through speech recognition. This text data is text data in which the speaker and the content of each speaker's speech are transcribed into text through speaker recognition. The text data extraction unit 152 stores the extracted text data in the storage unit 120, associating it with identification information that identifies the call.

[0055] Furthermore, the text data extraction unit 152 extracts multiple emotional values ​​of the speaker, such as a customer and an operator, from the audio data for each utterance, for example, using known techniques. The text data extraction unit 152 stores the extracted emotional values ​​in the storage unit 120, associating them with the text data. Here, speaker recognition is performed by recognizing the voices of multiple speakers using existing speech separation technologies, etc. Furthermore, the emotion value is an evaluation value that represents the speaker's emotion.

[0056] The text data may also correspond to audio data for a predetermined period from the start of the call, for example, one minute. In this case, the specified period may be the period until the end of the conversation. Alternatively, the specified period may be a certain period of time preceding the end of the conversation. Alternatively, the specified period may be a predetermined time elapsed from the time a specific keyword was uttered during the conversation. Furthermore, sentiment values ​​may also be statistical measures such as the mean, median, or variance obtained by averaging each sentiment value over a predetermined period.

[0057] <Employee Turnover Probability Prediction Unit 153> The employee turnover probability prediction unit 153 obtains the data by reading text data 121 containing sentiment values ​​from the memory unit 120. The employee turnover probability prediction unit 153 converts the text data 121 containing sentiment values ​​into a format that can be input into the first model 124. This conversion to a format that can be input into the first model 124 involves, for example, aggregating the text data 121 containing sentiment values ​​of operators and customers so that each utterance corresponds to a certain period. This aggregation is performed to predict the predicted value of the employee turnover probability for each operator over a certain period. The certain period can be annual, monthly, weekly, daily, etc. For the aggregation of text data 121 containing sentiment values, statistics such as the mean, median, maximum, minimum, standard deviation, and variance (calculated by averaging each sentiment value) are used.

[0058] The employee turnover probability prediction unit 153 may also integrate the emotion value and text data for each utterance and use the integrated text data. Furthermore, the employee turnover probability prediction unit 153 may extract words (frequently occurring words) or phrases for each utterance in the text data, quantify and aggregate the importance of each extracted word or phrase using, for example, TF-IDF (Term Frequency - Inverse Document Frequency), or it may aggregate the frequency of word occurrences. Furthermore, the employee turnover probability prediction unit 153 may obtain only the emotion value from the memory unit 120, or it may obtain only text data that does not include the emotion value.

[0059] The employee turnover probability prediction unit 153 inputs the aggregated text data 121, which includes emotional values, into the first model 124 stored in the memory unit 120, thereby obtaining a predicted value of employee turnover probability as output. The predicted value of employee turnover probability output by the first model 124 is the predicted value of employee turnover probability for each operator and for each period of time. The employee turnover probability prediction unit 153 stores the predicted value of employee turnover probability as the employee turnover probability prediction value 122 in the memory unit 120.

[0060] <Customer Service Skill Prediction Unit 154> The response skill prediction unit 154 acquires the data by reading text data 121 from the memory unit 120. The response skill prediction unit 154 converts the text data 121, which includes the emotional values ​​of the operator and the customer, into a format that can be input into the second model 125. This conversion to a format that can be input into the second model 125 involves, for example, aggregating the text data 121 containing emotional values ​​so that each utterance in the text data 121 corresponds to a call unit. This aggregation is performed to predict the predicted value of the operator's response skill for each call. For the aggregation of text data 121 containing sentiment values, statistics such as the mean, median, maximum, minimum, standard deviation, and variance (calculated by averaging each sentiment value) are used.

[0061] The response skill prediction unit 154 may also integrate the emotion value and text data for each utterance and use the integrated text data. Furthermore, the response skill prediction unit 154 may extract words (frequently occurring words) and phrases for each utterance in the text data, quantify and aggregate the importance of each extracted word or phrase using, for example, TF-IDF, or aggregate the frequency of word occurrences. Furthermore, the customer service skill prediction unit 154 may acquire only the emotion value from the memory unit 120, or it may acquire only text data that does not include the emotion value. Also, the emotion value may be the emotion value of the customer only, or it may be the emotion value of the operator only.

[0062] Here, the evaluation value (predicted value) of the operator's response skills is an indicator that represents the operator's knowledge of the product or service and whether they are handling calls appropriately. Examples include customer satisfaction, NPS (Net Promoter Score), and scores based on response quality rubrics set by the company (users, operators).

[0063] The response skill prediction unit 154 inputs the aggregated text data 121, which includes emotional values, into the second model 125 stored in the memory unit 120, thereby obtaining a predicted value of response skill as output. The predicted value of response skill output by the second model 125 is the predicted value of response skill for each call. The response skill prediction unit 154 stores the predicted value of response skill as the predicted response skill value 123 in the memory unit 120.

[0064] <Output processing unit 155> When the output processing unit 155 receives a data acquisition request from the operating terminal device 300 via the network NW and communication unit 110, it reads out the employee turnover probability prediction value 122 and the customer service skill prediction value 123 stored in the storage unit 120. The output processing unit 155 then transmits the employee turnover probability prediction value 122 and the customer service skill prediction value 123 to the operating terminal device 300.

[0065] Next, the functional configuration of the call terminal device 200 will be described.

[0066] <Telephone terminal device 200> Figure 4 is a block diagram showing an example of the functional configuration of the telephone terminal device 200 according to this embodiment. The telephone terminal device 200 comprises a communication unit 210, a storage unit 220, an input unit 230, an output unit 240, and a control unit 250. The communication unit 210, storage unit 220, input unit 230, output unit 240, and control unit 250 are interconnected via a bus.

[0067] <Communications Department 210> The communication unit 210 has the function of communicating with the information processing device 100 and the operation terminal device 300. The communication unit 210 outputs various information received from the information processing device 100 to the control unit 150. The communication unit 210 also outputs various information received from the operation terminal device 300 to the control unit 150. Furthermore, the communication unit 210 transmits information input from the control unit 250 to the information processing device 100.

[0068] <Storage section 220> The storage unit 220 is composed of a storage medium, such as an HDD, flash memory, EEPROM, RAM, ROM, or any combination thereof. This storage unit 220 can, for example, use non-volatile memory.

[0069] The memory unit 220 stores the audio data 221. The voice data 221 is, for example, voice data relating to a call between an operator and a customer. This voice data also includes information related to the call, such as the date and time of the call, the time of day of the call, the holding time, the type of call (incoming or outgoing), and information identifying the call terminal device 200.

[0070] <Input section 230> The input unit 230 is an input device such as a mouse, keyboard, microphone, or headset that is connected to the communication terminal device 200. The input unit 230 receives operation input from an external source. The input unit 230 outputs an operation signal to the control unit 250 in accordance with the operation input.

[0071] <Output section 240> The output unit 240 is an output device such as a display device or a speaker. The output unit 240 outputs the information output from the control unit 250 to the output device or other devices.

[0072] <Control Unit 250> The control unit 250 has the function of controlling the telephone terminal device 200. The control unit 250 reads various data, applications, programs, etc. stored in the storage unit 220 and controls the telephone terminal device 200.

[0073] The processing of the control unit 250 will be described in more detail.

[0074] The control unit 250 is configured to include a sound collection unit 251. <Sound collection section 251> The sound collection unit 251 acquires the audio data input from the input unit 230 and stores it as audio data 221 in the storage unit 220. The sound collection unit 251 transmits the audio data 221 to the information processing device 100 via the communication unit 210 and the network NW.

[0075] Next, the functional configuration of the operating terminal device 300 will be described.

[0076] <Operating terminal device 300> Figure 5 is a block diagram showing an example of the functional configuration of the operating terminal device 300 according to this embodiment. The operating terminal device 300 comprises a communication unit 310, a storage unit 320, an input unit 330, an output unit 340, and a control unit 350. The communication unit 310, storage unit 320, input unit 330, output unit 340, and control unit 350 are interconnected via a bus.

[0077] <Communications Department 310> The communication unit 310 has the function of communicating with the information processing device 100 and the telephone terminal device 200. The communication unit 310 outputs various information received from the information processing device 100 to the control unit 350. The communication unit 310 also outputs various information received from the telephone terminal device 200 to the control unit 350. Furthermore, the communication unit 310 transmits information input from the control unit 350 to the information processing device 100.

[0078] <Storage section 320> The storage unit 320 is composed of a storage medium, such as an HDD, flash memory, EEPROM, RAM, ROM, or any combination thereof. This storage unit 320 can, for example, use non-volatile memory.

[0079] The memory unit 320 stores employee turnover probability prediction data 321 and customer service skill prediction data 322. The employee turnover probability prediction data 321 is the employee turnover probability prediction value 122 obtained from the information processing device 100. The customer service skill prediction data 322 is the customer service skill prediction value 123 obtained from the information processing device 100.

[0080] <Input section 330> The input unit 330 is an input device such as a mouse or keyboard connected to the operation terminal device 300. The input unit 330 receives operation input from an external source. The input unit 330 outputs an operation signal corresponding to the operation input to the control unit 350.

[0081] <Output section 340> The output unit 340 is an output device such as a display device or a speaker. The output unit 340 outputs the information output from the control unit 350 to the output device or other devices.

[0082] <Control Unit 350> The control unit 350 has the function of controlling the operation terminal device 300. The control unit 350 reads various data, applications, programs, etc. stored in the storage unit 320 and controls the operation terminal device 300.

[0083] The processing of the control unit 350 will be explained in more detail.

[0084] The control unit 350 is comprised of a data acquisition unit 351 and a display processing unit 352. <Data acquisition unit 351> The data acquisition unit 351 transmits a data acquisition request to the information processing device 100 based on user operations such as those of an administrator. When the data acquisition unit 351 acquires the predicted employee turnover rate 122 and the predicted customer service skills 123 from the information processing device 100 in response to the data acquisition request, it stores them in the storage unit 320 as predicted employee turnover rate data 321 and predicted customer service skills data 322.

[0085] <Display Processing Unit 352> The display processing unit 352 receives input of the display unit and display period from a user such as an administrator. The display unit is information that specifies a unit such as a year, month, week, or day. The display period is information that specifies the start and end dates for display. The display processing unit 352 reads the employee turnover probability prediction data 321 from the storage unit 320. The display processing unit 352 aggregates the predicted employee turnover probability for each operator within the display unit and display period. The mean, median, etc., of the predicted employee turnover probability are used to aggregate the predicted employee turnover probability.

[0086] Furthermore, the display processing unit 352 reads the customer service skill prediction data 322 from the storage unit 320. The display processing unit 352 aggregates the predicted values ​​of customer service skills for each operator within the display unit and display period. The average value, median value, etc., of the predicted values ​​are used to aggregate the predicted values ​​of customer service skills.

[0087] The display processing unit 352 combines the aggregated predicted values ​​of employee turnover probability and customer service skills. Specifically, based on identification information that identifies the operator, the display processing unit 352 combines the predicted values ​​of employee turnover probability and customer service skills for which the display unit and display period match. The display processing unit 352 generates an output image based on the combined predicted values ​​of employee turnover probability and customer service skills and outputs it to the output unit 340.

[0088] Next, the text data storage process in the telephone terminal device 200 and the information processing device 100 will be described.

[0089] Figure 6 is a flowchart showing an example of processing performed by the telephone terminal device 200 and the information processing device 100 according to this embodiment. In step S11, the call terminal device 200 collects call data on a call-by-call basis and acquires voice data. The voice data is stored in the storage unit 220 as voice data 221, on a call-by-call and utterance-by-utterance basis. Next, the call terminal device 200 executes the process in step S12. Call-based memory refers to remembering a call in a way that allows for the identification of the entire call, from its start to its end. Utterance-based memory refers to remembering a call in a way that allows for the identification of the speaker as either an operator or a customer.

[0090] In step S12, the call terminal device 200 transmits the voice data 221 to the information processing device 100. Then, the call terminal device 200 terminates processing.

[0091] In step S13, the information processing device 100 acquires voice data from the call terminal device 200. Next, the information processing device 100 executes the process in step S14.

[0092] In step S14, the information processing device 100 extracts emotion values ​​for each speaker and text data representing the content of the utterance from the audio data. Next, the information processing device 100 executes the process in step S15.

[0093] In step S15, the information processing device 100 stores the text data and emotion value as text data 121 in the storage unit 120. Then, the information processing device 100 terminates the process. In the flowchart shown in Figure 6, the sound collection in step S11 and the transmission of audio data in step S12 do not necessarily have to be processed in a continuous sequence. In this case, for example, since the audio data 221 is stored in the storage unit 220 in step S11, the processing from step S12 onward can be configured to be performed at regular intervals, such as every hour or every day.

[0094] Next, we will explain the process for predicting employee turnover probability and customer service skills in the information processing device 100.

[0095] Figure 7 is a flowchart showing an example of processing in the information processing device 100 according to this embodiment. In step S21, the information processing device 100 inputs text data 121 containing sentiment values ​​into the first model 124 to predict the predicted value of the probability of each operator leaving the company. Next, the information processing device 100 executes the process in step S22.

[0096] The processing in step S21 can be executed when a certain amount of text data 121 has been accumulated, or at regular intervals (for example, daily, weekly, or monthly). Furthermore, the emotion value entered into the first model 124 in step S21 may be the emotion value between the operator and the customer, the emotion value of the operator only, or the emotion value of the customer only.

[0097] In step S22, the information processing device 100 stores the predicted value of the employee turnover probability as the employee turnover probability prediction value 122 in the storage unit 120. Next, the information processing device 100 executes the process in step S23.

[0098] In step S23, the information processing device 100 inputs text data 121, including emotion values, into the second model 125 to predict the predicted value of the response skill for each call. Next, the information processing device 100 performs the process in step S24. Furthermore, the emotion value entered into the second model 125 in step S23 may be the emotion value between the operator and the customer, the emotion value of the operator only, or the emotion value of the customer only.

[0099] In step S24, the information processing device 100 stores the predicted value of the response skill as the predicted response skill value 123 in the storage unit 120. Then, the information processing device 100 terminates the process shown in Figure 7.

[0100] Furthermore, the information processing device 100 may execute steps S21 and S22 after executing steps S23 and S24. Also, if the information processing device 100 does not output based on predicted values ​​of customer service skills, it does not need to perform steps S23 and S24, and if it does not output based on the probability of employee turnover, it does not need to perform steps S21 and S22.

[0101] Next, the process of step S21 in the information processing device 100 will be described in detail.

[0102] Figure 8 is a flowchart showing an example of information processing in the information processing device 100 according to this embodiment. In step S211, the information processing device 100 obtains text data 121 containing emotion values ​​from the storage unit 120. Next, the information processing device 100 executes the process in step S212.

[0103] In step S211, the information processing device 100 may acquire attribute information related to the call and the operator, such as the date and time of the call, the time of the call, the holding time, the type of call (incoming or outgoing), the operator's affiliation, the operator's years of experience, and the operator's years of service. In this case, for example, the storage unit 120 stores the operator attribute information, and the information processing device 100 can read and acquire said operator attribute information.

[0104] In step S212, the information processing device 100 processes (converts) the text data 121 into a format that can be input into the first model 124. This conversion to a format that can be input into the first model 124 involves, for example, aggregating the text data 121 containing emotion values ​​so that it is organized into units for each operator and for each period of time, based on the utterance units in the text data 121 that contain emotion values. Next, the information processing device 100 executes the process in step S213.

[0105] In step S213, the information processing device 100 inputs the processed text data containing the emotion values ​​into the first model 124. Next, the information processing device 100 executes the process in step S214.

[0106] In step S214, the information processing device 100 obtains predicted values ​​of employee turnover probability for each operator and for each period of time from the first model 124 in response to the input. The information processing device 100 stores these predicted employee turnover probability values ​​as employee turnover probability prediction values ​​122 in the storage unit 120. Then, the information processing device 100 terminates the process shown in Figure 8.

[0107] Next, the process in step S23 of the information processing device 100 will be described in detail.

[0108] Figure 9 is a flowchart showing an example of information processing in the information processing device 100 according to this embodiment. In step S231, the information processing device 100 obtains text data 121 containing emotion values ​​from the storage unit 120. Next, the information processing device 100 executes the process in step S232.

[0109] In step S231, the information processing device 100 may acquire attribute information related to the call and the operator, such as the date and time of the call, the time of the call, the holding time, the operator's department, the operator's years of experience, and the operator's years of service. In this case, for example, the storage unit 120 stores the operator attribute information, and the information processing device 100 can read and acquire said operator attribute information.

[0110] In step S232, the information processing device 100 processes (converts) the text data 121 into a format that can be input into the second model 125. This conversion to a format that can be input into the second model 125 involves, for example, aggregating the text data 121 containing emotion values ​​so that each utterance in the text data 121 containing emotion values ​​becomes a unit for each call. Next, the information processing device 100 executes the process in step S233.

[0111] In step S233, the information processing device 100 inputs the processed text data containing the emotion values ​​into the second model 125. Next, the information processing device 100 executes the process in step S234.

[0112] In step S234, the information processing device 100 obtains a predicted value of the response skill for each call from the second model 125 in response to the input. The information processing device 100 stores the predicted value of the response skill as the predicted response skill value 123 in the storage unit 120. Then, the information processing device 100 terminates the process shown in Figure 9.

[0113] Next, the processing in the information processing device 100 and the operation terminal device 300 will be described.

[0114] Figure 10 is a flowchart showing an example of processing in the information processing device 100 and the operation terminal device 300 according to this embodiment. In step S31, the information processing device 100 receives a data acquisition request from the operating terminal device 300. Next, the information processing device 100 executes the process in step S32.

[0115] In step S32, the information processing device 100 retrieves the predicted employee turnover rate 122 and the predicted customer service skills 123 from the storage unit 120 in response to a request, and transmits the predicted employee turnover rate 122 and the predicted customer service skills 123 to the operation terminal device 300.

[0116] In step S33, the operating terminal device 300 sends a data acquisition request to the information processing device 100. Next, the operating terminal device 300 executes the process in step S34.

[0117] In step S34, the operating terminal device 300 obtains the predicted employee turnover rate 122 and the predicted customer service skill rate 123 from the information processing device 100 as a response to the request. The operating terminal device 300 stores the predicted employee turnover rate 122 and the predicted customer service skill rate 123 in the storage unit 320 as employee turnover rate prediction data 321 and customer service skill prediction data 322. Next, the operating terminal device 300 executes the process in step S35.

[0118] In step S35, the operating terminal device 300 accepts the specification of the display unit and display period based on the user's operation. Next, the operating terminal device 300 executes the process in step S36.

[0119] In step S36, the operating terminal device 300 performs display processing on the employee turnover probability prediction data 321 and the customer service skill prediction data 322. Next, the operating terminal device 300 performs the processing in step S37.

[0120] In step S37, the operating terminal device 300 switches the display data based on the user's input and repeats the processes in steps S35 and S36. Then, the operating terminal device 300 terminates the process shown in Figure 10.

[0121] Note that the information processing device 100 may perform steps S34 and S35 instead of the operating terminal device 300. In this case, the operating terminal device 300 does not need to perform step S33.

[0122] Next, the process of step S36 in the operating terminal device 300 will be explained in detail.

[0123] Figure 11 is a flowchart showing an example of processing in the operation terminal device 300 according to this embodiment. In step S361, the operating terminal device 300 obtains the employee turnover probability prediction data 321 and the customer service skill prediction data 322 from the storage unit 320. Then, the operating terminal device 300 executes the process in step S362.

[0124] In step S362, the operation terminal device 300 aggregates the predicted values ​​of the probability of employee turnover for each operator within the display unit and display period. Next, the operation terminal device 300 executes the process in step S363.

[0125] In step S363, the operating terminal device 300 aggregates predicted values ​​of each operator's response skills for the display unit and display period. Next, the operating terminal device 300 executes the process in step S364.

[0126] In step S364, the operating terminal device 300 combines the predicted values ​​of the turnover probability for each operator in the aggregated display unit and display period with the predicted values ​​of the customer service skills for each operator in the display unit and display period. Then, the operating terminal device 300 completes the process shown in Figure 11.

[0127] Next, we will explain the various types of data stored in the storage unit 120 of the information processing device 100.

[0128] Figure 12 shows an example of text data 121 stored in the storage unit 120 of the information processing device 100 according to this embodiment. The text data shown in the diagram is data that associates the call identification number, start time, end time, speaker, content of speech, emotion value (A), emotion value (B), and emotion value (C). A call identification number is a number that identifies a call. The start time and end time refer to time information that represents the start time and date / time of an utterance, and the end time and date / time of an utterance. The start and end times are obtained, for example, from time information contained in the audio data.

[0129] The speaker is the speaker whose voice has been separated by speaker recognition. Note that if the speakers can be distinguished, such as Speaker A, Speaker B, etc., it is not necessary to distinguish the type of person, such as "operator" or "customer." The operator ID is operator identification information identified by the information identifying the call terminal device 200 and the operator attribute information included in the voice data 221. In this embodiment, it is sufficient to identify at least the operator, so it is not necessary to identify the person in the customer's utterance. The utterance content is text data representing what the speaker said. This text data is generated by speech recognition.

[0130] Emotional Value (A), Emotional Value (B), and Emotional Value (C) are numerical values ​​that represent pre-defined emotions. For example, if Emotional Value (A) represents joy and Emotional Value (B) represents anger, then known technology outputs multiple emotion-corresponding emotional values ​​from the voice data for each utterance. As illustrated, the text data is associated with the content of each speaker's utterance and multiple emotion values ​​for each utterance. The text data 121 may be configured to be stored in the storage unit 120, including other information related to the call contained in the voice data 221.

[0131] Figure 13 shows an example of a predicted employee turnover rate value 122 stored in the storage unit 120 of the information processing device 100 according to this embodiment. The predicted employee turnover rate 122 shown in the diagram is data that associates the target date, operator ID, and turnover rate. The example shown is an example of the predicted employee turnover rate for each operator, aggregated on a daily basis. The target date represents the date for which the data is to be compiled. The operator ID is identification information that identifies the operator. The probability of leaving an employee is the predicted value of the probability of leaving an employee, which was output by inputting the text data 121 into the first model 124.

[0132] Figure 14 shows an example of a response skill prediction value 123 stored in the storage unit 120 of the information processing device 100 according to this embodiment. The illustrated predicted response skill value 123 is data that associates the target date, operator ID, call ID, and response skill. The target date represents the date for which the data is to be compiled. The operator ID is identification information that identifies the operator. The call ID is identification information that identifies a call. The call ID corresponds to the call identification information in Figure 12. The customer service skill is a predicted value of customer service skill output by inputting text data 121 into the second model 125.

[0133] Next, an example of a display screen in the operating terminal device 300 will be described.

[0134] Figure 15 shows an example of the display screen in the operating terminal device 300 according to this embodiment. The display screen shows tabs labeled "Display 1," "Display 2," and "Display 3" as controls for switching display modes. The example shown is an instance where "Display 1" is selected by the user. The display mode for "Display 1" is a sorting mode that displays the predicted turnover probabilities of multiple operators, for example, in descending order of the predicted turnover probability. Additionally, radio buttons are displayed on the screen to specify the display unit as "year," "month," "week," or "day," and the display unit is determined by the user's selection.

[0135] Additionally, a slider bar is displayed on the screen to specify the display period, allowing the user to select the start and end dates to define the display period. The operating terminal device 300 performs display processing based on user input specifying the display period and display unit, and displays the predicted values ​​of the employee turnover probability for each operator in a comparable manner, for example, in descending order of employee turnover probability, as the analysis results of employee turnover probability.

[0136] In the "Display 1" display mode, the aggregated values ​​of the employee turnover probability for each operator ID, generated in the process described using Figure 11, are used. Furthermore, the aggregation of text data 121 including sentiment values ​​in the information processing device 100 must at least match the display period specified above. If the aggregation of text data 121 including sentiment values ​​in the information processing device 100 does not match the display period, the information processing device 100 may reaggregate according to the display period.

[0137] Figure 16 shows another example of the display screen in the operating terminal device 300 according to this embodiment. The display screen shows tabs labeled "Display 1," "Display 2," and "Display 3" as controls for switching display modes. The example shown is an instance where "Display 2" is selected by the user. The "Display 2" display mode is a two-axis display mode in which points identified by the relationship between the predicted turnover probability of multiple operators and the predicted customer service skills are plotted in one of several regions divided by the two axes of the predicted turnover probability and the predicted customer service skills, for example, region A, region B, region C, and region D. Furthermore, the display units and display period are the same as in the example shown in Figure 15. In the illustrated example, each plot represents each operator.

[0138] In the "Display 2" display mode, the predicted values ​​of the probability of employee turnover and the predicted values ​​of customer service skills for each operator ID, which were combined in the process described using Figure 11, are used. In the example shown in Figure 16, the data is divided into four areas—Area A, Area B, Area C, and Area D—based on two axes: predicted probability of employee turnover and predicted customer service skills. Area A is the area where operators requiring special attention are plotted. Managers can use this area to consider whether improvements can be made to the treatment and working environment of operators plotted in Area A: Operators requiring special attention. Area B is the area where the group of operators under consideration for improvement is plotted. Managers can use this to consider whether improvements can be made to the treatment and work environment of the operators plotted in Area B: the group of operators under consideration for improvement. Area C is the area where operators requiring training are plotted. Managers can recognize that operators plotted in Area C: Operators requiring training need improvement in their communication skills. Area D is the area where the group of operators designated as priority maintenance are plotted. Managers can use Area D: Priority Maintenance Operators group to focus on maintaining the current treatment and work environment for the plotted operators, and to gain insights into the treatment and work environment of the plotted operators.

[0139] In this way, the information processing device 100 can provide administrators with clues to consider how to respond to operators in each area through the display mode of display 2, thereby increasing the likelihood of reducing operator turnover. Furthermore, the information processing device 100 can indicate which of the four regions the plotted points are concentrated in. For example, if the plots are concentrated in region B, it may indicate that some kind of action is needed regarding the work environment. Conversely, if the plots are concentrated in region D, it may indicate that the work environment is good.

[0140] In this case, each of the four regions may be colored with a different background color. For example, region B may have a red background to indicate that it is a region requiring attention, while region D may have a blue background to indicate that it is a good region. Furthermore, if you want to select operators who need countermeasures, you can use this to select them, for example, from area B. Alternatively, you can select operators from area A with the expectation of improving their customer service skills in the future. In short, managers can arbitrarily select individuals to receive care based on the relationship between predicted turnover probability and predicted customer service skills. Furthermore, it becomes possible to color-code the plotted points based on the operator's attributes, or to plot individuals who have actually left the company in a specific color.

[0141] Note that while Figure 16 shows four divisions, the divisions are not limited to four; they can be divided into six, nine, or any other number of divisions. Increasing the number of divisions may make the relationship between the predicted employee turnover rate and the predicted customer service skills clearer. Furthermore, the system may be configured to allow users to zoom in on specific areas. This makes it easier for administrators to view the areas they are interested in. Furthermore, the operator's past data may be used to display the movement from past points to the current point using arrows. Alternatively, instead of arrows, points with large fluctuations compared to the past may be displayed as larger points than others.

[0142] Figure 17 shows another example of the display screen in the operating terminal device 300 according to this embodiment. The display screen shows tabs labeled "Display 1," "Display 2," and "Display 3" as controls for switching display modes. The example shown is an instance where "Display 3" is selected by the user. The "Display 3" display mode is a turnover probability prediction trend display mode that shows the temporal changes in the predicted turnover probability values ​​for multiple operators. Furthermore, the display units and display period are the same as in the example shown in Figure 15. The operator ID is identification information that identifies the operator. In the illustrated example, the predicted monthly turnover rate for the operator with operator ID "ID0001" is displayed for each month corresponding to the display unit "month" from "April 2023 to May 2024," which corresponds to the display period "April 2023 to May 2024." This display is shown in multiple stages (e.g., 5 stages) depending on the predicted turnover rate.

[0143] The multi-stage display method may be changed according to the predicted value of the turnover probability and the threshold of the multi-stage method, or according to the relative evaluation of multiple operators, or according to the ratio of each operator's predicted value of the turnover probability to the average value. The display order of operator IDs should preferably be in an order that is noteworthy when looking at changes over time, such as when the turnover probability remains high. By doing this, it is possible to observe the time-series changes in the predicted turnover rate for each operator.

[0144] In the "Display 3" display mode, the predicted values ​​of the probability of employee turnover for each operator ID and the display period, which were combined in the process described using Figure 11, are used. The display shown in Figure 17 allows administrators and other users to be notified that operators with a consistently high probability of leaving require care. Furthermore, for operators whose probability of leaving has changed from high to low, the relationship with the predicted response skill value can be visualized by switching the display mode to Display 2. The upper and lower limits for the multiple stages of employee turnover probability could be, for example, 100% and 0%, or they could be configured to be dynamically set from aggregated values.

[0145] In the examples shown in Figures 15 to 17, filtering and sorting may be performed according to attribute information such as the operator's age and years of experience. Additionally, a flag may be assigned to the identification information of operators who have already left the company, and the trend of the predicted probability of leaving the company before their departure may be displayed according to the flag.

[0146] As described above, the information processing system SYS according to this embodiment includes: an audio data acquisition unit 151 that acquires audio data; an extraction unit (text data extraction unit 152) that extracts a plurality of emotion values ​​representing the speaker's emotions and text data representing the content of the speaker's utterance from the audio data for each speaker; a first prediction unit (employee turnover probability prediction unit 153) that predicts a predicted value of the employee turnover probability by inputting a plurality of emotion values ​​and text data into a first model 124 based on the correspondence between the emotion values ​​and text data and the employee turnover probability; a second prediction unit (employee turnover skill prediction unit 154) that predicts a predicted value of the employee turnover skill by inputting a plurality of emotion values ​​and text data into a second model 125 based on the correspondence between the emotion values ​​and text data and the employee turnover skill; and an output unit (output processing unit 155) that outputs the predicted value of the employee turnover probability and the predicted value of the employee turnover skill.

[0147] In this way, the SYS information processing system can display analysis results based on predicted employee turnover rates for operators and other staff. Furthermore, administrators can review the output and provide support to operators, thereby reducing employee turnover. They can also encourage improvements in customer service skills.

[0148] Furthermore, managers can consider measures to reduce employee turnover from several perspectives by switching to one of the following display modes: Display 1, sorted display mode based on employee turnover probability; Display 2, two-axis display mode, to check the relationship between employee turnover probability and customer service skills; or Display 3, trend display mode for predicted employee turnover probability, to check the temporal changes in employee turnover probability.

[0149] Furthermore, for example, if an administrator notices an operator ID in Display 2, they can select that operator ID using the mouse, and then switch to Display 1 or Display 3 to change the display state so that the relevant operator ID can be identified. Specifically, it is possible to change the display state by enclosing the data of the relevant operator ID in a frame or making it blink. This allows administrators to check the trend of the relevant operator's turnover rate from multiple perspectives across multiple displays.

[0150] Although each embodiment of this invention has been described in detail above with reference to the drawings, the specific configuration is not limited to those described above, and various design changes can be made without departing from the spirit of this invention.

[0151] For example, in the embodiments described above, an example was described in which the system is composed of an information processing device 100, a call terminal device 200, and an operation terminal device 300. However, one aspect of the present invention may be realized by a system that combines some or all of these systems, or by a system in which some of these systems are rearranged.

[0152] Furthermore, the programs that operate in the information processing device 100, the communication terminal device 200, and the operation terminal device 300 in one aspect of the present invention may be one or more programs that control a processor such as a CPU (Central Processing Unit) (a program that makes a computer function) in order to realize the functions shown in the above embodiments and modifications relating to one aspect of the present invention. The term "computer" here includes quantum computers. The information handled by each of these devices may be temporarily stored in RAM (Random Access Memory) during processing, and then stored in various storage devices such as flash memory and HDD (Hard Disk Drive), and may be read, modified, and written by the CPU or the like as needed.

[0153] Furthermore, some or all of the information processing device 100, communication terminal device 200, and operation terminal device 300 in each of the embodiments and modifications described above may be implemented using a computer equipped with one or more processors. In that case, the program for implementing this control function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read by a computer system and executed.

[0154] In this context, "computer system" refers to the computer system built into the information processing device 100, the telephone terminal device 200, and the operating terminal device 300, and includes hardware such as the OS and peripheral devices. Furthermore, "computer-readable recording medium" refers to portable media such as flexible disks, magneto-optical disks, ROMs, and CD-ROMs, as well as storage devices such as hard disks built into computer systems.

[0155] Furthermore, "computer-readable recording media" may include those that dynamically hold programs for a short period of time, such as communication lines used when transmitting programs via networks such as the Internet or communication lines such as telephone lines, as well as those that hold programs for a certain period of time, such as volatile memory inside a computer system that acts as a server or client in such cases. In addition, the above-mentioned program may be for the purpose of realizing some of the functions described above, and may also be a program that can realize the above-mentioned functions in combination with a program already recorded in the computer system.

[0156] Furthermore, some or all of the information processing device 100, call terminal device 200, and operation terminal device 300 in each of the embodiments and modifications described above may be implemented as LSIs, which are typically integrated circuits, or as chipsets. Also, each functional block of the information processing device 100, call terminal device 200, and operation terminal device 300 in each of the embodiments and modifications described above may be individually chipped, or some or all of them may be integrated into a chip. In addition, the method of integrated circuit implementation is not limited to LSIs; it may also be implemented using dedicated circuits and / or general-purpose processors. Furthermore, if an integrated circuit implementation technology that can replace LSIs emerges due to advances in semiconductor technology, it is also possible to use integrated circuits based on that technology.

[0157] Although various embodiments and modifications have been described in detail above with reference to the drawings as one aspect of this invention, the specific configuration is not limited to these embodiments and modifications, and includes design changes and the like that do not depart from the gist of this invention. Furthermore, various modifications are possible within the scope of the claims for one aspect of this invention, and embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of this invention. In addition, configurations in which elements described in the above embodiments and modifications that produce similar effects are substituted for each other are also included. [Explanation of Symbols]

[0158] SYS Information Processing System 100 Information Processing Devices 101 CPU 102 Storage medium interface section 103 Storage medium 104 Input device 105 Output device 106 ROM 107 RAM 108 Auxiliary storage 109 Network Interface Section 110 Communications Department 120 Storage section 121 Text data 122 Predicted probability of leaving the company 123 Predicted Customer Service Skills 124 First Model 125 Second Model 130 Input section 140 Output section 150 Control Unit 151 Audio data acquisition unit 152 Text Data Extraction Unit 153 Employee Turnover Probability Prediction Department 154 Customer Service Skill Prediction Department 155 Output Processing Unit 200 Telephone terminal equipment 210 Communications Department 220 Storage section 221 Audio data 230 Input section 240 Output section 250 Control Unit 251 Sound collection section 300 Operating terminal device 310 Communications Department 320 Storage section 321 Employee Turnover Probability Prediction Data 322 Customer Service Skill Prediction Data 330 Input section 340 Output section 350 Control Unit 351 Data Acquisition Unit 352 Display Processing Unit NW Network

Claims

1. Audio data acquisition unit that acquires audio data, An extraction unit extracts multiple emotion values ​​representing the speaker's emotions and text data representing the content of the utterance made by the speaker from the audio data for each speaker. A first prediction unit predicts a predicted value of the probability of leaving an employee by inputting the multiple emotional values ​​and the text data into a first model based on the correspondence between emotional values ​​and text data and the probability of leaving an employee. A second prediction unit predicts a predicted value of response skills by inputting the multiple emotional values ​​and the text data into a second model based on the correspondence between emotional values ​​and text data and response skills. An output unit that outputs the predicted value of the probability of leaving the company and the predicted value of customer service skills, Equipped with, Information processing system.

2. The aforementioned audio data acquisition unit acquires multiple audio data, The extraction unit extracts, for each of the multiple audio data, a plurality of emotion values ​​representing the speaker's emotions and text data representing the content of the utterance made by the speaker, for each of the speaker. The first prediction unit inputs the multiple emotion values ​​for each of the multiple voice data and the text data into the first model to predict the predicted value of the probability of leaving the company for each of the multiple voice data. The output unit outputs the predicted value of the probability of leaving the company in a way that allows for confirmation of its temporal progression. The information processing system according to claim 1.

3. A display processing unit generates an output image in a two-axis display mode in which points identified by the predicted value of the probability of employee turnover and the predicted value of the customer service skills are plotted in one of a plurality of regions divided by the two axes of the predicted value of the probability of employee turnover and the predicted value of the customer service skills. Furthermore, The output unit outputs the output image. The information processing system according to claim 1.

4. The aforementioned audio data acquisition unit acquires multiple audio data, The extraction unit extracts, for each of the multiple audio data, a plurality of emotion values ​​representing the speaker's emotions and text data representing the content of the utterance made by the speaker, for each of the speaker. The first prediction unit inputs the multiple emotion values ​​for each of the multiple voice data and the text data into the first model to predict the predicted value of the probability of leaving the company for each of the multiple voice data. The output unit outputs the predicted values ​​of the probability of employee departure for each of the plurality of audio data in a comparable manner. The information processing system according to claim 2.

5. Based on user operations, the system generates an output image in a two-axis display mode in which points identified by the predicted value of the employee turnover rate and the predicted value of the customer service skills are plotted in one of several regions divided by the two axes of the predicted value of the employee turnover rate and the predicted value of the customer service skills. An output image representing the temporal change in the predicted value of the probability of leaving the company is generated as a display mode for the predicted value of the probability of leaving the company. A display processing unit that generates the output image so that it can switch between the two-axis display mode and the shift in predicted employee turnover rate display mode. Furthermore, The output unit outputs the output image. The information processing system according to claim 2.

6. Based on user operations, the system generates an output image in a two-axis display mode in which points identified by the predicted value of the employee turnover rate and the predicted value of the customer service skills are plotted in one of several regions divided by the two axes of the predicted value of the employee turnover rate and the predicted value of the customer service skills. An output image representing the temporal change in the predicted value of the probability of leaving the company is generated as a display mode for the predicted value of the probability of leaving the company. The output images, which allow comparison of each of the predicted values ​​of the probability of leaving the company, are generated in a display mode sorted by probability of leaving the company. A display processing unit that generates the output image so that it can switch between the two-axis display mode and the predicted turnover rate trend display mode, Furthermore, The output unit outputs the output image. The information processing system according to claim 4.

7. Audio data acquisition unit that acquires audio data, An extraction unit extracts multiple emotion values ​​representing the speaker's emotions and text data representing the content of the utterance made by the speaker from the audio data for each speaker. A first prediction unit predicts a predicted value of the probability of leaving an employee by inputting the multiple emotional values ​​and the text data into a first model based on the correspondence between emotional values ​​and text data and the probability of leaving an employee. A second prediction unit predicts a predicted value of response skills by inputting the multiple emotional values ​​and the text data into a second model based on the correspondence between emotional values ​​and text data and response skills. An output unit that outputs the predicted value of the probability of leaving the company and the predicted value of customer service skills, Equipped with, Information processing device.

8. A method of information processing performed by a computer, The steps for acquiring audio data and acquiring audio data, An extraction step of extracting multiple emotion values ​​representing the speaker's emotions and text data representing the content of the utterance made by the speaker from the audio data for each speaker, A first prediction step involves inputting the multiple emotional values ​​and the text data into a first model based on the correspondence between emotional values ​​and text data and the probability of leaving the job, thereby predicting a predicted value for the probability of leaving the job. A second prediction step involves inputting the multiple emotional values ​​and the text data into a second model based on the correspondence between emotional values ​​and text data and response skills, thereby predicting the response skill value. An output step that outputs the predicted value of the probability of leaving the company and the predicted value of customer service skills, Having, Information processing methods.

9. On the computer, The steps for acquiring audio data and acquiring audio data, An extraction step of extracting multiple emotion values ​​representing the speaker's emotions and text data representing the content of the utterance made by the speaker from the audio data for each speaker, A first prediction step involves inputting the multiple emotional values ​​and the text data into a first model based on the correspondence between emotional values ​​and text data and the probability of leaving the job, thereby predicting a predicted value for the probability of leaving the job. A second prediction step involves inputting the multiple emotional values ​​and the text data into a second model based on the correspondence between emotional values ​​and text data and response skills, thereby predicting the response skill value. An output step that outputs the predicted value of the probability of leaving the company and the predicted value of customer service skills, A program to execute.

Images