Operator support system, operator support device, and operator support method

Abstract

To use a voice recognition technique and a generative AI to extract registration information of each item of a customer service record from a voice recognition result related to a call voice between a customer and an operator, and appropriately reflect an item extraction result on a registration screen while avoiding manual input of the operator from being wasted due to overwriting when reflecting the item extraction result on the registration screen.SOLUTION: An operator support system instructs a voice recognition section to execute voice recognition processing for converting a call voice into text by voice recognition, instructs a text analysis section to execute item extraction processing for extracting a draft of registration information of each item of a customer service record from a voice recognition result, displays a registration screen 51 on which an operator performs operation of registration regarding each item of the customer service record on an operator terminal, determines a state of manual input of the operator to an input field 52 of each item on the registration screen when acquiring an item extraction result from the text analysis section, and changes a mode of reflecting the item extraction result in the input field according to the state of the manual input.SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] The present invention relates to an operator support system, an operator support device, and an operator support method for assisting the work of an operator in creating a customer response record based on the content of an inquiry made by a customer by phone.

Background Art

[0002] In a call center, when an operator finishes responding to an inquiry call from a customer, the operator creates a customer response record (ticket) in which the content of the customer's inquiry is divided into items according to a predetermined format. Since creating this customer response record is laborious, it is desirable to assist the operator's work and reduce the operator's burden.

[0003] As a technology for assisting such operator work and reducing the operator's burden, conventionally, by using speech recognition technology, the conversation voice between the customer and the operator is converted into text, and customer information is obtained from the speech recognition result, thereby saving the labor of manual input by the operator (see Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] When creating customer service records, a registration screen is displayed on the operator's terminal where the operator enters the information to be registered in each field of the customer service record. At this time, similar to conventional technologies, if the results of speech recognition of the call audio between the customer and the operator can be used, the operator's manual data entry effort can be reduced. Furthermore, it is conceivable to use a generation AI for the item extraction process, which extracts the registration information for each field of the customer service record from the speech recognition results. In this case, once the speech recognition and item extraction processes are complete, the item extraction results should be reflected on the registration screen, allowing the operator to confirm the registered content.

[0006] On the other hand, during a call or while waiting for the speech recognition and item extraction processes to complete, the operator may manually enter information into the registration screen using an input device. For example, proper nouns such as customer names are prone to errors in speech recognition, so the operator may manually enter them. In this case, when the speech recognition and item extraction processes are completed and the item extraction results are reflected on the registration screen, the operator's manual input and the item extraction results will conflict. This can lead to problems such as the operator's manual input being overwritten and wasted, or the item extraction results not being properly reflected on the registration screen.

[0007] Therefore, the main objective of the present invention is to provide an operator support system, operator support device, and operator support method that, when using speech recognition technology and generation AI to extract registration information for each item of a customer service record from the speech recognition results of a call between a customer and an operator, and reflect the item extraction results on the registration screen, can appropriately reflect the item extraction results on the registration screen while avoiding the operator's manual input being wasted due to overwriting. [Means for solving the problem]

[0008] The operator support system of the present invention is an operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records, wherein the processor instructs the speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition result, instructs the text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition result, displays a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction result is obtained from the text analysis unit, it determines the status of the operator's manual input to each input field of the registration screen, and if it is determined that the operator has not yet made a manual input and is in an unentered state, it automatically reflects the item extraction result in the input field, and if it is determined that the operator has completed manual input, it automatically reflects the item extraction result in the input field. Furthermore, if the input is complete and manual input by the operator has been completed, the item extraction results will not be reflected in the input field. If the operator selects an input field, the item extraction results will be displayed as input candidates in a pop-up on the registration screen, and the item extraction results will be reflected in the input field according to the operator's operation to select the item extraction results. This will be the structure.

[0009] Furthermore, the operator support device of the present invention is an operator support device in which a processor performs processing to support the work of an operator who creates customer service records, wherein the processor instructs the speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs the text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, instructs the GUI control unit to display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction results are obtained from the text analysis unit, the processor determines the status of the operator's manual input to each input field of the registration screen, and if it is determined that the operator has not yet made a manual input and is in an unentered state, the item extraction results are automatically reflected in the input field, and if it is determined that the operator has completed manual input, the item extraction results are automatically reflected in the input field. Furthermore, if the input is complete and manual input by the operator has been completed, the item extraction results will not be reflected in the input field. If the operator selects an input field, the item extraction results will be displayed as input candidates in a pop-up on the registration screen, and the item extraction results will be reflected in the input field according to the operator's operation to select the item extraction results. This will be the structure.

[0010] Furthermore, the operator support method of the present invention is an operator support method that causes one or more processors to execute a process to support the work of an operator who creates customer service records, and includes: instructing a speech recognition unit to execute a speech recognition process to convert the voices of the customer and the operator into text by speech recognition and obtaining the speech recognition result; instructing a text analysis unit to execute an item extraction process to extract draft registration information for each item of the customer service record from the speech recognition result; displaying a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record; and, upon obtaining the item extraction result from the text analysis unit, determining the status of the operator's manual input in the input fields of each item on the registration screen, and if it is determined that the operator has not yet made a manual input and is in an unentered state, the item extraction result is automatically reflected in the input field; and if it is determined that the operator has completed the manual input, the item extraction result is automatically reflected in the input field. Furthermore, if the input is complete and manual input by the operator has been completed, the item extraction results will not be reflected in the input field. If the operator selects an input field, the item extraction results will be displayed as input candidates in a pop-up on the registration screen, and the item extraction results will be reflected in the input field according to the operator's operation to select the item extraction results. This will be the structure. [Effects of the Invention]

[0011] According to the present invention, the manner in which the item extraction results are reflected in the input fields is changed according to the state of manual input by the operator in each item input field on the registration screen. Therefore, even when a conflict occurs between the operator's manual input results and the item extraction results, the item extraction results can be appropriately reflected on the registration screen while avoiding the operator's manual input being wasted by being overwritten. [Brief explanation of the drawing]

[0012] [Figure 1] Overall configuration diagram of the operator support system according to this embodiment. [Figure 2] Block diagram showing the schematic configuration of the main control unit and GUI control unit according to this embodiment. [Figure 3] This diagram illustrates the overview of the process for reflecting item extraction results performed in the main control unit according to this embodiment. [Figure 4] This diagram illustrates the registration screen displayed on the operator terminal according to this embodiment. [Figure 5]Explanatory drawing showing the registration screen displayed on the operator terminal according to this embodiment [Figure 6] Explanatory drawing showing an overview of a second example of the input completion determination process performed by the main control unit according to this embodiment [Figure 7] Explanatory drawing showing an example of a prompt and its response in a third example of the input completion determination process performed by the main control unit according to this embodiment [Figure 8] Explanatory drawing showing the registration screen displayed on the operator terminal according to the first modification example [Figure 9] Explanatory drawing showing the registration screen displayed on the operator terminal according to the first modification example [Figure 10] Explanatory drawing showing the registration screen displayed on the operator terminal according to the second modification example [Figure 11] Explanatory drawing showing the transition status of the main part of the registration screen displayed on the operator terminal according to the second modification example [Figure 12] Explanatory drawing showing an example of a table of modification histories stored in the storage unit of the main control unit according to the third modification example [Figure 13] Explanatory drawing showing the registration screen displayed on the operator terminal according to the third modification example

Modes for Carrying Out the Invention

[0013] A first invention made to solve the above problems is an operator support system in which one or more processors execute a process for supporting the work of an operator who creates customer response records. The processor instructs a speech recognition unit to execute a speech recognition process for converting the voices of a customer and an operator into text by speech recognition, obtains a speech recognition result, and from the speech recognition result, instructs a text analysis unit to execute an item extraction process for extracting a draft of registration information for each item of the customer response record. A registration screen for the operator to perform registration operations for each item of the customer response record is displayed on the operator terminal. When the item extraction result is obtained from the text analysis unit, the state of the operator's manual input to the input field for each item on the registration screen is determined. If it is determined that the state is an uninput state where the operator's manual input has not yet been performed, the item extraction result is automatically reflected in the input field. If it is determined that the operator's manual input has been completed, the item extraction result is automatically reflected in the input field. Furthermore, if the input is complete and manual input by the operator has been completed, the item extraction results will not be reflected in the input field. If the operator selects an input field, the item extraction results will be displayed as input candidates in a pop-up on the registration screen, and the item extraction results will be reflected in the input field according to the operator's operation to select the item extraction results. It is configured as follows. The second invention, made to solve the aforementioned problems, is an operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records, wherein the processor instructs the speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs the text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, displays a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and obtains the item extraction results from the text analysis unit, for each item on the registration screen The system determines the status of manual input by the operator in the input field. If it is determined that the field is empty and no manual input has been made by the operator, the item extraction results are automatically reflected in the input field. If it is determined that manual input by the operator has been completed, the item extraction results are not automatically reflected in the input field. Depending on the operator's operation to select the input field, if the caret is located at the beginning of the input field, the item extraction results are displayed as input candidates in a pop-up on the registration screen. If the caret is not located at the beginning of the input field, in addition to the item extraction results, words related to the customer inquiry are displayed as input candidates in a pop-up on the registration screen. The third invention, made to solve the aforementioned problems, is an operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records, wherein the processor instructs the speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs the text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, displays a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction results are obtained from the text analysis unit, the input fields for each item on the registration screen The system determines the status of manual input by the operator. If it determines that the manual input by the operator has not yet been performed and the input is in an unentered state, the item extraction results are automatically reflected in the input field. If it determines that the manual input by the operator has been completed, the item extraction results are not automatically reflected in the input field. Instead, the operator's manual corrections to the item extraction results are stored in the memory as a correction history. In response to the operator's operation to select a phrase from the item extraction results in the input field, the system uses the correction history to determine if the selected phrase matches a phrase that has been previously corrected in relation to the item extraction results. In such cases, the correction results are displayed as input candidates on the registration screen. The fourth invention, made to solve the aforementioned problems, is an operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records, wherein the processor instructs the speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs the text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, displays a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and obtains the item extraction results from the text analysis unit. The system determines the status of manual input by the operator for each item in the input field of the registration screen. If it determines that the item has not yet been manually entered by the operator, the item extraction results are automatically reflected in the input field. If it determines that the manual input by the operator has been completed, the item extraction results are not automatically reflected in the input field. Instead, the correction results made by the operator to the item extraction results are stored in the storage unit as a correction history. Based on the correction history, the system selects items that are prone to errors in the item extraction results and highlights the input field for the item that is prone to errors in the item extraction results on the registration screen. The fifth invention, made to solve the aforementioned problems, is an operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records. The processors instruct a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtain the speech recognition results. The processors instruct a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results. A registration screen is displayed on the operator terminal in which the operator performs registration operations for each item of the customer service record. When the item extraction results are obtained from the text analysis unit, the processors determine the status of the operator's manual input to each input field of the registration screen. If it is determined that the operator has not yet made a manual input and the input is in an unentered state, the item extraction results are automatically reflected in the input field. If it is determined that the operator has completed a manual input, the item extraction results are not automatically reflected in the input field. In addition to determining whether the manual input is unentered or completed, the processors also determine whether the manual input is incorrect. If it is determined to be incorrect, the item extraction results are automatically reflected in the input field. The sixth invention, made to solve the aforementioned problems, is an operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records. The processors instruct a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtain the speech recognition results. The processors instruct a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results. A registration screen is displayed on the operator terminal in which the operator performs registration operations for each item of the customer service record. When the item extraction results are obtained from the text analysis unit, the processors determine the status of the operator's manual input to each input field of the registration screen. If it is determined that the operator has not yet made a manual input and the input is in an unentered state, the item extraction results are automatically reflected in the input fields. If it is determined that the operator has completed manual input, the item extraction results are not automatically reflected in the input fields. If it is determined that the manual input is incorrect, the manually entered content is deleted and the item extraction results are automatically reflected in the input fields.

[0014] According to this, in order to change the mode of reflecting the item extraction result in the input field according to the state of the operator's manual input to the input field for each item on the registration screen, even if a conflict occurs between the operator's manual input result and the item extraction result, it is possible to appropriately reflect the item extraction result on the registration screen while avoiding the operator's manual input being wasted by overwriting.

[0015] Also, a second 7 invention is configured such that when the state is an input incomplete state where the operator's manual input has not been completed, the processor reflects the item extraction result in the input field.

[0016] According to this, the results of item extraction can be appropriately reflected in the input fields according to the status of the operator's manual input (empty, completed, or incomplete).

[0017] Also, the 8 The invention is configured such that when the processor obtains the item extraction result from the text analysis unit, it reflects the item extraction result in the input field in an unconfirmed state, and changes the item extraction result in the input field to a confirmed state in response to the operation of the operator who selects the input field.

[0018] According to this, in order to confirm the item extraction results reflected in the input field, an operator must act to approve the item extraction results, thus ensuring that the operator can reliably verify whether or not the item extraction results are appropriate.

[0019] Also, the 9 The invention is configured such that when the processor obtains the item extraction result from the text analysis unit, it reflects the item extraction result in the input field in an unconfirmed state, displays an operable button for approving the item extraction result on the registration screen, and changes the item extraction result in the input field to a confirmed state in response to the operator's operation on the button.

[0020] According to this, in order to confirm the item extraction results reflected in the input field, an operator must act to approve the item extraction results, thus ensuring that the operator can reliably verify whether or not the item extraction results are appropriate.

[0021] Also, the 10 The invention is In the third invention described above, The processor is configured to display the item extraction results in the input field in different ways depending on whether the item is in an undetermined state or a determined state. Furthermore, the eleventh invention is configured such that, in the fourth invention, the processor displays the item extraction results in the input field in different ways depending on whether the result is in an undetermined state or a determined state.

[0022] According to this, the operator can easily determine whether the item extraction result displayed in the input field is in an unconfirmed state or a confirmed state. In this case, for example, the unconfirmed and confirmed states may be distinguished by changing the text format, such as italics, font size, or font color.

[0035] Furthermore, the twelfth invention is that the processor in the call unit , Gu From the customer Wear When a signal is detected, a new customer response record relating to the inquiry is registered, and the operator terminal is notified accordingly. Upon request from the operator terminal, the registration screen is displayed on the operator terminal in a state that allows the operator to manually input data.

[0036] According to this system, operators can, during a call, enter any text they wish to record as notes from the conversation with the customer into a registration screen as needed. Furthermore, operators can perform both the input of notes during the call and the correction of the extracted items after the call ends on the same screen. This reduces the effort required to switch screens, thereby improving the convenience for operators. 。

[0037] Also, the 13The invention relates to an operator support device in which a processor performs processes to support the work of an operator creating customer service records, wherein the processor instructs a speech recognition unit to perform a speech recognition process to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs a text analysis unit to perform an item extraction process to extract draft registration information for each item of the customer service record from the speech recognition results, instructs a GUI control unit to display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and, upon obtaining the item extraction results from the text analysis unit, determines the status of the operator's manual input in the input fields of each item on the registration screen, and if it is determined that the operator has not yet made a manual input and is in an unentered state, the item extraction results are automatically reflected in the input fields, and if it is determined that the operator has completed manual input, the item extraction results are automatically reflected in the input fields. Furthermore, if the input is complete and manual input by the operator has been completed, the item extraction results will not be reflected in the input field. If the operator selects an input field, the item extraction results will be displayed as input candidates in a pop-up on the registration screen, and the item extraction results will be reflected in the input field according to the operator's operation to select the item extraction results. This will be the structure. Furthermore, the 14th invention is an operator support device in which a processor performs processing to support the work of an operator who creates customer service records, wherein the processor instructs a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, instructs a GUI control unit to display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction results are obtained from the text analysis unit, the processor applies to the input fields of each item on the registration screen The system determines the status of manual input by the operator. If it is determined that the operator has not yet made any manual input, the item extraction results are automatically reflected in the input field. If it is determined that the operator has completed manual input, the item extraction results are not automatically reflected in the input field. Depending on the operator's operation to select the input field, if the caret is located at the beginning of the input field, the item extraction results are displayed as input candidates in a pop-up on the registration screen. If the caret is not located at the beginning of the input field, in addition to the item extraction results, words related to the customer inquiry are displayed as input candidates in a pop-up on the registration screen. Furthermore, the 15th invention is an operator support device in which a processor performs processing to support the work of an operator who creates customer service records, wherein the processor instructs the speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs the text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, instructs the GUI control unit to display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction results are obtained from the text analysis unit, the processor performs operation for each item in the input field of the registration screen The system determines the status of manual input by the user. If it determines that the user has not yet made any manual input, the item extraction results are automatically reflected in the input field. If it determines that the user has completed manual input, the item extraction results are not automatically reflected in the input field. Instead, the user's manual corrections to the item extraction results are stored in the memory as a correction history. In response to the user's operation to select a phrase from the item extraction results in the input field, the system uses the correction history to determine if the selected phrase matches a phrase that has been previously corrected in relation to the item extraction results. In such cases, the correction results are displayed as input candidates on the registration screen. Furthermore, the 16th invention is an operator support device in which a processor performs processing to support the work of an operator who creates customer service records, wherein the processor instructs a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, instructs a GUI control unit to display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction results are obtained from the text analysis unit, the processor performs the registration The system determines the status of manual input by the operator for each item in the input field on the screen. If it is determined that the item has not yet been manually entered by the operator, the item extraction results are automatically reflected in the input field. If it is determined that the manual input by the operator has been completed, the item extraction results are not automatically reflected in the input field. Instead, the correction results made by the operator to the item extraction results are stored in the memory as a correction history. Based on the correction history, the system selects items that are prone to errors in the item extraction results and highlights the input fields related to those items on the registration screen. Furthermore, the 17th invention is an operator support device in which a processor performs processing to support the work of an operator who creates customer service records, wherein the processor instructs a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, instructs a GUI control unit to display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction results are obtained from the text analysis unit, the processor determines the status of the operator's manual input to each input field of the registration screen, and if it is determined that the operator has not yet made a manual input and is in an unentered state, the item extraction results are automatically reflected in the input field, and if it is determined that the operator has completed the manual input, the item extraction results are not automatically reflected in the input field, and in addition to determining whether the manual input is unentered or completed, the processor also determines whether the manual input is incorrect, and if it is determined to be incorrect, the item extraction results are automatically reflected in the input field. Furthermore, the 18th invention is an operator support device in which a processor performs processing to support the work of an operator who creates customer service records, wherein the processor instructs a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtains the speech recognition results, instructs a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, instructs a GUI control unit to display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when it obtains the item extraction results from the text analysis unit, it determines the status of the operator's manual input to each input field of the registration screen, and if it is determined that the operator has not yet made a manual input and is in an uninputted state, it automatically reflects the item extraction results in the input field, if it is determined that the operator has completed the manual input, it does not automatically reflect the item extraction results in the input field, and if it is determined that the manual input is incorrect, it erases the manually entered content and then automatically reflects the item extraction results in the input field.

[0038] According to this, similar to the first invention, the manner in which the item extraction results are reflected in the input fields is changed according to the state of manual input by the operator for each item on the registration screen. Therefore, even when a conflict occurs between the operator's manual input and the item extraction results, the item extraction results can be appropriately reflected on the registration screen while avoiding the operator's manual input being wasted by being overwritten.

[0039] Also, the 19 The invention relates to an operator support method that causes one or more processors to perform processing to support the work of an operator who creates customer interaction records, The one or more processors described aboveThe system instructs the speech recognition unit to perform a speech recognition process to convert the voices of the customer and operator into text using speech recognition, obtains the speech recognition results, instructs the text analysis unit to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results, displays a registration screen on the operator's terminal for the operator to perform registration operations for each item of the customer interaction record, obtains the item extraction results from the text analysis unit, determines the status of the operator's manual input in the input fields for each item on the registration screen, and if it is determined that the operator has not yet made a manual input and the input field is empty, the item extraction results are automatically reflected in the input field; if it is determined that the operator has completed the manual input, the item extraction results are automatically reflected in the input field. Furthermore, if the input is complete and manual input by the operator has been completed, the item extraction results will not be reflected in the input field. If the operator selects an input field, the item extraction results will be displayed as input candidates in a pop-up on the registration screen, and the item extraction results will be reflected in the input field according to the operator's operation to select the item extraction results. This will be the structure. Furthermore, the 20th invention is an operator support method in which one or more processors perform processing to support the work of an operator who creates customer service records, wherein the one or more processors instruct a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtain the speech recognition results, instruct a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction results are obtained from the text analysis unit, the input fields for each item on the registration screen The system determines the status of manual input by the operator. If it determines that the manual input by the operator has not yet been performed and the input is in an empty state, the item extraction results are automatically reflected in the input field. If it determines that the manual input by the operator has been completed, the item extraction results are not automatically reflected in the input field. Depending on the operator's operation to select the input field, if the caret is located at the beginning of the input field, the item extraction results are displayed as input candidates in a pop-up on the registration screen. If the caret is not located at the beginning of the input field, in addition to the item extraction results, words related to the customer inquiry are displayed as input candidates in a pop-up on the registration screen. Furthermore, the 21st invention is an operator support method in which one or more processors perform processing to support the work of an operator who creates customer service records, wherein the one or more processors instruct a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtain the speech recognition results, instruct a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction results are obtained from the text analysis unit, the input fields for each item on the registration screen The system determines the status of manual input by the operator. If it determines that the system is in an unentered state where no manual input has yet been made by the operator, the item extraction results are automatically reflected in the input field. If it determines that manual input by the operator has been completed, the item extraction results are not automatically reflected in the input field. Instead, the operator's manual modifications to the item extraction results are stored in the memory as a modification history. In response to the operator's operation to select a phrase from the item extraction results in the input field, the system uses the modification history to determine if the selected phrase matches a phrase that has been modified in the past regarding the item extraction results. In such cases, the modification result is displayed as an input candidate on the registration screen. Furthermore, the 22nd invention is an operator support method in which one or more processors perform processing to support the work of an operator who creates customer service records, wherein the one or more processors instruct a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtain the speech recognition results, instruct a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and obtain the item extraction results from the text analysis unit, The system determines the status of manual input by the operator for each item in the input field on the registration screen. If it is determined that the item has not yet been manually entered by the operator, the item extraction results are automatically reflected in the input field. If it is determined that the manual input by the operator has been completed, the item extraction results are not automatically reflected in the input field. Instead, the correction results made by the operator to the item extraction results are stored in the memory as a correction history. Based on the correction history, the system selects items that are prone to errors in the item extraction results and highlights the input field for the item that is prone to errors in the item extraction results on the registration screen. Furthermore, the 23rd invention is an operator support method in which one or more processors perform processing to support the work of an operator who creates customer service records, wherein the one or more processors instruct a speech recognition unit to perform speech recognition processing to convert the voices of the customer and the operator into text by speech recognition and obtain the speech recognition results, instruct a text analysis unit to perform item extraction processing to extract draft registration information for each item of the customer service record from the speech recognition results, display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction results are obtained from the text analysis unit, the status of the operator's manual input to each input field of the registration screen is determined, and if it is determined that the operator has not yet made a manual input and is in an unentered state, the item extraction results are automatically reflected in the input field, and if it is determined that the operator has completed a manual input, the item extraction results are not automatically reflected in the input field, and in addition to determining whether the manual input is unentered or completed, it is determined whether the manual input is incorrect, and if it is determined to be incorrect, the item extraction results are automatically reflected in the input field. Furthermore, the 24th invention is an operator support method that causes one or more processors to execute a process to support the work of an operator who creates customer service records, wherein the one or more processors instruct a speech recognition unit to execute a speech recognition process to convert the voices of the customer and the operator into text by speech recognition and obtain the speech recognition result, instruct a text analysis unit to execute an item extraction process to extract a draft of the registration information for each item of the customer service record from the speech recognition result, display a registration screen on the operator terminal in which the operator performs registration operations for each item of the customer service record, and when the item extraction result is obtained from the text analysis unit, the status of the operator's manual input to each input field of the registration screen is determined, and if it is determined that the operator has not yet made a manual input and is in an uninputted state, the item extraction result is automatically reflected in the input field, if it is determined that the operator has completed the manual input, the item extraction result is not automatically reflected in the input field, and if it is determined that the manual input is an incorrect input, the manually entered content is deleted and then the item extraction result is automatically reflected in the input field.

[0040] According to this, similar to the first invention, the manner in which the item extraction results are reflected in the input fields is changed according to the state of manual input by the operator for each item on the registration screen. Therefore, even when a conflict occurs between the operator's manual input and the item extraction results, the item extraction results can be appropriately reflected on the registration screen while avoiding the operator's manual input being wasted by being overwritten.

[0041] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0042] Figure 1 is an overall configuration diagram of the operator support system according to this embodiment.

[0043] The operator support system assists operators in call centers in creating customer service records (tickets) based on the content of customer inquiries made over the phone. These customer service records document the customer's inquiry, categorized into items according to a predetermined format.

[0044] The operator support system comprises a call unit 1, a voice interception unit 2, a main control unit 3 (operator support device), a voice recognition unit 4, a text analysis unit 5, a management unit 6, a GUI control unit 7, and an operator terminal 8.

[0045] The call unit 1 relays the transmission and reception of voice between the customer's telephone 9 and the operator's terminal 8, enabling communication between the customer and the operator. The call unit 1 may be implemented using a publicly known cloud service, or it may be implemented using a telephone system that builds "PBX (main unit)" functionality on the cloud and utilizes business phone functionality via an internet connection.

[0046] The voice interception unit 2 intercepts (collects) the audio of conversations between customers and operators that take place in the call unit 1. The voice interception unit 2 may be implemented using a known cloud service.

[0047] The speech recognition unit 4 converts the audio of the call acquired by the audio interception unit 2 into text using speech recognition. The speech recognition unit 4 may be implemented using a known cloud service.

[0048] The text analysis unit 5 performs item extraction processing on the speech recognition results output from the speech recognition unit 4. In the item extraction processing, a generation AI is used to extract strings from the speech recognition results that will serve as drafts (original versions) of the registration information for each item in the customer interaction record. The text analysis unit 5 also performs summary text generation processing on the speech recognition results. In the summary text generation processing, a generation AI is used to generate a summary text of the inquiry content. The text analysis unit 5 may be implemented using cloud services such as Google Bard®, Microsoft Bing AI®, or ChatGPT®.

[0049] Management Unit 6 manages various types of data, such as customer interaction records, using a database. Management Unit 6 may be implemented using, for example, a CRM tool that manages customer contact history or a general relational database. Alternatively, it may be implemented using a publicly known incident management tool or a cloud service that includes web server functionality.

[0050] The main control unit 3 controls each part of the operator support system to generate registration information for each item of the customer interaction record from the audio of the call between the customer and the operator, and performs the necessary processing to register that registration information in the database of the management unit 6.

[0051] The GUI control unit 7 generates display information (e.g., HTML data) for various screens to be displayed on the operator terminal 8 and displays these screens on the operator terminal 8. The GUI control unit 7 also acquires operation information corresponding to the operator's operations performed on the operator terminal 8. The GUI control unit 7 may be implemented as a software package including web server functionality, such as a CRM tool or incident management tool, or as a cloud service including web server functionality.

[0052] The operator terminal 8 consists of a PC, smartphone, tablet, or the like. The operator terminal 8 includes a display device 11 such as an LCD panel, input devices 12 such as a touch panel, keyboard, and mouse, a microphone 13, and a speaker 14. The microphone 13 and speaker 14 may be a headset.

[0053] The operator terminal 8 allows the operator to make calls with customers using the microphone 13 and speaker 14. The operator terminal 8 also displays various screens on the display device 11 based on display information transmitted from the GUI control unit 7, using an application such as a browser. Furthermore, the operator terminal 8 detects screen operations performed by the operator using the input device 12 and transmits the operation information to the GUI control unit 7.

[0054] Next, the schematic configuration of the main control unit 3 and the GUI control unit 7 according to this embodiment will be described. Figure 2 is a block diagram showing the schematic configuration of the main control unit 3 and the GUI control unit 7.

[0055] The main control unit 3 comprises a communication unit 21, a storage unit 22, and a processor 23.

[0056] The communication unit 21 communicates with the voice interception unit 2, the voice recognition unit 4, the text analysis unit 5, the management unit 6, and the GUI control unit 7.

[0057] The memory unit 22 stores programs and other data executed by the processor 23.

[0058] The processor 23 performs various processes by executing programs stored in the memory unit 22. In this embodiment, the processor 23 performs processes such as speech recognition control, item extraction control, input status determination, incorrect input determination, item extraction result reflection, correction history management, selection of frequently error items, and screen control.

[0059] In the speech recognition control process, the processor 23 instructs the speech recognition unit 4 to perform speech recognition processing to convert the voice of the call between the customer and the operator into text using speech recognition, and obtains the speech recognition result.

[0060] In the item extraction control process, the processor 23 instructs the text analysis unit 5 to execute an item extraction process that uses a generation AI to extract drafts of the registration information for each item of the customer service record based on the speech recognition results, and obtains the item extraction results.

[0061] In the input status determination process, the processor 23 determines the status of manual input performed by the operator using the input device 12 for each input field 52 of the registration screen 51 (see Figures 4 and 5). Specifically, first, it is determined whether manual input has already been performed by the operator. If manual input has already been performed by the operator, then it is determined whether manual input by the operator has been completed (input completion determination process). This allows for the determination of whether manual input has not been performed by the operator (not entered state), whether manual input by the operator has not been completed (input incomplete state), and whether manual input by the operator has been completed (input completed state).

[0062] In the error detection process, processor 23 determines whether the manual input by the operator was made incorrectly. Specifically, if the characters entered by the operator are neither words nor sentences and are meaningless, the manual input by the operator is determined to be an error.

[0063] In the process of reflecting the item extraction results, the processor 23 reflects the item extraction results output from the text analysis unit 5 onto the registration screen 51 (see Figures 4 and 5) displayed on the operator terminal 8. In particular, the manner in which the item extraction results are reflected on the registration screen 51 is changed depending on the status of manual input by the operator.

[0064] In the revision history management process, the processor 23 stores in the storage unit 22 as revision history instances where the item extraction results reflected on the registration screen 51 (see Figures 4 and 5) have been modified by manual input from an operator. The revision history may also be stored in the management unit 6.

[0065] In the process of selecting frequently error-prone items, the processor 23 selects items that are prone to errors in the item extraction results based on the correction history. Specifically, items that include phrases that are prone to errors in speech recognition, such as proper nouns like the names of customer representatives, are selected as items prone to errors in the item extraction results. Items prone to errors in the item extraction results are highlighted on the registration screen 61 by the GUI control unit 7.

[0066] In the screen control process, the processor 23 instructs the GUI control unit 7 to display a registration screen 51 (see Figures 4 and 5) on the operator terminal 8, where the operator performs registration operations for each item of the customer interaction record. A screen showing the operator the speech recognition results of the call audio during a call, or a screen allowing the operator to confirm the speech recognition results after the call ends, may also be displayed. Furthermore, the system may transition to the registration screen 61 after the operator confirms the speech recognition results and instructs the execution of the item extraction process.

[0067] The GUI control unit 7 comprises a communication unit 31, a storage unit 32, and a processor 33.

[0068] The communication unit 31 communicates with the main control unit 3.

[0069] The memory unit 32 stores programs and other data that are executed by the processor 33.

[0070] The processor 33 performs various processes by executing programs stored in the memory unit 32. In this embodiment, the processor 33 performs processes such as screen information generation and operation information acquisition.

[0071] In the screen information generation process, the processor 33 generates display information for the screen to be displayed on the operator terminal 8. In this embodiment, display information for the registration screen 51 (see Figures 4 and 5) is generated.

[0072] In the process of acquiring operation information, the processor 33 acquires operation information generated by detecting the operator's actions at the operator terminal 8.

[0073] Next, the process of reflecting the item extraction results performed by the main control unit 3 according to this embodiment will be described. Figure 3 is an explanatory diagram showing an overview of the process of reflecting the item extraction results.

[0074] In this embodiment, for each input field 52 of the registration screen 51 (see Figures 4 and 5), the status of manual input performed by the operator using the input device 12 is determined as either not entered, incomplete, or completed. In addition, in this embodiment, an incorrect input state is determined for each input field 52 of the item.

[0075] In the empty state, that is, when no manual input has been made by an operator, there are no manually entered strings (manual input results) in the input field 52, and the item extraction results output from the text analysis unit 5 are automatically reflected in the input field 52.

[0076] In the case of an incomplete input state, that is, when manual input by an operator has already been performed but is not yet complete, the manual input result in input field 52 remains as is, and the item extraction result is automatically reflected by being appended after the manual input result. In this case, since operator confirmation is required, both the manual input result and the item extraction result are displayed in input field 52, allowing the operator to make corrections by manual input.

[0077] When the input is complete, that is, when the operator has finished manual input, the manual input result in input field 52 remains as is, and the item extraction results are not automatically reflected. In this case, there is no need to reflect the item extraction results, and operator confirmation is not required, so the item extraction results are not reflected.

[0078] In the event of an incorrect input state, that is, when an operator has made a mistake in manual input, the item extraction result is automatically reflected in input field 52, and the manual input result in input field 52 is overwritten and erased. In this case, there is no need to retain the manual input result, and operator confirmation is not required, so it is forcibly overwritten and erased.

[0079] Next, the registration screen 51 displayed on the operator terminal 8 according to this embodiment will be described. Figures 4 and 5 are explanatory diagrams showing the registration screen 51.

[0080] The example shown in Figure 4(A) illustrates the initial state of the registration screen 51. The registration screen 51 has input fields 52 for each of the following items: customer number, customer name, person in charge, reception category, causative equipment, inquiry details, fault details, and lamp status. In the input field 52 for each item, a string of characters that will be used as registration information for each item in the customer service record is entered.

[0081] The example shown in Figure 4(B) illustrates a case where the operator manually enters data before the text analysis unit 5 outputs the item extraction results.

[0082] In the input fields 52 for the items of person in charge, inquiry details, and problem details, characters are entered manually by the operator using the input device 12. In the input field 52 for the person in charge item, the operator enters the name of the person in charge based on what the customer said. In the input field 52 for the inquiry details item, the operator started to enter the information, but was unable to immediately enter the correct content, resulting in incomplete characters. In the input field 52 for the problem details item, meaningless characters are entered due to an error by the operator.

[0083] Furthermore, on the registration screen 51, input fields 52 for items where errors are likely to occur in the item extraction results are highlighted. In the example shown in Figure 4(B), a warning mark 54 is displayed in the input field 52 for the person in charge. This allows the operator to recognize that errors are likely to occur in the person in charge item in the item extraction results, and to enter the person in charge's name based on the customer's utterance without waiting for the item extraction results to be reflected. Also, on the registration screen 51, a thick border is displayed around input fields 52 that are selected by the operator.

[0084] The example shown in Figure 5(A) shows the state after manual input by the operator, when the item extraction results are output from the text analysis unit 5 and reflected on the registration screen 51.

[0085] In the input fields 52 for customer number and customer name, the customer number and customer name determined from the customer's telephone number are entered. The main control unit 3 determines the customer number and customer name based on the telephone number by referring to the customer database when a customer inquiry call is received.

[0086] In the input field 52 for the person in charge, manual input by the operator has been completed (input completed state). In this case, even if the item extraction results are output from the text analysis unit 5, the item extraction results are not reflected on the registration screen 51, and the state in which the manually entered string (manual input result) is displayed in input field 52 is maintained.

[0087] In the input fields 52 for the reception category, causative device, and lamp status, manual input by the operator has not yet been performed (they are empty). In this case, the item extraction results are reflected on the registration screen 51, and the string representing the item extraction results is displayed in the input field 52.

[0088] In the inquiry details input field 52, a string has been manually entered by the operator, but because this string does not match the attribute of the item, the manual entry by the operator is judged to be incomplete (input incomplete state). Specifically, the inquiry details field requires a string to be entered as a sentence, but only the word "signage" has been entered. In this case, since it is expected that the operator will edit the string, the item extraction result is reflected on the registration screen 51 while leaving the string manually entered by the operator (manual entry result). Specifically, in input field 52, the string from the item extraction result is added after the string from the manual entry result.

[0089] In the input field 52 for the error details, characters have already been entered manually by the operator, but since these characters are neither a word nor a sentence and are meaningless, the operator's manual entry is judged as an incorrect entry (incorrect entry state). Specifically, the meaningless character "あ" has been entered in input field 52 (see Figure 4(B)). In this case, the characters entered manually by the operator (manual entry result) are discarded, and the item extraction result is reflected on the registration screen 51. Specifically, in input field 52, the string from the item extraction result is reflected in a state where it overwrites the characters from the manual entry result.

[0090] Furthermore, in the input field 52 for each item, the displayed string is shown in a different manner depending on whether it is in an unconfirmed state or a confirmed state.

[0091] In the example shown in Figure 5(A), the input fields 52 for customer number and customer name display the text in the standard font because the text is already finalized and does not require modification. Similarly, the input field 52 for the person in charge also displays the text in the standard font because the text is already finalized and does not require modification.

[0092] On the other hand, in the input fields 52 for each item of reception category, causative equipment, fault details, and lamp status, the string of the item extraction result is in an undetermined state, and the string is displayed in a manner different from the standard in order to prompt the operator to check whether there are any errors in the item extraction result. In the example shown in Figure 5(A), the string of the item extraction result is displayed in input field 52 in a gray, italicized font. Alternatively, for example, the string may be displayed in a font smaller than the standard.

[0093] Furthermore, in the input field 52 for the inquiry details, the string entered manually by the operator (manual input result) is displayed in a confirmed state, while the string from the item extraction result is displayed in an unconfirmed state. Specifically, the string from the manual input result is displayed in a standard font, while the string from the item extraction result is displayed in a gray, italicized font.

[0094] The example shown in Figure 5(B) illustrates a case where the item extraction results output from the text analysis unit 5 are reflected on the registration screen 51, and then the string is modified by the operator's manual input.

[0095] In the input field 52 for the inquiry details, as shown in Figure 5(A), when the item extraction results were reflected in the input field 52, the string of the item extraction results was added after the string of the manually entered result. However, in the example shown in Figure 5(B), the string displayed in the input field 52 has been modified by the operator. Specifically, the word "signage" has been deleted by the operator. Also, in the input field 52 for the inquiry details, because the string was modified by the operator's manual input, the string displayed in the input field 52 changed to a confirmed state, and the modified string is displayed in the standard font.

[0096] Furthermore, in this embodiment, when the operator performs an operation to select an input field 52 for each item (for example, a tap operation or a click operation), the unconfirmed string displayed in the input field 52 changes to a confirmed state. In the example shown in Figure 5(B), the strings in the input fields 52 for each item—reception category, causative equipment, fault details, and lamp status—are displayed in an unconfirmed state. However, when the operator performs an operation to select an input field 52 for each item, the strings displayed in the input field 52 change to a confirmed state. This ensures that the operator can reliably confirm whether or not there are any errors in the item extraction results.

[0097] Furthermore, the registration screen 51 is provided with a registration button 53. When an operator operates the registration button 53, a registration process is performed in which the strings displayed in the input fields 52 for each item are registered in the database of the management unit 6 as registration information for each item of the customer service record. At this time, the autocorrect function may be used to correct typographical errors. Also, if the registration button 53 is operated while the strings displayed in the input fields 52 are still in an unconfirmed state, a message prompting the operator to confirm the strings in the input fields 52 may be displayed on the screen.

[0098] In this embodiment, the registration screen 51 is displayed on the operator terminal 8, but a registration confirmation screen may be displayed after the registration screen 51 to allow the operator to confirm the registration details. That is, when the operator operates the registration button 53 on the registration screen 51, the system transitions to the registration confirmation screen, and when the operator approves the registration details on the registration confirmation screen, the registration process may be performed.

[0099] Incidentally, when the call unit 1 detects an incoming call from a customer making an inquiry, a contact ID is assigned to the call, and the contact ID, the customer's phone number, and the operator ID are registered in the database of the management unit 6. At this time, a new customer interaction record (ticket) may also be registered.

[0100] In this case, fields other than Contact ID, Phone Number, and Operator ID may be left blank. Alternatively, fields related to Contact ID, Phone Number, and Operator ID may be registered. For example, if there is a separate database for managing customer information, the customer's phone number will be associated with the customer number and store name. Similarly, if there is a separate database for managing operators, the operator ID will be associated with the operator's name. In other words, using registration screen 51 (see Figure 4) as an example, fields related to Contact ID, Phone Number, and Operator ID would include the customer number, customer name (store name), and person in charge (operator's name).

[0101] Furthermore, when a new customer interaction record (ticket) is registered in the management unit 6, the operator may be notified accordingly. Specifically, when the main control unit 3 receives a response from the management unit 6 indicating that the new customer interaction record has been successfully registered, it may request external software to send a notification that a new customer interaction record has been registered to the operator's contact information (operator terminal 8) associated with the operator ID. Examples of such external software include email and communication tools with business chat functionality.

[0102] Furthermore, the notification that a new customer interaction record has been registered may include access information to display the customer interaction record viewing screen on the operator terminal 8. An example of access information is a URL. The operator can view the contents of the newly registered customer interaction record by performing an operation to open the customer interaction record viewing screen on the operator terminal 8.

[0103] Furthermore, operators can manually input text using the input device 12 on the customer interaction record viewing screen. This allows operators to input text as needed during a call to record any information they wish to remember from the conversation with the customer.

[0104] Furthermore, the screen for viewing customer interaction records can be, for example, the registration screen 51 (see Figure 4(A)). In this case, the registration screen 51 is displayed when the call begins. On the registration screen 51, the operator can enter text of content they want to note down in the input fields 52 of the relevant items. Also, on the registration screen 51, the item extraction results are reflected after the call ends. At this time, the operator can correct the item extraction results while checking the text they entered as notes. In this way, the operator can enter notes during the call and correct the item extraction results after the call ends on the same screen. This reduces the effort required to switch screens, thus improving the convenience for the operator.

[0105] Next, a first example of the input completion determination process performed by the main control unit 3 will be described. In the input completion determination process, if manual input by the operator has already been performed, it is determined whether or not the manual input by the operator has been completed, that is, whether the input is in an incomplete state or an input completed state.

[0106] In this example, each item in the customer interaction record is assigned a predetermined attribute, and the input completion determination process is performed based on whether the string manually entered by the operator in the input field 52 of the registration screen 51 (manual input result) matches the attribute of each item (matching determination process).

[0107] First, for a specific item, if the string to be registered exceeds a predetermined minimum number of characters, the minimum number of characters is set as an attribute of that item. In the conformance determination process, the conformance of the manually entered result is determined based on whether or not the manually entered result exceeds the minimum number of characters. For example, in the "Inquiry Details" item, the minimum number of characters is set to 10, and it is determined whether or not the manually entered result is 10 characters or more. If the manually entered result is 10 characters or more, it is determined that the operator's manual entry is complete; if the manually entered result is less than 10 characters, it is determined that the operator's manual entry is incomplete.

[0108] Furthermore, for specific items, if the registered string is represented in a predetermined pattern of character combinations, a character combination pattern (notation rule) is set as an attribute of that item. In the conformance determination process, it is determined whether the manually entered result conforms to the character combination pattern. For example, in the "Model Number" item, if the model number is set based on the notation rule "two uppercase alphabet letters + a number", a regular expression pattern is set as follows, and it is determined whether the manually entered result conforms to this predetermined regular expression pattern. "Patent=[AZ][AZ]\d * " Here, if the manual input result conforms to a predetermined pattern, the manual input by the operator is determined to be complete; if the manual input result does not conform to a predetermined pattern, the manual input by the operator is determined to be incomplete.

[0109] Furthermore, for specific items, if the registered string contains a predetermined type of character, the inclusion of that predetermined type of character is set as an attribute of that item. In the conformance determination process, it is determined whether or not the predetermined type of character is included in the manually entered result. For example, for the "lamp status" item, it is determined whether or not the manually entered result contains a character representing a color. If the manually entered result contains a character representing a color, it is determined that the operator's manual entry is complete; if the manual entry result does not contain a character representing a color, it is determined that the operator's manual entry is incomplete.

[0110] Next, a second example of the input completion determination process performed by the main control unit 3 according to this embodiment will be described. Figure 6 is an explanatory diagram showing an overview of the second example of the input completion determination process.

[0111] In this example, the input completion determination process is performed based on the similarity between the string manually entered by the operator in the input field 52 of the registration screen 51 (manual input result) and the string obtained by the generating AI in the item extraction process (item extraction result) (similarity determination process). That is, if the manual input result is similar to the item extraction result, the manual input by the operator is determined to be complete; if the manual input result is not similar to the item extraction result, the manual input by the operator is determined to be incomplete.

[0112] In the similarity determination process, for example, first, the manually entered results and item extraction results are decomposed into words by morphological analysis. Next, vectorization is performed for each of the manually entered results and item extraction results based on the word occurrences. Here, for example, words that appear in the text are assigned a value of 1, and words that do not appear in the text are assigned a value of 0. The tf-idf (term frequency-inverse document frequency) method may also be used. Next, the cosine similarity is calculated based on the vectorization results for each of the manually entered results and item extraction results.

[0113] Next, by comparing the cosine similarity to a predetermined threshold (e.g., 0.5), it is determined whether the manual input result is similar to the item extraction result, and based on this determination, it is determined whether the operator's manual input is complete. That is, if the cosine similarity is greater than or equal to the threshold, it is determined that the manual input result is similar to the item extraction result, and the operator's manual input is determined to be complete. On the other hand, if the cosine similarity is less than the threshold, it is determined that the manual input result is not similar to the item extraction result, and the operator's manual input is determined to be incomplete.

[0114] Here, if we set the cosine similarity threshold to 0.5, for example, in the example shown in Figure 6, the cosine similarity is 0.408, meaning the similarity between the manual input result and the item extraction result is low, and it is determined that the manual input by the operator is incomplete. In this case, the item extraction result is reflected on the registration screen 51 (see Figure 5). Specifically, the string "The signage screen is not displaying" which is the manual input result, is followed by the string "The screen remains blue" which is the item extraction result, and the string "The signage screen is not displaying, the screen remains blue" is displayed in the input field 52 of the registration screen 51.

[0115] Next, a third example of the input completion determination process performed by the main control unit 3 will be described. Figure 7 is an explanatory diagram showing an example of a prompt to be input to the text analysis unit 5 and its response.

[0116] In this example, the input completion determination process is performed in the text analysis unit 5 using a generation AI. Specifically, in addition to the process of extracting drafts (original versions) of the registration information for each item of the customer service record (item extraction process), the text analysis unit 5 also performs a process to determine whether or not manual input by an operator has been completed for each item of the customer service record (input completion determination process).

[0117] In this case, the main control unit 3 instructs the text analysis unit 5 to execute the input completion determination process. At this time, for example, a prompt (instruction statement) as shown in Figure 7(A) is generated by the main control unit 3, and that prompt is input to the text analysis unit 5, and the response (determination result) shown in Figure 7(B) is output from the text analysis unit 5.

[0118] In the example shown in Figure 7(A), the prompt includes a string manually entered by the operator in the input field 52 of a specific item (in this example, the result summary item) on the registration screen 51 (manual input result), and the speech recognition result of the call between the customer and the operator. Based on the speech recognition result, it is determined whether the manual input result is valid as registration information for the specific item.

[0119] The response (judgment result) output from the text analysis unit 5 can be either "Yes" or "No". If the response is "Yes", it is determined that manual input by the operator is complete. On the other hand, if the response is "No", it is determined that manual input by the operator is incomplete.

[0120] (First variation) Next, a first modified example will be described. Points not specifically mentioned here are the same as in the previously described embodiment. Figures 8 and 9 are explanatory diagrams showing the registration screen 61 displayed on the operator terminal 8 according to the first modified example.

[0121] In this modified example, the registration screen 61 shown in Figures 8 and 9 is displayed on the operator terminal 8. The registration screen 61 is provided with input fields 52 for each item, similar to the embodiment described above. Also in this modified example, similar to the embodiment described above, when the item extraction results are output from the text analysis unit 5, the item extraction results are reflected in the input fields 52 for each item in an undetermined state.

[0122] On the other hand, in this modified version, an assist button 62 is provided on the registration screen 61. The assist button 62 is provided for each item: person in charge, reception category, causative device, inquiry content, fault content, and lamp status. When the operator operates the assist button 62, the item extraction results reflected in the input field 52 in an unconfirmed state change to a confirmed state.

[0123] Furthermore, since the customer number and customer name fields reflect the customer number and customer name determined from the customer's phone number, the assist button 62 is not provided.

[0124] The example shown in Figure 8(A) is the stage before the item extraction results are output from the text analysis unit 5. In this case, since the item extraction results have not yet been output from the text analysis unit 5, the item extraction results cannot be reflected on the registration screen 61, and therefore the assist button 62 is inoperable. In the example shown in Figure 8(A), the assist button 62 is displayed in gray.

[0125] Furthermore, the input fields 52 for customer number and customer name display the customer number and customer name obtained from the customer's phone number. In addition, the input field 52 for the person in charge displays the string manually entered by the operator during the call. The input fields 52 for the other items are left blank.

[0126] The example shown in Figure 8(B) is when the item extraction results are output from the text analysis unit 5. In this case, the item extraction results can be reflected on the registration screen 61, and the assist button 62 changes to an operable state.

[0127] Furthermore, in this embodiment, when the input is complete, that is, when a string has already been entered into the input field 52 by the operator, if the operator selects the input field 52, the string from the item extraction results will be presented as an input candidate.

[0128] In the example shown in Figure 8(B), the operator selects the input field 52 for the "person in charge" item. In this case, a caret 63 is displayed in the input field 52. Near the caret 63, a selection box 64 pops up for the operator to select a string from the item extraction results as an input candidate. When the operator selects a string from the item extraction results in the selection box 64, that string is reflected in the input field 52.

[0129] On the other hand, in the input fields 52 for each item—reception category, causative device, inquiry details, fault details, and lamp status—the item extraction results are displayed in an undetermined state. Specifically, the string of the item extraction result is displayed in input field 52 in a light gray color with italicized text formatting.

[0130] When the operator operates the assist button 62, the string representing the item extraction result, which is displayed in the input field 52 in an unconfirmed state, changes to a confirmed state. This ensures that the operator can reliably confirm whether the item extraction result is appropriate or not.

[0131] The examples shown in Figures 9(A) and (B) illustrate the case where an operator selects input field 52 in order to edit the unconfirmed item extraction result displayed in input field 52. In this example, input field 52 for the inquiry item is selected. In this case, as shown in Figure 9(A), a caret 63 is first displayed at the beginning of input field 52. A selection box 64 also pops up near the caret 63. The selection box 64 displays the strings of the item extraction result as input candidates.

[0132] Note that the selection box 64 may display the entire string of the extracted items, or it may display only a portion of the string, for example, only the beginning of the string.

[0133] The example shown in Figure 9(B) is when an operator moves the caret 63 displayed in input field 52 from the beginning of input field 52. In this case, in addition to the item extraction results, words related to the customer's inquiry are displayed as input candidates in the selection box 64. For example, words related to the word where the caret 63 is located, words extracted from the speech recognition results, words related to the customer, and words related to the type of equipment or system used by the customer are displayed in the selection box 64.

[0134] In this modified example, the string of the item extraction result is displayed as an input candidate in the selection box 64. Therefore, if the operator has already made a manual input and a conflict occurs between the manually entered string and the string of the item extraction result, the operator can select either of them. This prevents the operator's manual input from being overwritten and wasted, while ensuring that the item extraction result is properly reflected in the registration screen 61.

[0135] The selection box 64 may also pop up as a predictive text function, displaying phrases containing characters entered by the operator as conversion candidates. In this case, the string of the item extraction result will be displayed in the selection box 64 along with the predicted conversion candidate phrases.

[0136] Furthermore, in this modified example, since an assist button 62 is provided for each item, the operator can instruct the system to reflect the item extraction results on an item-by-item basis. Alternatively, an assist button 62 corresponding to all items may be provided, and when the operator operates that assist button 62, the item extraction results may be reflected for all items at once. In addition, multiple items may be grouped together, and an assist button 62 may be provided for each group, so that the item extraction results are reflected on a group-by-group basis.

[0137] Furthermore, in this modified example, the display mode of the assist button 62 changes depending on whether or not the item extraction results can be reflected in the input field 52, that is, whether it is in an inoperable state or an operable state. Thus, the assist button 62 functions as an assist status display unit that indicates whether or not it is possible to reflect the item extraction results in the input field 52. On the other hand, an assist status display unit may be provided that has the same form as the assist button 62 but does not have the function of an operation unit for reflecting the item extraction results in the input field 52, and only has the function of indicating whether or not it is possible to reflect the item extraction results in the input field 52.

[0138] Incidentally, in the above embodiment and the first modified example, the registration screen 61 is configured to fit a small, vertically oriented display device 11 when the operator terminal 8 is a smartphone. Furthermore, in this embodiment and the first modified example, the burden on the operator's input operations can be reduced by utilizing AI, so that customer interaction records can be registered even on a small display device 11, such as a smartphone. As a result, the operator terminal 8 lent to operators when they work as a home-based call center can be made cheaper, and the operating costs of the system can be reduced. In addition, when the operator terminal 8 is a smartphone, it becomes easier to use voice input for text input, further reducing the burden on the operator's input operations.

[0139] (Second variation) Next, a second modified example will be described. Points not specifically mentioned here are the same as in the previous embodiment. Figure 10 is an explanatory diagram showing the registration screen 71 displayed on the operator terminal 8 according to the second modified example. Figure 11 is an explanatory diagram showing the transition status of the main parts of the registration screen 71.

[0140] In this modified example, the registration screen 71 shown in Figure 10 is displayed on the operator terminal 8. Similar to the embodiment described above, the registration screen 71 is provided with input fields 52 for each item, but in this modified example, a display field 72 for displaying the item extraction results is provided next to the input fields 52, in addition to the input fields 52.

[0141] The example shown in Figure 10 shows the state after the item extraction results have been output from the text analysis unit 5 and reflected in the display field 72. Before the item extraction results are output, the display fields 72 for each item—person in charge, reception category, causative equipment, inquiry content, fault content, and lamp status—are blank. On the other hand, the customer number and customer name are obtained based on the customer's telephone number, so the customer number and customer name are displayed in the display field 72.

[0142] Furthermore, in the above embodiment, the reflection of the item extraction results in the input field 52 differs depending on the status of manual input by the operator (input completed, input not completed). However, in this modified example, the item extraction results are reflected in the display field 72 for each item regardless of the status of manual input by the operator. In addition, while the operator can manually input strings in the input field 52, the display field 72 only displays the item extraction results, and the operator cannot manually edit the strings displayed in the display field 72.

[0143] In this modified version, the input field 52 and the display field 72 are arranged side by side for each item, and the item extraction results are displayed in the display field 72. Therefore, the operator can manually enter a string into the input field 52 while referring to the item extraction results displayed in the display field 72.

[0144] Furthermore, the registration screen 71 is provided with an assist button 73. By operating the assist button 73, the operator can reflect the item extraction results displayed in the display field 72 into the input field 52. This allows the operator to choose whether or not to reflect the item extraction results displayed in the display field 72 into the input field 52. In the above embodiment and the first modified example, the item extraction results are reflected in the input field 52 in an unconfirmed state, but in this modified example, the item extraction results are reflected in the input field 52 in a confirmed state.

[0145] Here, if, after the operator manually enters an appropriate string into the input field 52 during a call, the item extraction results are displayed in the display field 72, and the item extraction results are more appropriate than the manually entered results, the operator can overwrite the manually entered results in the input field 52 entirely with the item extraction results by operating the assist button 73.

[0146] On the other hand, if there are appropriate words or expressions in part of the item extraction results, the effort of inputting can be reduced by quoting part of the item extraction results. In this modified example, the operator can partially reflect the item extraction results in the input field 52 by selecting part of the string of the item extraction results in the display field 72.

[0147] Specifically, when an operator has a caret displayed in the input field 52, selects a portion of the string from the item extraction results in the display field 72, and then operates the assist button 73, the selected string in the display field 72 is inserted at the caret's position in the input field 52 (insertion operation). Also, when an operator selects a portion of the manually entered string in the input field 52, and simultaneously selects a portion of the string from the item extraction results in the display field 72, and then operates the assist button 73, the selected string in the input field 52 is replaced with the selected string in the display field 72 (replacement operation).

[0148] The example shown in Figure 11 is a replacement operation. As shown in Figure 11(A), the input field 52 displays the string entered manually by the operator. The display field 72 displays the string extracted from the items. As shown in Figure 11(B), the operator selects a portion of the manually entered string in the input field 52 and a portion of the extracted string in the display field 72, and then operates the assist button 73. As a result, as shown in Figure 11(C), the string selected in the input field 52 is replaced with the string selected in the display field 72.

[0149] In this modified example, the item extraction results are partially reflected in the input field 52 in response to the operator's operation of selecting a portion of the item extraction results in the display field 72, thereby saving the operator the trouble of manual input.

[0150] (Third variation) Next, a third modified example will be described. Points not specifically mentioned here are the same as in the previous embodiment. Figure 12 is an explanatory diagram showing an example of a table of modification history stored in the storage unit 22 of the main control unit 3 according to the third modified example. Figure 13 is an explanatory diagram showing the registration screen 61 displayed on the operator terminal 8 according to the third modified example.

[0151] If there is an error in the item extraction result output from the text analysis unit 5, the item extraction result is corrected by the operator. In this modified example, cases in which the item extraction result is corrected by the operator are stored as a correction history in the storage unit 22 of the main control unit 3. Specifically, as shown in Figure 12, the item extraction result that was corrected by the operator and the correction result manually entered by the operator are registered in the correction history table in association. The correction history may also be stored in the management unit 6.

[0152] In this modified example, as shown in Figure 13, in the input field 52 of the registration screen 61, in response to the operator's operation to select a phrase from the item extraction results, if the selected phrase matches a phrase that has been previously modified in relation to the item extraction results, the modified result is displayed as an input candidate on the registration screen 61, based on the modification history.

[0153] The example shown in Figure 13(A) is the state immediately after the item extraction results output from the text analysis unit 5 are reflected on the registration screen 61. The example shown in Figure 13(B) is when the operator determines that there is an error in some of the item extraction results displayed in the input fields 52 for each item and that correction is necessary, and performs the operation to select the input field 52 that needs correction. In the example shown in Figure 13(B), the input field 52 for the inquiry content item is selected. When a selection operation is performed on the input field 52, the input field 52 switches to an editable state, and a caret 63 is displayed in the input field 52.

[0154] When the operator places the caret 63 over a phrase that needs correction, if the phrase where the caret 63 is located is registered in the correction history, that is, if it matches a phrase in the item extraction result that has been corrected by the operator in the past, a selection box 64 will appear near the caret 63. Based on the correction history, the selection box 64 will display as input candidates the phrases that the operator manually entered when they previously corrected the item extraction result.

[0155] Specifically, in the input field 52 for the inquiry details, the phrase "reference item" included in the reflected item extraction results was previously corrected to "3 products," so the phrase "3 products" is displayed as an input suggestion in the selection box 64. When the operator selects the phrase "3 products" in the selection box 64, the phrase "reference item" displayed in the input field 52 is replaced with the phrase "3 products."

[0156] Furthermore, for the item extraction result where caret 63 is located, multiple corrected phrases entered by the operator may be displayed as input candidates in the selection box 64.

[0157] In this modified version, if the same error as in the past occurs in the item extraction results, the operator can correct it simply by selecting an input candidate, thus reducing the burden on the operator.

[0158] In the example shown in Figure 13(B), an input field 52 and an assist button 62 are provided for each item, similar to the first modified example (see Figures 8 and 9). However, this modified example can also be applied to cases where an input field 52, a display field 72, and an assist button 73 are provided for each item, similar to the second modified example (see Figure 10).

[0159] As described above, embodiments have been explained as examples of the technology disclosed in this application. However, the technology in this disclosure is not limited to these embodiments and can be applied to embodiments that have been modified, replaced, added, or omitted. Furthermore, it is possible to create new embodiments by combining the components described in the above embodiments. [Industrial applicability]

[0160] The operator support system, operator support device, and operator support method according to the present invention utilize speech recognition technology and generative AI to extract registration information for each item of customer service records from speech recognition results relating to call audio between a customer and an operator, and reflect the item extraction results on the registration screen. This has the effect of appropriately reflecting the item extraction results on the registration screen while avoiding the waste of the operator's manual input due to overwriting, and is useful as an operator support system, operator support device, and operator support method to support the work of operators who create customer service records based on the content of customer inquiries by telephone. [Explanation of Symbols]

[0161] 1: Call section 2: Audio interception unit 3: Main control unit (operator support device) 4: Voice recognition unit 5: Text Analysis Department 6: Management Department 7: GUI Control Unit 8: Operator terminal 9: Telephone 11: Display device 12: Input Devices 13: Mike 14: Speakers 23: Processor 33: Processor 51: Registration screen 52: Input field 53: Register button 61: Registration screen 62: Assist button 63: Carret 64: Selection Box 71: Registration screen 72:Display field 73: Assist button

Claims

1. In an operator support system in which one or more processors perform processes to assist operators in creating customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. When the input is complete and manual input by the operator has been completed, the item extraction results are not reflected in the input field. When the operator selects the input field, the item extraction results are displayed as input candidates in a pop-up on the registration screen. In response to the operator's action in selecting the item extraction result, the item extraction result is reflected in the input field. An operator support system characterized by the following features.

2. An operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. Depending on the operator's operation to select the input field, if the caret is located at the beginning of the input field, the item extraction result is displayed as an input candidate in a pop-up on the registration screen; if the caret is not located at the beginning of the input field, in addition to the item extraction result, words related to the customer inquiry are displayed as input candidates in a pop-up on the registration screen. An operator support system characterized by the following features.

3. An operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. The correction results made by the operator to the item extraction results are stored in the memory unit as a correction history. In response to the operator's operation of selecting a phrase from the item extraction results in the input field, if the selected phrase matches a phrase that has been previously modified in relation to the item extraction results, based on the modification history, the modification result will be displayed as an input candidate on the registration screen. An operator support system characterized by the following features.

4. An operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. The correction results made by the operator to the item extraction results are stored in the memory unit as a correction history. Based on the aforementioned revision history, select the items that are prone to errors in the item extraction results, In the registration screen, the input fields for items that are prone to errors in the item extraction results are highlighted. An operator support system characterized by the following features.

5. An operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. In addition to determining whether the manual input is incomplete or completed, the system also determines whether the manual input is incorrect, and if it is determined to be incorrect, it automatically reflects the item extraction result in the input field. An operator support system characterized by the following features.

6. An operator support system in which one or more processors perform processing to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. If the manual input is determined to be an error, the manually entered content is deleted, and the item extraction result is automatically reflected in the input field. An operator support system characterized by the following features.

7. The aforementioned processor, The operator support system according to claim 1, characterized in that, if the input is incomplete and manual input by the operator has not been completed, the item extraction results are reflected in the input field.

8. The aforementioned processor, When the item extraction results are obtained from the text analysis unit, the item extraction results are reflected in the input field in an unconfirmed state. The operator support system according to claim 1, characterized in that it changes the item extraction result of the input field to a confirmed state in response to the operation of the operator who selects the input field.

9. The aforementioned processor, When the item extraction results are obtained from the text analysis unit, the item extraction results are reflected in the input field in an unconfirmed state, and a button to approve the item extraction results is displayed on the registration screen in an operable state. The operator support system according to claim 1, characterized in that the result of extracting the items in the input field is changed to a confirmed state in response to the operator's operation on the button.

10. The aforementioned processor, The operator support system according to claim 3, characterized in that the item extraction results are displayed in the input field in different ways depending on whether the state is undetermined or confirmed.

11. The processor is The operator support system according to claim 4, characterized in that the item extraction results are displayed in the input field in different ways depending on whether the state is undetermined or confirmed.

12. The aforementioned processor, When the call unit detects an incoming call from a customer, it registers a new customer interaction record related to that inquiry and notifies the operator terminal accordingly. The operator support system according to claim 1, characterized in that, in response to a request from the operator terminal, the registration screen is displayed on the operator terminal in a state that allows the operator to manually input data.

13. An operator support device in which a processor performs processes to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. The GUI control unit is instructed to display a registration screen on the operator's terminal, in which the operator performs the registration operation for each item of the customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. When the input is complete and manual input by the operator has been completed, the item extraction results are not reflected in the input field. When the operator selects the input field, the item extraction results are displayed as input candidates in a pop-up on the registration screen. In response to the operator's action in selecting the item extraction result, the item extraction result is reflected in the input field. An operator support device characterized by the following features.

14. An operator support device in which a processor performs processes to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. The GUI control unit is instructed to display a registration screen on the operator's terminal, in which the operator performs the registration operation for each item of the customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. Depending on the operator's operation to select the input field, if the caret is located at the beginning of the input field, the item extraction result is displayed as an input candidate in a pop-up on the registration screen; if the caret is not located at the beginning of the input field, in addition to the item extraction result, words related to the customer inquiry are displayed as input candidates in a pop-up on the registration screen. An operator support device characterized by the following features.

15. An operator support device in which a processor performs processes to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. The GUI control unit is instructed to display a registration screen on the operator's terminal, in which the operator performs the registration operation for each item of the customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. The correction results made by the operator to the item extraction results are stored in the memory unit as a correction history. In response to the operator's operation of selecting a phrase from the item extraction results in the input field, if the selected phrase matches a phrase that has been previously modified in relation to the item extraction results, based on the modification history, the modification result will be displayed as an input candidate on the registration screen. An operator support device characterized by the following features.

16. An operator support device in which a processor performs processes to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. The GUI control unit is instructed to display a registration screen on the operator's terminal, in which the operator performs the registration operation for each item of the customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. The correction results made by the operator to the item extraction results are stored in the memory unit as a correction history. Based on the aforementioned revision history, select the items that are prone to errors in the item extraction results, In the registration screen, the input fields for items that are prone to errors in the item extraction results are highlighted. An operator support device characterized by the following features.

17. An operator support device in which a processor performs processes to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. The GUI control unit is instructed to display a registration screen on the operator's terminal, in which the operator performs the registration operation for each item of the customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. In addition to determining whether the manual input is incomplete or completed, the system also determines whether the manual input is incorrect, and if it is determined to be incorrect, it automatically reflects the item extraction result in the input field. An operator support device characterized by the following features.

18. An operator support device in which a processor performs processes to support the work of an operator who creates customer service records, The aforementioned processor, The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. The GUI control unit is instructed to display a registration screen on the operator's terminal, in which the operator performs the registration operation for each item of the customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. If the manual input is determined to be an error, the manually entered content is deleted, and the item extraction result is automatically reflected in the input field. An operator support device characterized by the following features.

19. An operator support method that causes one or more processors to execute processes that support the work of operators creating customer service records, The one or more processors The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. When the input is complete and manual input by the operator has been completed, the item extraction results are not reflected in the input field. When the operator selects the input field, the item extraction results are displayed as input candidates in a pop-up on the registration screen. In response to the operator's action in selecting the item extraction result, the item extraction result is reflected in the input field. An operator support method characterized by the following:

20. An operator support method that causes one or more processors to execute processes that support the work of operators creating customer service records, The one or more processors The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. Depending on the operator's operation to select the input field, if the caret is located at the beginning of the input field, the item extraction result is displayed as an input candidate in a pop-up on the registration screen; if the caret is not located at the beginning of the input field, in addition to the item extraction result, words related to the customer inquiry are displayed as input candidates in a pop-up on the registration screen. An operator support method characterized by the following:

21. An operator support method that causes one or more processors to execute processes that support the work of operators creating customer service records, The one or more processors The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. The correction results made by the operator to the item extraction results are stored in the memory unit as a correction history. In response to the operator's operation of selecting a phrase from the item extraction results in the input field, if the selected phrase matches a phrase that has been previously modified in relation to the item extraction results, based on the modification history, the modification result will be displayed as an input candidate on the registration screen. An operator support method characterized by the following:

22. An operator support method that causes one or more processors to execute processes that support the work of operators creating customer service records, The one or more processors The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. The correction results made by the operator to the item extraction results are stored in the memory unit as a correction history. Based on the aforementioned revision history, select the items that are prone to errors in the item extraction results, In the registration screen, the input fields for items that are prone to errors in the item extraction results are highlighted. An operator support method characterized by the following:

23. An operator support method that causes one or more processors to execute processes that support the work of operators creating customer service records, The one or more processors The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. In addition to determining whether the manual input is incomplete or completed, the system also determines whether the manual input is incorrect, and if it is determined to be incorrect, it automatically reflects the item extraction result in the input field. An operator support method characterized by the following:

24. An operator support method that causes one or more processors to execute processes that support the work of operators creating customer service records, The one or more processors The speech recognition unit is instructed to perform speech recognition processing to convert the voices of the customer and operator into text using speech recognition, and the speech recognition results are obtained. The text analysis unit is instructed to perform an item extraction process to extract draft registration information for each item of the customer interaction record from the speech recognition results. A registration screen is displayed on the operator's terminal, allowing the operator to perform registration operations for each item of the aforementioned customer service record. When the item extraction results are obtained from the text analysis unit, the status of manual input by the operator for each item in the registration screen is determined. If it is determined that the data is in an unentered state and has not yet been manually entered by an operator, the item extraction results will be automatically reflected in the input field. If it is determined that the data has been manually entered by an operator, the item extraction results will not be automatically reflected in the input field. If the manual input is determined to be an error, the manually entered content is deleted, and the item extraction result is automatically reflected in the input field. An operator support method characterized by the following:

Images