JP2021081452A - Presentation support method using computer and support system - Google Patents

Abstract

To provide a system and a program capable of easily checking whether contents to be spoken have been accurately spoken.SOLUTION: A plurality of keywords to be spoken (or the keywords which are not allowed to be spoken) depending on contents to be spoken are previously stored. Voice recognition is performed for presentation etc. Whether the plurality of keywords have been spoken is checked, and the keywords are stored so as to be displayed. Even if the conversation is not reproduced, it can be easily grasped that appropriate explanation etc. has been presented.SELECTED DRAWING: Figure 4

Description

The present invention relates to a presentation support method using a computer and a support system.

For example, Japanese Patent No. 4551105 discloses a conference support system using voice recognition. It is very convenient to perform voice recognition and memorize the contents of the meeting. On the other hand, there may be an obligation to explain necessary information to the other party, such as informed consent, sales of mobile terminals, and real estate contracts. In addition, there are matters that the sales staff of a company should talk about as a company and things that should not be talked about. It was basically left to the speaker, such as the sales staff, to decide whether or not they talked about these matters to be talked about.

Japanese Patent No. 4551105

It would be cumbersome to record, play back, and check the conversation as to whether the speaker said what he should or shouldn't say. In addition, speakers often forget to talk about what they should talk about because they concentrate too much on the story.

Therefore, it is an object of the invention described in this specification to provide a system that makes it easy to check whether or not the content to be spoken is properly spoken. In addition, in a preferable example of the invention, when a matter that is omitted is called attention or a matter that should not be talked about, an opportunity for correction is given and a material for evaluating the person who talks is used. The purpose is to be able to obtain.

The present invention solves the above-mentioned problems. Basically, a plurality of keywords (or keywords that should not be spoken) to be spoken are stored in advance according to the content to be spoken, and a presentation or the like is voice-recognized. However, it is based on the knowledge that by checking whether multiple keywords are separated and memorizing them so that they can be displayed, it is possible to easily grasp that appropriate explanations have been given without playing the conversation. Furthermore, a preferred example of the present invention is that if there is a keyword that is omitted in the explanation during a presentation or the like, the keyword can be read out and displayed to draw the attention of the speaker and prevent the explanation from being omitted. Based on knowledge.

The invention disclosed in the present specification first relates to a presentation support method using a computer. This method is implemented by a computer. The subject of this method is basically a computer. This method first receives input about the content of the presentation. Then, a plurality of keywords related to the input are read from the storage unit. Next, the voice including the presentation is voice-recognized, and the words recognized by the voice are analyzed. Then, matching is performed between a plurality of keywords and words recognized by voice.

This method preferably records the presentation and stores the matching results. Then, among a plurality of keywords, the terms matched by voice recognition are checked to the effect that they are included in the story. Then, the check result is recorded so that it can be displayed. When the sales representative, etc. checks whether all the necessary items have been spoken, the administrator can check whether the necessary items have been spoken without playing the recording, so explanation of important matters, informed consent, etc. Checking becomes easier.

This method preferably further includes a step of outputting a keyword that is not included when it is determined as a result of matching that some of the plurality of keywords are not included in the speech-recognized terms. According to this method, it is possible to issue an alert so that there is no omission of explanation, thereby preventing omission of explanation. In addition, by recording the fact that there was a keyword omission in a displayable manner, it becomes a material for evaluating the aptitude as a sales representative.

This method preferably includes keywords that should be presented and keywords that should not be presented. Then, when it is determined that the speech-recognized term contains a keyword that should not be presented, an alert is output to notify that the keyword that should not be presented is included. As a result, if an NG word that should not be explained is included in the presentation or explanation, an alert can be issued to prompt correction or the like. In addition, by recording the number of times the NG word is spoken, it becomes a material for evaluating the aptitude as a sales representative.

According to the present invention, it is possible to provide a system that can easily check whether or not the content to be spoken is properly spoken. In addition, in a preferable example of the invention, when a matter that is omitted is called attention or a matter that should not be talked about, an opportunity for correction is given and a material for evaluating the person who talks is used. Obtainable.

FIG. 1 is a block diagram showing a basic configuration of a computer. FIG. 2 is a conceptual diagram showing an example of a display screen of a person in charge. FIG. 3 is an example showing a plurality of keywords and NG words read out for the drug AAA. FIG. 4 is a conceptual diagram showing the matching situation between the keyword and the presentation during the presentation. FIG. 5 is a partial flow for determining whether or not a keyword is included. FIG. 6 is a conceptual diagram showing an example of a screen on which keywords are displayed. FIG. 7 is an example of the evaluation screen.

Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. The present invention is not limited to the forms described below, and includes those which are appropriately modified by those skilled in the art from the following forms to the extent obvious to those skilled in the art.

The invention disclosed in the present specification first relates to a presentation support method using a computer. This method is implemented by a computer (including a computer alone, a mobile terminal, or one connected to a server or the like).

The computer may be any one or a combination of two or more of a mobile terminal, a desktop personal computer, and a server. These are usually connected so that information can be exchanged via the Internet (intranet) or the like. Functions may be shared by using multiple computers, such as assigning some functions to one of the computers.

As shown in FIG. 1, which is a block diagram showing a basic configuration of a computer, the computer has an input unit 31, an output unit 33, a control unit 35, a calculation unit 37, and a storage unit 39. Each element is connected by a bus 41 or the like so that information can be exchanged. For example, the control program may be stored in the storage unit, or various types of information may be stored in the storage unit. When the specified information is input from the input unit, the control unit reads out the control program stored in the storage unit. Then, the control unit appropriately reads out the information stored in the storage unit and transmits it to the arithmetic unit. In addition, the control unit transmits the appropriately input information to the calculation unit. The arithmetic unit performs arithmetic processing using various received information and stores it in the storage unit. The control unit reads the calculation result stored in the storage unit and outputs it from the output unit. In this way, various processes are executed. Each means executes these various processes.

The subject of this method is basically a computer. In the following example, a presentation and presentation support system when a person in charge such as an MR recommends his / her own medicine to a doctor will be described as an example.

FIG. 2 is a conceptual diagram showing an example of a display screen of a person in charge. As shown in FIG. 2, the tablet of the person in charge displays the drug AAA, the drug BBB, and the drug CCC. Then, for example, suppose that the person in charge introduces the medicine AAA to the doctor. A tablet or a server (computer) connected to the tablet stores a presentation document for explaining the pharmaceutical AAA. The computer may also be capable of reading presentations associated with pharmaceutical AAA. In addition, the computer stores a plurality of keywords, Keys 1 to 5, which should be spoken, and NG words 1 and 2, which are keywords which should not be spoken, in relation to the pharmaceutical AAA. In addition, the computer stores ○ minutes, which is the presentation time (or explanation time), in relation to the drug AAA.

The person in charge clicks the drug AAA button or icon on the tablet. Then, information indicating that the drug AAA has been selected is input to the computer. In this way, the computer receives input about the content of the presentation. Examples of input regarding the content of the presentation are the title of the presentation, the explanation of important matters related to the mobile terminal contract, the explanation of the real estate contract, the target medicine, and multiple cases related to the conversation and the presentation. It is the object to judge whether the keyword is spoken properly.

When the computer receives the input information that the drug AAA has been selected, it is a keyword that should not be spoken with Keys 1 to 5, which are multiple keywords to be spoken, which are stored in connection with the drug AAA from the storage unit. Read NG words 1 and 2. At this time, only Keys 1 to 5, which are a plurality of keywords to be spoken, may be read out. In addition, the computer reads out the average presentation time of ○ minutes in relation to the drug AAA. In this way, information including a plurality of keywords related to the input is read from the storage unit.

FIG. 3 is an example showing a plurality of keywords and NG words read out for the drug AAA. In this example, in connection with the pharmaceutical AAA, a plurality of keywords that should be spoken, Keys 1 to 5, and keywords that should not be spoken, NG words 1 and 2, are read out. In addition, these can be displayed separately on the monitor. Since multiple keywords and NG words can be displayed on the monitor, the person in charge can check what should be said and what should not be said before giving a concrete explanation or presentation.

When the person in charge selects the drug AAA, the presentation material related to the drug AAA is read out and displayed on, for example, the monitor of the person in charge's tablet or the monitor that the other party can see. The person in charge gives a presentation and explanation based on this presentation material. Then, the voice recognition device in the tablet recognizes the voice including the presentation. At the time of voice recognition, the frequency band and the tempo of the talk may be stored together. By memorizing the frequency band and the tempo of the story, when the speaker changes, it is possible to recognize that the speaker has changed. It is optional to adopt the speaker identification method, and if so, what method to adopt.

The computer analyzes the presented terms stored in the storage unit and stores the analyzed terms in the storage unit. Then, the computer analyzes whether or not the words recognized by voice and the keywords match. Then, matching is performed between a plurality of keywords and words recognized by voice.

FIG. 4 is a conceptual diagram showing the matching situation between the keyword and the presentation during the presentation. In the example shown in FIG. 4, it can be seen that Keys 1 to 3 of the plurality of keywords to be spoken, Keys 1 to 5, are spoken. In other words, the computer matched the read keyword group with the terms obtained by the analysis, and as for Keys 1 to 3, they matched the terms, and therefore, the keys to the effect that they matched were added to Keys 1 to 3. For example, the situation shown in FIG. 4 is memorized so as to be displayable together with the conversation of the speaker. Then, by looking at the display screen as shown in FIG. 4, the speaker's boss or the like can grasp that the NG word was not spoken, but not all the keywords that need to be spoken were spoken. Further, for example, when the situation that all the keywords are spoken is displayed, the boss or the like can have a guess that the necessary items are properly spoken without playing back the recorded content.

That is, this method preferably records the presentation and stores the matching results. Then, among a plurality of keywords, the terms matched by voice recognition are checked to the effect that they are included in the story. Then, the check result is recorded so that it can be displayed. When the sales representative, etc. checks whether all the necessary items have been spoken, the administrator can check whether the necessary items have been spoken without playing the recording, so explanation of important matters, informed consent, etc. Checking becomes easier.

In a preferred example of this method, the computer becomes a predetermined percentage of the read presentation time (explanation time) or a predetermined time (this time may be greater than the presentation time (or explanation time)). ， If it is determined that all Key words are not spoken, the monitor may be controlled to display those keywords (for example, at the bottom of the part where the presentation is made). For example, a measuring means in a computer measures the start of a presentation (for example, the time after the drug AAA is selected or the time after voice recognition of the voice including the presentation), and the time is stored in the storage unit. When the memorized predetermined time is reached, the program that determines whether the keyword and the term match is started. Then, the control unit reads out a plurality of keywords and the analyzed terms. The predetermined time may be stored in the storage unit in advance, and for example, a predetermined ratio (for example, a numerical value between 70 to 95% and 80 to 90%) with respect to the standard presentation time). In that case, matching is performed to see if there are any keywords that have not been presented, and if there are keywords that have not been presented, for example, a message indicating that there are keywords that have not been spoken and keywords that have not been spoken are displayed. It may be output to a monitor or the like.

FIG. 5 is a flowchart for matching keywords and terms. The control unit reads Key1 from the storage unit. Then, it is determined whether or not the analyzed term matches key1. If any of the terms matches Key1, the process of setting a flag indicating that Key1 has been spoken is performed. For example, the control unit stores in the storage unit that Key1 has been spoken in connection with this presentation and explanation. The same work is performed for multiple keywords Key 1 to 5 that should be spoken and NG words 1 and 2 that are keywords that should not be spoken. In this case, for example, when the presentation material is displayed on both the monitor of the person in charge of presentation and the monitor of the other party, the Key word is displayed only on the monitor of the speaker (the person giving the presentation). May be good. Further, this Key word is preferably displayed at the lower part of the presentation.

FIG. 6 is a conceptual diagram showing an example of a screen on which keywords are displayed. In the example of FIG. 6, two keywords, Key4 and Key5, without the matching flag, are displayed at the bottom of the presentation screen related to the pharmaceutical AAA. In this way, among the keywords to be explained, those that have not been spoken even after a predetermined time has passed are displayed on the monitor, so that the speaker can be alerted.

That is, this method preferably further includes a step of outputting a keyword that is not included when it is determined as a result of matching that some of the plurality of keywords are not included in the speech-recognized terms. According to this method, it is possible to issue an alert so that there is no omission of explanation, thereby preventing omission of explanation. In addition, by recording the fact that there was a keyword omission in a displayable manner, it becomes a material for evaluating the aptitude as a sales representative.

Then, when it is determined that the speech-recognized term contains a keyword that should not be presented, an alert may be output to notify that the keyword that should not be presented is included. In this case, it is preferable that the NG word is displayed instead of the Key 4 and Key 5 in FIG. 6, and a warning to the effect that the NG word is spoken is displayed on the display unit. This alert display is stored in the storage unit, and when the computer determines that the NG word has been spoken (the analyzed term matches the NG word), the alert display stored in the storage unit. Information about the above should be read out and displayed on the display. As a result, if an NG word that should not be explained is included in the presentation or explanation, an alert can be issued to prompt correction or the like. In addition, by recording the number of times the NG word is spoken, it becomes a material for evaluating the aptitude as a sales representative.

FIG. 7 is an example of the evaluation screen. In this example, the number of keywords to be explained that were not matched by the predetermined time, the number of keywords that were not matched by the predetermined time to be explained that were explained by the end of the presentation, and the matched NG keywords. The number of (keywords that should not be spoken) is displayed. This number is stored in the storage unit and is used for various evaluations of the sales person AAA. This person is in charge of explaining multiple medicines. Then, in the presentation shown on the left side of the pharmaceutical AAA in the example of this screen, although one keyword to be explained was failed to be explained by the preset time, the keyword was displayed after the keyword was displayed. You can see that it was explained and corrected. It can also be seen that the NG keyword was not spoken. The administrator who sees this management screen can play the conversation at the time of the presentation by pressing the play button existing in the column, for example. However, since all the necessary keywords are spoken and the NG keyword is not spoken, the administrator can grasp that an appropriate presentation was made without playing. In this case, the approval button is displayed on the management screen, so when the administrator presses the approval button, the computer receives input of information that the approval has been made. Then, the computer records in the storage unit information that the presentation has been approved in connection with this presentation.

On the other hand, in the central presentation of FIG. 7, one keyword to be spoken is leaked within a predetermined time, and the keyword is not recognized by the end of the presentation. In addition, one NG keyword is recognized. In this case, the boss can play the audio stored in connection with this presentation by pressing the play button. Further, in this case, as shown in FIG. 7, the computer cannot approve until the input of the play button and the computer determines that the voice has been played, and approves the voice after it has been played. The button may be displayed. The boss who played the presentation may be able to modify the evaluation of this presentation as appropriate. In this case, the computer receives input of correction information about this presentation. Then, the evaluation of the presentation stored in the storage unit (for example, required-1) is corrected, and the updated information is recorded in the storage unit. After playing the audio, you can approve this presentation by pressing the approve button. In this way, it is possible to easily evaluate each person in charge, and it is possible to easily grasp whether the presentation and explanation were made properly.

This specification also provides a presentation support computer system for implementing the above method. The system has a means of receiving input related to the content of the presentation, a means of reading multiple keywords related to the input from the storage unit, a means of voice-recognizing the presentation and analyzing the voice-recognized terms, and a means of voice-recognizing multiple keywords. It includes means for matching terms and means for recording presentations and storing matching results. The system also includes various means for implementing the methods described above. Examples of various means are timekeeping means and various display means.

The specification also provides a presentation support program for implementing the above method and a computer-readable information recording medium containing the program. The program is. It controls the computer to perform the above steps. The program also includes a means for the computer to receive input related to the content of the presentation, a means for reading a plurality of keywords related to the input from the storage unit, a means for voice-recognizing the presentation, and a means for analyzing the voice-recognized term. It functions to include a means for matching keywords and speech-recognized terms, and a means for recording a presentation and memorizing the matching result.

The present invention can be used in the display terminal industry, the information and communication industry, software development, the pharmaceutical industry, the financial industry, and the like.

Claims (7)

It is a presentation support method using a computer.
The computer
The process of receiving input about the content of the presentation and
A process of reading a plurality of keywords related to the input from the storage unit, and
The process of speech recognition of speech including presentations and analysis of speech-recognized terms,
The process of matching the plurality of keywords with the voice-recognized term, and
How to include. The method according to claim 1.
The computer
A method further comprising recording the presentation and storing the matching results. The method according to claim 1.
The computer
As a result of the matching, when it is determined that there is a keyword that is not included in the voice-recognized term among the plurality of keywords, a method further including a step of outputting the keyword that is not included. The method according to claim 3.
The computer
It further includes the step of measuring the time of the presentation after receiving the input regarding the content of the presentation.
A method in which the computer outputs the keywords that are not included after the presentation time has elapsed for a predetermined time. The method according to claim 1.
The plurality of keywords include keywords that should be presented and keywords that should not be presented.
In the matching step, when the voice-recognized term contains the keyword that should not be presented,
A method further comprising the step of outputting an alert notifying that the computer contained a keyword that should not be presented. A means of receiving input about the content of the presentation and
A means for reading a plurality of keywords related to the input from the storage unit, and
A means of speech-recognizing presentations and analyzing speech-recognized terms,
A means for matching the plurality of keywords with the voice-recognized term, and
A means for recording the presentation and storing the matching result,
Presentation support computer system, including. It is a presentation support program
Computer,
A means of receiving input about the content of the presentation and
A means for reading a plurality of keywords related to the input from the storage unit, and
A means of speech-recognizing presentations and analyzing speech-recognized terms,
A means for matching the plurality of keywords with the voice-recognized term, and
A means for recording the presentation and storing the matching result,
A program that works to include.

Images