Scenario creation system

Abstract

[Problem] To provide a system capable of automatically creating a scenario (script) related to a presentation. [Solution] A scenario creation system according to the present invention includes a speech analysis unit 3 for analyzing speech, a presentation material related information storage unit 5 for storing information related to presentation material, a keyword storage unit 7, a key phrase storage unit 9, a presentation evaluation unit 11, and a scenario creation unit 13. The presentation evaluation unit 11: finds one or a plurality of keywords related to the presentation material on the basis of a plurality of presentations related to the presentation material, updates one or a plurality of the keywords stored in the keyword storage unit 7, and stores the updated keywords in the keyword storage unit 7; finds one or a plurality of key phrases related to the presentation material, updates one or a plurality of the key phrases stored in the key phrase storage unit 9, and stores the updated key phrases in the key phrase storage unit 9; and selects a presentation out of the plurality of presentations that has a greater count of one or both of the updated keywords and the updated key phrases, as a presentation that is highly evaluated. The scenario creation unit 13 creates a scenario related to the presentation material on the basis of the presentation that is highly evaluated.

Description

Scenario Creation System

[0001] The present invention relates to a scenario creation system.

[0002] Japanese Patent No. 7195674 describes a presentation evaluation device that can evaluate presentations. However, even if a scenario is created based on this evaluation and machine learning, it has not been possible to create a good scenario.

[0003] Patent No. 7195674

[0004] The object of the present invention is to provide a system that can automatically create a scenario (script) related to a presentation.

[0005] This invention is based on the finding that, for example, by using key phrases to evaluate a presentation, an appropriate scenario can be automatically created.

[0006] The first invention relates to a scenario creation system 1. The scenario creation system 1 has a voice analysis unit 3, a presentation material related information storage unit 5, a keyword storage unit 7, a key phrase storage unit 9, a presentation evaluation unit 11, and a scenario creation unit 13.

[0007] The audio analysis unit 3 is a component for analyzing audio. The presentation material-related information storage unit 5 is a component for storing information related to the presentation materials. The information related to the presentation materials includes, for example, information for identifying each page of the presentation materials and one or more explanatory sentences stored in association with each page. The keyword storage unit 7 is a component for storing one or more keywords related to the presentation materials (or one or more explanatory sentences). If a page has multiple explanatory sentences, a keyword may be stored in association with each explanatory sentence. The key phrase storage unit 9 is a component for storing one or more key phrases related to the presentation materials.

[0008] The presentation evaluation unit 11 is an element for evaluating presentations. The presentation evaluation unit 11 obtains one or more keywords associated with the presentation materials based on multiple presentations related to the presentation materials. The presentation evaluation unit 11 then updates one or more keywords stored in the keyword storage unit 7 and stores the updated keywords in the keyword storage unit 7. The presentation evaluation unit 11 obtains one or more key phrases associated with the presentation materials, updates one or more key phrases stored in the key phrase storage unit 9, and stores the updated key phrases in the key phrase storage unit 9. The presentation evaluation unit 11 selects, for example, from each of the multiple presentations, a presentation with a large number of updated keywords and / or updated key phrases as a presentation with a high evaluation.

[0009] The scenario creation section 13 is an element for creating a scenario relating to presentation materials based on presentations that have received high praise.

[0010] In a preferred example of the scenario creation system 1, the updated keywords and updated key phrases are keywords and key phrases that are included in a predetermined percentage or more of the multiple presentations.

[0011] In a preferred example of the scenario creation system 1, the scenario creation unit 13 creates a scenario relating to presentation materials using keywords and key phrases contained in presentations that have received high ratings.

[0012] In a preferred example of the scenario creation system 1, the presentation evaluation unit 11 uses the updated keywords and updated key phrases to further evaluate the communication ability and listening ability in a new presentation related to the presentation material.

[0013] A preferred example of the scenario creation system 1 further includes a required-to-speak explanation display unit 21. The required-to-speak explanation display unit 21 is an element for displaying required-to-speak explanations on the display unit. The required-to-speak explanations are explanations that the presentation evaluation unit 11 has determined not to have been explained among the explanations in the presentation material stored in the presentation material related information storage unit 5. The required-to-speak explanation display unit 21 reads out the required-to-speak explanations from the presentation material related information storage unit 5 and displays them on the display unit.

[0014] A preferred example of the scenario creation system 1 further includes an explanation status display unit 23. The explanation status display unit 23 is an element for displaying, on the display unit, information about each page, information about one or more explanatory sentences, and information about one or more explanatory sentences that have been explained by the input presentation.

[0015] A preferred example of the scenario creation system 1 further includes a keyword adoption rate evaluation unit 25. The keyword adoption rate evaluation unit 25 is an element for further evaluating the input presentation using information regarding keywords contained in phonetic terms that are stored in association with the presentation materials.

[0016] A preferred example of the scenario creation system 1 further includes a topic word extraction unit 27. The topic word extraction unit 27 is an element for extracting topic words, which are included in the speech terms acquired by the speech analysis unit 3 and serve as a trigger for understanding the content of the presentation, using the speech terms acquired by the speech analysis unit 3. In this case, it is preferred that the presentation evaluation unit 11 further evaluates the input presentation using information regarding topic words that are stored in the keyword storage unit 7 and correspond to one or more keywords related to the presentation.

[0017] A preferred example of the scenario creation system 1 is to store in the keyword storage unit 7 as new keywords topic words that are common to the highly rated input presentations evaluated by the presentation evaluation unit 11 and that are not registered as one or more keywords in the keyword storage unit 7.

[0018] The next invention relates to a program for causing a computer to function as any of the above-mentioned scenario creation systems.

[0019] The next invention relates to an information recording medium on which the above program is recorded.

[0020] According to the present invention, a scenario related to a presentation can be automatically created based on the presentation.

[0021] Fig. 1 is a conceptual diagram of a scenario creation system. Fig. 2 is a conceptual diagram showing an example of a diagram of cluster classification. Fig. 3 is a conceptual diagram showing an example of a display by an explanation status display unit. Fig. 4 is a conceptual diagram showing an example of a file being input into the system.

[0022] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The present invention is not limited to the embodiments described below, but also includes appropriate modifications of the embodiments below within the scope obvious to those skilled in the art.

[0023] FIG. 1 is a conceptual diagram of a scenario creation system. As shown in FIG. 1, this system 1 includes a voice analysis unit 3, a presentation material related information storage unit 5, a keyword storage unit 7, a key phrase storage unit 9, a presentation evaluation unit 11, and a scenario creation unit 13. This system 1 may further include any of the following elements: a required-speech explanation sentence display unit 21, an explanation status display unit 23, a keyword adoption rate evaluation unit 25, and a topic word extraction unit 27. This system 1 may also include a sentence storage unit 29, or may include the sentence storage unit 29 instead of the keyword storage unit 7. This system 1 is a device implemented by a computer or a processor.

[0024] The computer has an input unit, an output unit, a control unit, a calculation unit, and a memory unit, and each element is connected by a bus or the like to enable information exchange. The various information is digital information, and the computer can process the digital information to perform various calculations, and the memory unit can store the digital information. For example, the memory unit may store a control program or various information. When predetermined information is input from the input unit, the control unit reads the control program stored in the memory unit. The control unit then reads the information stored in the memory unit as appropriate and transmits it to the calculation unit. The control unit also transmits the input information as appropriate to the calculation unit. The calculation unit performs calculation processing using the received various information and stores it in the memory unit. The control unit reads the calculation results stored in the memory unit and outputs them from the output unit. In this manner, various processes and steps are performed. The various units and means execute these various processes. The computer may have a processor, and the processor may realize various functions and steps. The computer may be standalone. Some of the computer's functions may be distributed between a server and a terminal. In this case, it is preferable that the server and the terminal are capable of transmitting and receiving information via a network such as the Internet or an intranet. The computer may include a processor and a memory coupled to the processor. The memory may store instructions that, when executed by the processor, cause the computer to perform various processes or function as various elements. The computer may be provided with various training data to construct a learning model through machine learning, and may obtain desired results by inputting various information into the trained model. The obtained results may also be input into the computer and fed back to improve the accuracy of the trained model. In this case, the computer may perform various analyses using a learning model created through machine learning and deep learning of AI (artificial intelligence).

[0025] Scenario Creation System 1 Scenario creation system 1 is a system for automatically creating a scenario (script) based on a presentation (e.g., audio). A scenario is not a game scenario, but a record of what a speaker will say. An example scenario is a presentation example (example of a speech based on the presentation material) that serves as a model for explaining or giving a presentation based on certain presentation materials. Another example of a scenario is a collection of questions and answers, and it may be a presentation to which a collection of questions and answers has been added. Presentations include anything in which speech is given based on materials, such as lectures, speeches, explanations, telephone conversations, video calls, and conversations. Presentation can be used as a verb as well as a noun. System 1 may be a single terminal or a system including multiple terminals.

[0026] Speech Analysis Unit 3 The speech analysis unit 3 is a component for analyzing speech. Speech (presentation) input to a computer is converted, for example, into a digital signal and then stored in a memory unit. The computer's control unit then reads a program stored in the memory unit, analyzes information based on the digitalized speech, and stores the analyzed speech in the memory unit as digital information, i.e., speech terms. In this manner, the speech analysis unit 3 stores presentation-related speech input from an input unit such as a microphone in the memory unit as digital information, i.e., speech terms. Multiple presentations may be evaluated and scenarios created by inputting presentations previously stored in the memory unit into the system 1. In other words, the speech analysis unit 3 need only be an element on a certain terminal and does not need to be present on the same terminal as other elements. In this case, the speech analysis unit 3 may be referred to as a speech terminology input unit. The speech terminology input unit is a component for inputting digitalized speech terms into the system 1.

[0027] Presentation Material-Related Information Storage Unit 5 The presentation material-related information storage unit 5 is an element for storing information related to presentation materials. A computer's storage unit functions as the presentation material-related information storage unit 5. The information related to the presentation materials includes, for example, information for identifying each page of the presentation materials and one or more explanatory sentences stored in association with each page. The explanatory sentences are, for example, sentences that are expected to be spoken during the presentation or explanations of the content of the presentation. For example, explanatory sentences are sometimes written in the notes section of PowerPoint (registered trademark). Each explanatory sentence may be a keyword, a single sentence, or may include multiple sentences. The system 1 can read the presentation or a relevant page of the presentation using the information related to the presentation materials.

[0028] Keyword Storage Unit 7 The keyword storage unit 7 is an element for storing one or more keywords associated with one or more descriptions. A computer's storage unit functions as the keyword storage unit 7. When a page has multiple descriptions, one or more keywords are stored in association with each description. Examples of keywords are "scenario," "improvement," and "ability."

[0029] Key Phrase Storage Unit 9 The key phrase storage unit 9 is an element for storing one or more key phrases associated with presentation materials. A computer's storage unit functions as the key phrase storage unit 9. If a page has multiple explanatory sentences, one or more key phrases may be stored in association with each explanatory sentence. A key phrase is a collection of two or more words (phrases) that are highly important and serve as keys (clues). Examples of key phrases are "standard scenario," "improvement of medical condition," and "expressive ability."

[0030] Sentence Storage Unit 29 The system 1 may have a sentence storage unit 29. The sentence storage unit 29 is an element for storing one or more sentences associated with presentation materials or one or more sentences associated with one or more explanatory texts. A computer's memory functions as the sentence storage unit 29. If there are multiple explanatory texts for a page, one or more sentences may be stored in association with each explanatory text. A sentence is a sentence. A sentence may be a complete sentence, such as a sentence including a subject and a predicate, or a sentence including a subject, predicate, and object, or an incomplete sentence, such as a sentence with an omitted subject.

[0031] Presentation Evaluation Unit 11 The presentation evaluation unit 11 is an element for evaluating presentations. For example, the input unit, control unit, calculation unit, and storage unit of a computer function as the presentation evaluation unit 11. For example, the presentation evaluation unit 11 uses the speech terms acquired by the speech analysis unit 3 to determine whether one or more keywords or key phrases are included in one or more explanatory sentences. The presentation evaluation unit 11 may then determine whether one or more explanatory sentences have been spoken and evaluate the input presentation based on information about what has been explained in the explanatory sentences of the presentation material stored in the presentation material-related information storage unit 5. The presentation evaluation unit 11 determines one or more keywords associated with the presentation material based on multiple presentations related to the presentation material. The presentation evaluation unit 11 then updates one or more keywords stored in the keyword storage unit 7 and stores the updated keywords in the keyword storage unit 7.

[0032] In addition, the presentation evaluation unit 11 determines one or more key phrases associated with the presentation materials, updates one or more key phrases stored in the key phrase storage unit 9, and stores the updated key phrases in the key phrase storage unit 9.

[0033] For example, multiple presentations (speech terms) related to a certain presentation material are input to the system 1. The speech analysis unit 3 of the system 1 then analyzes each presentation and stores the speech terms related to each presentation in the storage unit. The presentation evaluation unit 11 references the keyword storage unit 7 and the key phrase storage unit 9 to read any keywords and key phrases stored at a given time. The presentation evaluation unit 11 then reads the speech terms related to the multiple presentations and analyzes whether the read keywords and key phrases are included. Specifically, the presentation evaluation unit 11 performs a calculation to determine whether any keywords and key phrases converted into digital information match the speech terms, which are also digital information. Meanwhile, the presentation evaluation unit 11 reads the speech terms related to the multiple presentations and compares them to extract keywords and key phrases that are included in many of the speech terms of the multiple presentations. The extracted keywords and key phrases are then compared with the keywords and key phrases stored in the keyword storage unit 7 and the key phrase storage unit 9. At this time, a term dictionary can be referenced as appropriate. The term dictionary stores words and phrases. By referencing the words and phrases stored in the term dictionary, the words and phrases included in the presentation can be obtained. Then, if there are keywords and key phrases that have been used in more presentations than the keywords and key phrases originally stored in the keyword storage unit 7 and the key phrase storage unit 9, the keywords and key phrases stored in the keyword storage unit 7 and the key phrase storage unit 9 are updated.

[0034] Next, the presentation evaluation unit 11 selects, for example, a presentation with a large number of updated keywords and / or updated key phrases from among the multiple presentations as a highly rated presentation. In other words, the keywords and key phrases stored in the keyword storage unit 7 and the key phrase storage unit 9 are updated in the manner described above. Furthermore, multiple presentations related to a certain presentation material are stored in the storage unit. The presentation evaluation unit 11 then calculates the number of updated keywords and / or updated key phrases included in each presentation and stores the calculated number in the storage unit. The presentation evaluation unit 11 then compares the numbers stored in the storage unit and selects the presentation with the largest number as a highly rated presentation. Information about the selected presentation may be stored in the storage unit as appropriate. The highly rated presentation may be for the entire presentation (all pages) of a certain presentation material, or a highly rated presentation may be selected for each page of a certain presentation material.

[0035] The keyword storage unit 7 and the keyphrase storage unit 9 may store evaluation values ​​and coefficients for each keyword and keyphrase, respectively. When a keyword is included in each presentation, the evaluation values ​​and coefficients for that keyword or keyphrase may be read and added together to determine an evaluation value for each presentation. The presentation with the highest evaluation value may then be selected as the presentation with the highest evaluation value. For example, the more frequently a keyword is used in multiple presentations, the higher the evaluation value (or coefficient) it is assigned. Furthermore, for example, a keyphrase may be assigned a higher evaluation value than a keyword (e.g., 1.5 times the evaluation value if used equally frequently). However, after a presentation, users who listened to the presentation may be able to input their evaluations and comments into the system 1 from their terminals, and the highly rated presentation may be determined based on the evaluations (or evaluation values) and comments. Furthermore, when multiple people watch a presentation, scores may be input into the system 1 from the multiple users' terminals, and the highly rated presentation may be selected based on an evaluation value, such as the sum or average of the multiple input scores. Also, if the same user gives multiple presentations based on the same presentation materials, and the user's terminal selects the best presentation, the selection information may be input into system 1, thereby selecting the presentation with the highest rating.

[0036] The above is an example of evaluation, and the evaluation method of the presentation evaluation unit 11 is arbitrary. Therefore, the presentation evaluation unit 11 may evaluate the input presentation by appropriately performing calculations based on instructions from a program stored in the storage unit based on information about the explanations in the presentation materials stored in the presentation material-related information storage unit 5. For example, a case will be described in which evaluation is performed on only a certain page. This page contains four explanations A, B, C, and D, with A being the key sentence. For each explanation, multiple keywords, such as A1, A2, A3, and A4, are stored in association with each explanation. The computer compares the spoken terms with the keywords. If the spoken terms contain all of A1, A2, A3, and A4, the computer determines that explanation A has been explained. However, it may also be determined that explanation A has been explained if the spoken terms contain a predetermined percentage or more of the keywords, such as if three or more of the four keywords are included. Furthermore, the computer's storage unit may store related words for each keyword, and if a related word is included in the spoken terms, it may determine that a keyword related to the related word is included in the spoken terms. Scores are stored in a memory unit in association with explanatory sentences A, B, C, and D, and the computer reads out the scores for the explanatory sentences that have been explained and adds up the read scores. In this way, the computer may evaluate the presentation. Furthermore, if the presentation material includes multiple pages, the evaluation score for the entire presentation material may be added up. Furthermore, different coefficients may be used for important pages and unimportant pages. The above is just one example, and various calculation methods may be used to evaluate presentations so that the more explanatory sentences there are, the higher the evaluation.

[0037] The presentation evaluation unit 11 evaluates whether a sentence has been explained. The presentation evaluation unit 11 reads one or more sentences (statements) for one or more explanatory sentences from the sentence storage unit 29. The presentation evaluation unit 11 uses the phonetic terms acquired by the speech analysis unit 3 to determine whether one or more sentences for one or more explanatory sentences have been included. For example, Japanese Patent Application Laid-Open No. 2021-297025 describes a neural machine translation model for inferring a target sentence from a source sentence. As described above, a sentence agreement analysis device that determines whether a sentence matches another sentence is well known. The presentation evaluation unit 11 of the present invention includes a sentence agreement analysis device and uses the phonetic terms acquired by the speech analysis unit 3 to determine whether one or more sentences for one or more explanatory sentences have been included. The presentation evaluation unit 11 then determines whether one or more explanatory sentences have been spoken, and evaluates the input presentation based on information about what has been explained in the explanatory sentences of the presentation material stored in the presentation material-related information storage unit 5. The evaluation method used by the presentation evaluation unit 11 is arbitrary. Therefore, the presentation evaluation unit 11 may evaluate the input presentation by performing calculations based on instructions from a program stored in the appropriate storage unit, based on information about what has been explained in the explanatory sentences of the presentation material stored in the presentation material-related information storage unit 5. A specific evaluation example may be similar to the one using keywords described above. In this way, the presentation evaluation unit 11 can obtain an evaluation value for the presentation based on the sentences.

[0038] Scenario Creation Unit 13 The scenario creation unit 13 is an element for creating a scenario relating to the presentation materials based on a presentation that has received high praise.

[0039] In a preferred example of the scenario creation system 1, the updated keywords and key phrases are those that are included in a predetermined percentage or more of multiple presentations. Examples of the predetermined percentage are 50%, 60%, 70%, 80%, and 90%. When the percentage is 50%, for example, keywords and key phrases that are used in 50 or more of 100 presentations related to a certain presentation material are adopted as the updated keywords and key phrases. Keywords and key phrases may also be updated to a predetermined number.

[0040] In a preferred example of the scenario creation system 1, the scenario creation unit 13 creates a scenario relating to presentation materials using keywords and key phrases contained in presentations that have received high ratings.

[0041] The scenario creation unit 13 may use a presentation and a scenario based on the presentation as training data to construct a learning model through machine learning. Alternatively, the scenario creation unit 13 may use keywords and key phrases and a scenario based on the keywords and key phrases as training data to construct a learning model through machine learning. In this manner, a trained model can be obtained. The scenario creation unit 13 can then create a scenario by, for example, inputting keywords and key phrases into the trained model. By changing the training data for constructing the learning model, various input data can be input into the trained model to create a scenario. In this example of system 1, for example, for a presentation or each page of the presentation, keywords and key phrases included in a highly rated presentation are read from the storage unit and input into the trained model. The scenario creation unit 13 then outputs a scenario based on the highly rated presentation. The obtained scenario is appropriately stored in the storage unit, for example, as a standard scenario. Note that learning models that use machine learning to generate scenarios from input information are publicly known, as described in, for example, Japanese Patent No. 6629259 (Dialogue Scenario Generation Apparatus, Method, and Program) and Japanese Patent No. 6945909 (Message Transmission Scenario Generation Method and Message Transmission Scenario Generation Program). As described above, it is preferable that the number of update keywords and update key phrases is sufficient to create a scenario. Each scenario includes, for example, one or more sentences containing keywords and key phrases.

[0042] The scenario creation unit 13 may select a highly rated presentation for each explanatory statement related to the presentation material, and create a scenario for each explanatory statement based on the highly rated presentation (or keywords and key phrases included in the highly rated presentation).The scenario creation unit 13 may then read out scenarios for multiple explanatory statements related to a certain presentation material from the storage unit and combine them to obtain a scenario related to a certain presentation material.

[0043] The scenario creation unit 13 may classify a presentation related to a certain presentation material into multiple clusters and create a scenario for each cluster. Each of the multiple clusters includes a different topic or point of discussion. Methods for classifying conversations or talks into clusters are publicly known, as described in, for example, Japanese Patent Publication No. 7176443. The cluster classification unit, for example, reads speech terms related to the presentation from a storage unit and classifies them into multiple clusters using a learning model. Each cluster may include one or more sentences. The cluster classification unit may also analyze the points of the conversation or talk based on keywords or key phrases and classify the presentation into multiple clusters for each keyword or key phrase (or point of discussion). If the cluster classification is correct, the user may be able to input information indicating that the cluster classification is correct to the learned model. If the cluster classification is incorrect, the user may be able to input information indicating that the cluster classification is incorrect. Accuracy can be improved by repeating the cluster classification process. FIG. 2 is a conceptual diagram illustrating an example of cluster classification. In this example, the presentations were divided into smaller points based on keywords and key phrases as the content of what was being said changed, and visualized as clusters. For each cluster classification, the keywords and key phrases were updated, and based on the updated keywords and key phrases, presentations with high ratings were found for each cluster classification, and a scenario was created for each cluster classification.

[0044] The above describes the creation of a scenario based on a highly rated presentation. For example, a different invention would be one in which keywords and / or key phrases contained in a certain presentation are extracted and a scenario is created based on that presentation. In this case, the extracted keywords and key phrases may be used to create a scenario. In this case, for example, the extracted keywords and key phrases can be input into a trained model to obtain a scenario.

[0045] The scenario creation unit 13 may create a collection of questions and answers based on the updated keywords and key phrases. For example, the system 1 is connected to the Internet, can access various web pages, and can crawl the content of various web pages. The scenario creation unit 13 obtains information from various web pages based on the updated keywords and key phrases and stores it in the storage unit as appropriate. The scenario creation unit 13 may then automatically create a collection of questions and answers regarding the updated keywords and key phrases. For example, if the keyword is "Tablet A," the system 1 obtains various information from the package insert for Tablet A and stores it in the storage unit. The scenario creation unit 13 then creates a question and answer such as, "Are there any contraindications for using Tablet A?" "Yes. Tablet A cannot be prescribed to patients who have been prescribed a drug for disease B." The system 1 may, for example, have a term difficulty level dictionary. The keyword storage unit 7 may store keywords in association with difficulty levels. Furthermore, the key phrase storage unit 9 may store key phrases in association with difficulty levels. The difficulty levels can be determined by referencing the difficulty level dictionary. Examples of difficulty levels may be difficult, medium, and easy. Other examples of difficulty levels are for experts, for the general public, and easy. Other examples of difficulty levels are for doctors, for medical representatives, and for the general public. Another example of difficulty levels is by frequency. Examples of frequency levels are frequently used, occasionally used, and rarely used. In other words, keywords and key phrases may be stored along with their usage frequencies, and when usage frequency information is input to the system 1, keywords and key phrases associated with the usage frequencies may be read out, and a scenario may be created based on the keywords and key phrases corresponding to the read usage frequencies. Then, when difficulty information is input to the system 1, the scenario creation unit 13 may read keywords and key phrases associated with the difficulty levels from the keyword storage unit 7 and the key phrase storage unit 9 based on the difficulty information, and create a scenario (which may include a collection of questions and answers) corresponding to the difficulty levels.

[0046] In a preferred example of the scenario creation system 1, the presentation evaluation unit 11 uses the updated keywords and updated key phrases to further evaluate the communication ability and listening ability in a new presentation related to the presentation material. A new presentation refers to each presentation input into the system 1. In other words, the input presentation (and the presenter who gave the presentation) may be evaluated again using the updated keywords and updated key phrases.

[0047] The communication ability may include the ability to convey, the ability to explain, and the ability to persuade. The listening ability may include the ability to listen, the ability to understand, the ability to respond, the ability to tune in, and the ability to communicate. For example, when evaluating the communication ability, it is preferable to set the evaluation value of the key phrase higher than when evaluating the listening ability. In this way, the communication ability and the listening ability can be evaluated appropriately. For example, when the updated keyword is A, １ , A ２ , ...A ５ and the updated key phrase is B １ , B ２ , ...B ５ The presentation 1 input for evaluation is A １ , and A ５ Including B １ , B ２ , ...B ５ On the other hand, presentation 2 contains A １ , A ２ , ...A ５ Including B １ , and B ２ In this case, Presentation 1 (and the presenter of Presentation 1) may be evaluated as having higher communication ability and lower listening ability than Presentation 2. When evaluating, an evaluation value may be calculated based on the number of each keyword and key phrase. Alternatively, an evaluation value may be stored in association with each keyword and each key phrase, and the sum of the evaluation values ​​multiplied by a coefficient may be used as the evaluation value for evaluating the communication ability and listening ability of the presentation.

[0048] A scenario creation system 1 of one embodiment includes a voice analysis unit 3, a presentation material-related information storage unit 5, a sentence storage unit 29, a presentation evaluation unit 11, and a scenario creation unit 13. The sentence storage unit 29 is an element for storing one or more sentences associated with one or more explanatory sentences. If a page has multiple explanatory sentences, a sentence is stored in association with each explanatory sentence. The presentation evaluation unit 11 is an element for evaluating a presentation. The presentation evaluation unit 11 reads one or more sentences for each explanatory sentence from the sentence storage unit 29. Then, using the speech terms acquired by the voice analysis unit 3, the presentation evaluation unit 11 determines whether one or more sentences for each explanatory sentence are included and whether one or more explanatory sentences have been spoken, and evaluates the input presentation based on information about what has been explained in the explanatory sentences of the presentation material stored in the presentation material-related information storage unit 5.

[0049] The presentation evaluation unit 11 may obtain an evaluation value of the presentation based on the keywords and evaluate the presentation based on the evaluation value of the keyword-based presentation. For example, the presentation evaluation unit 11 may output an evaluation value or a grade based on the evaluation value. Alternatively, the presentation evaluation unit 11 may obtain a final evaluation value using the evaluation value of the presentation based on the keywords and the evaluation value of the presentation based on the sentences. The method of obtaining the final evaluation value is well known, and the final evaluation value may be obtained using a well-known calculation. The presentation evaluation unit 11 may read the evaluation values ​​of the presentation based on the keywords and the evaluation values ​​of the presentation based on the sentences from the storage unit, and cause the calculation unit to perform calculation processing based on the program read from the storage unit to obtain the final evaluation value. An example of the calculation is adding up the evaluation value of the presentation and the evaluation value of the sentence-based presentation. Another example of the calculation is multiplying one or both of the evaluation value of the presentation and the evaluation value of the presentation based on the sentences by an appropriate coefficient and then adding them to obtain the final evaluation value. Note that the presentation evaluation unit 11 evaluating the presentation based on the evaluation value of the presentation based on the sentence is a different embodiment from the above described embodiment in this specification.

[0050] In another embodiment, the computer reads one or more sentences for one or more explanatory texts, and then uses a speech terminology to determine whether one or more sentences for one or more explanatory texts have been included and whether one or more explanatory texts have been spoken, thereby evaluating the input presentation based on information about what has been explained in the explanatory texts of the presentation materials.

[0051] Required-to-speak explanation display unit 21 The required-to-speak explanation display unit 21 is an element for displaying required-to-speak explanations on the display unit. Required-to-speak explanations are explanations that the presentation evaluation unit 11 has determined not to have been explained among the explanations in the presentation material stored in the presentation material-related information storage unit 5. The required-to-speak explanation display unit 21 reads out the required-to-speak explanations from the presentation material-related information storage unit 5 and displays them on the display unit.

[0052] For example, assume that keywords B1, B2, B3, and B4 are stored in a storage unit for explanatory sentence B. The computer compares the audio terms with the keywords B1, B2, B3, and B4. For example, if the computer determines that the audio terms do not include any of the keywords B1, B2, B3, and B4, the computer determines that explanatory sentence B is not explained. In this case, for example, the keyword explanation rate may be stored in the storage unit, and if the explanation rate is lower than that rate, the computer may determine that explanatory sentence B is not explained. For example, if the explanation rate is 70%, the computer may determine that explanatory sentence B is not explained if the audio terms do not include two or more of the keywords B1, B2, B3, and B4. In this case, explanatory sentence B may be displayed on the speaker's monitor to prompt the speaker to explain explanatory sentence B. Furthermore, the speaker may be able to understand that he or she was unable to explain explanatory sentence B when reflecting on his or her presentation.

[0053] The explanation status display unit 23 is an element for displaying information about each page, information about one or more explanatory sentences, and information about the explanatory sentences that are explained in the input presentation on the display unit. An example of the information about each page is the name display of each page of the presentation. The name display is a small, deformed version of the presentation.

[0054] FIG. 3 is a conceptual diagram showing an example of a display by the explanation status display unit. In the example of FIG. 3, the presentation materials consist of five pages. Each page contains three explanatory sentences. A check mark is placed next to the explanatory sentences that were explained in a certain person's presentation. It can be seen that in this person's presentation, 10 of the 25 explanatory sentences were explained. By displaying the explanation status in this way, it is easy to understand which explanatory sentences the speaker was able to explain.

[0055] In this invention, it is preferable to output the evaluation of the communication ability and listening ability of the presentation as described above, and also output the explanation situation as described above. In this way, for each presenter, the explanation situation can be objectively displayed along with the evaluation of communication ability and listening ability, etc., and the evaluation of each presentation can be visually grasped.

[0056] Keyword adoption rate evaluation unit 25 The keyword adoption rate evaluation unit 25 is an element for further evaluating the input presentation using information about keywords stored in association with the presentation materials that are included in the phonetic terms.

[0057] For example, let us consider a case where evaluation is performed only on a certain page. This page contains four explanatory sentences A, B, C, and D. For each explanatory sentence, four keywords, i.e., A1, A2, A3, and A4, are stored in association with each explanatory sentence. This means that a total of 16 keywords are stored on this page. The computer reads these keywords from the storage unit and compares them with the speech vocabulary. In this way, the computer determines which keywords are included in the speech vocabulary and which are not. For example, if the input speech vocabulary related to a presentation contains 10 keywords, the computer can calculate the keyword adoption rate by dividing the number of included keywords by the number of keywords stored in the storage unit in association with the page. The keyword adoption rate may be calculated for the entire presentation material. Alternatively, the keyword adoption rate may be calculated by using an appropriate calculation such that the more keywords included, the higher the adoption rate.

[0058] Topic Word Extraction Unit 27 The topic word extraction unit 27 is an element for extracting topic words, which are included in the speech terms acquired by the speech analysis unit 3 and serve as a trigger for understanding the content of the presentation, using the speech terms acquired by the speech analysis unit 3. In this case, it is preferable that the presentation evaluation unit 11 further evaluates the input presentation using information on topic words that are stored in the keyword storage unit 7 and correspond to one or more keywords related to the presentation.

[0059] For example, consider the case of obtaining topic words for a certain page. This page has four explanatory sentences A, B, C, and D, and for each explanatory sentence, four keywords, such as A1, A2, A3, and A4, are stored in association with each explanatory sentence. A computer analyzes the phonetic terms, extracts multiple nouns, and stores them in a memory unit. The computer then reads the stored nouns and performs a calculation to match them with the keywords stored in association with the page. The computer then finds that some of the nouns stored in the memory unit do not correspond to any of the keywords. The computer stores these nouns in the memory unit as topic word candidates. These topic word candidates may be used as topic words. Alternatively, an Internet search may be performed using the topic word candidates and keywords related to presentation materials. If both a topic word candidate and a keyword exceeding a predetermined threshold are used on the same website, the topic word candidate may be used as a topic word. The calculation for determining topic word candidates as topic words may be adjusted as appropriate. In this way, the topic word extraction unit 27 extracts topic words from the speech terminology. The extracted topic words are stored in a storage unit as appropriate.

[0060] Keyword Update Unit A preferred example of the scenario creation system 1 is to store in the keyword storage unit 7 as new keywords topic words that are common to the highly rated input presentations evaluated by the presentation evaluation unit 11 and that are not registered as one or more keywords in the keyword storage unit 7.

[0061] For example, evaluation values ​​and topic words associated with multiple presentations are stored in a storage unit. The computer stores a threshold value and classifies evaluation values ​​above the threshold value as top performers (those with high evaluations). If a topic word is used at a rate above a predetermined threshold in association with the presentations of multiple highly rated people, the computer stores the topic word in the storage unit as a new keyword.

[0062] Next, an example of creating a scenario using the system 1 will be described. FIG. 4 is a conceptual diagram showing an example of inputting a file (a file containing a recorded presentation or an audio file containing a recorded presentation) into the system 1. As shown in FIG. 4, for example, when an audio file or a video file is dragged and dropped into the system 1, the presentation or audio terminology is input into the system 1. Also, if the video file is related to a presentation, audio related to the presentation is input into the sentence correction system 1, resulting in multiple sentences being input into the system 1. In this case, each input sentence may be input into the system 1 as a first sentence, a second sentence, a third sentence, etc., or characteristic sentences among the multiple sentences may be input into the system 1 as a first sentence, a second sentence, a third sentence, etc.

[0063] The computer (presentation material-related information storage unit 5) stores information about the presentation materials. The information about the presentation materials may include information for identifying each page of the presentation materials and one or more explanatory sentences stored in association with each page. The computer (keyword storage unit 7) stores one or more keywords associated with the presentation materials (or one or more explanatory sentences). The computer (keyphrase storage unit 9) stores one or more key phrases associated with the presentation materials (or one or more explanatory sentences). When a presentation is input to the computer (its audio input unit, for example, a microphone), the computer (speech analysis unit 3) converts the audio information into digital information, analyzes the presentation converted into digital information, and obtains audio terms, which are terms included in the input presentation. The audio terms are stored in the storage unit as appropriate. As described above, the audio terms may be read by the computer and input to the system 1 by drag-and-drop.

[0064] The computer (presentation evaluation unit 11) reads one or more keywords and one or more key phrases related to the presentation materials (for one or more explanatory sentences) from the storage unit. The computer (presentation evaluation unit 11) uses audio terms to determine whether one or more keywords and one or more key phrases related to the presentation materials (for one or more explanatory sentences) are included. At this time, the computer may determine whether one or more explanatory sentences are spoken. The presentation evaluation unit 11 obtains one or more keywords associated with the presentation materials based on multiple presentations related to the presentation materials, updates one or more keywords stored in the keyword storage unit 7, and stores the updated keywords in the keyword storage unit 7. The presentation evaluation unit 11 obtains one or more key phrases associated with the presentation materials, updates one or more key phrases stored in the key phrase storage unit 9, and stores the updated key phrases in the key phrase storage unit 9. The presentation evaluation unit 11 determines, among the multiple presentations, a presentation with a large number of updated keywords and / or updated key phrases as a presentation with a high evaluation. Furthermore, the presentation evaluation unit 11 may evaluate the presentations using keywords and key phrases stored in the keyword storage unit 7 and the key phrase storage unit 9. A computer (scenario creation unit 13) creates a scenario for the presentation materials based on highly rated presentations. The scenario creation unit 13 may, for example, input keywords and key phrases included in highly rated presentations into a trained model to create a scenario. Alternatively, the scenario creation unit 13 may input highly rated presentations into a trained model to create a scenario. However, a scenario may also be created using keywords and key phrases included in the input presentation without evaluating the presentation.The system 1 may store keywords and key phrases together with their frequency of use, and when frequency of use information is input to the system 1, the system 1 may read out keywords and key phrases associated with the frequency of use and create a scenario based on the keywords and key phrases corresponding to the read frequency of use. When difficulty level information is input to the system 1, the scenario creation unit 13 may read out keywords and key phrases associated with the difficulty level from the keyword storage unit 7 and the key phrase storage unit 9 based on the difficulty level information, and create a scenario (which may include a collection of questions and answers) corresponding to the difficulty level.

[0065] Furthermore, the scenario creation unit 13 may create a collection of questions and answers using keywords and key phrases included in the presentation.

[0066] In this way, when a predetermined number or more presentations relating to a certain material have been input into the system, the newly input presentations relating to that material can be evaluated using the keywords stored in the keyword storage unit 7 and the key sentences stored in the key sentence storage unit 9. The presentation evaluation unit 11 determines whether the speech terms relating to the newly input presentation match the keywords stored in the keyword storage unit 7 and the key sentences stored in the key sentence storage unit 9. The presentation evaluation unit 11 then evaluates the presentation based on the matching keywords and key sentences. As an example of the evaluation, evaluations may be stored according to the number of matching keywords and key sentences, and the presentation evaluation unit 11 may read out the stored evaluations based on the number of matches.

[0067] This specification also discloses a program for causing a computer to function as the above-mentioned device, a program for causing a computer to implement a method based on the above-mentioned device, and a computer-readable information recording medium storing such a program. The information recording medium is preferably a non-transitory recording medium. Examples of the information recording medium include a CD, a CD-ROM, a DVD, a USB memory, a hard disk, and a disk on a server.

[0068] The present invention relates to a presentation support device and can be used in the information-related industry.

[0069] 1 Scenario creation system 3 Speech analysis unit 5 Presentation material related information storage unit 7 Keyword storage unit 9 Key phrase storage unit 11 Presentation evaluation unit 13 Scenario creation unit 21 Explanation sentence to be spoken display unit 23 Explanation status display unit 25 Keyword adoption rate evaluation unit 27 Topic word extraction unit 29 Sentence storage unit

Claims

a presentation material related information storage unit (5) for storing information related to presentation materials; a keyword storage unit (7) for storing one or more keywords associated with the presentation materials; a key phrase storage unit (9) for storing one or more key phrases associated with the presentation materials; a presentation evaluation unit (11) for evaluating the presentations; and a scenario creation unit (13) for creating a scenario, wherein the presentation evaluation unit (11) obtains one or more keywords associated with the presentation materials based on a plurality of presentations related to the presentation materials, updates the one or more keywords stored in the keyword storage unit (7), and stores the updated keywords in the keyword storage unit (7), obtains one or more key phrases associated with the presentation materials, updates the one or more key phrases stored in the key phrase storage unit (9), and stores the updated key phrases in the key phrase storage unit (9); and among each of the plurality of presentations, a presentation having a large number of either or both of the updated keywords and the updated key phrases is determined to be a presentation having a high evaluation, The scenario creation section (13) creates a scenario relating to the presentation materials based on the highly rated presentation, in a scenario creation system.

2. A scenario creation system as described in claim 1, wherein the updated keywords and the updated key phrases are keywords and key phrases that are included in a predetermined percentage or more of the plurality of presentations.

3. A scenario creation system as described in claim 2, wherein the scenario creation unit (13) creates a scenario relating to the presentation materials using keywords and key phrases contained in the highly rated presentations.

4. A scenario creation system as described in claim 3, wherein the presentation evaluation unit (11) further evaluates the communication ability and listening ability in a new presentation related to the presentation material using the updated keywords and the updated key phrases.

5. A program for causing a computer to function as the scenario creation system according to claim 1.

Images