JP2024060181A - Vocabulary evaluation device, vocabulary evaluation method, and vocabulary evaluation program - Google Patents

Abstract

[Problem] To accurately evaluate the vocabulary ability of a subject. [Solution] A vocabulary evaluation device 1 having a word collection means 11 that collects words included in the subject's utterance via a terminal 2 connected for communication, an environment acquisition means 13 that acquires the environment in which the utterance was made based on information from the terminal, and an analysis means 14 that analyzes the subject's word acquisition status for each environment. The word collection means has a voice receiving means 111 that acquires an audio source that has picked up the contents of the utterance, and the vocabulary evaluation device may further have a voice recognition means 122 that recognizes the audio source and extracts words included in the audio source. [Selected Figure] Figure 2

Description

The present invention relates to a technology for evaluating a subject's vocabulary ability.

Various technologies have been proposed for evaluating children's language development. For example, Patent Document 1 describes a method of supporting vocabulary and language learning by collecting speech by arranging at least one microphone in the learner's listening environment. Patent Document 2 describes a technology for determining facial expressions detected based on image data. Patent Document 3 discloses a development support device that determines the development level of each user by tallying up the number of utterances of each type for multiple users. Patent Document 4 discloses a technology for digitizing a support chart that supports growth and development.

Published Patent Application No. 2008-513840 JP 2018-067017 A JP 2012-242732 A Patent Publication No. 3217617

In conversations between people, the vocabulary and pragmatic usage change depending on the conversation environment, such as the relationship with the conversation partner and the location. For example, the vocabulary used in conversations with close friends, such as family members, may differ from the vocabulary used in conversations with strangers, such as teachers or medical professionals. In addition, when a person is taught new words through language development training, they may repeat the words on the spot, but not use them in other environments. Thus, to accurately evaluate whether a subject has acquired words, it is essential to refer to the conversation environment, but conventional technology has not been able to perform a detailed evaluation.

Therefore, one of the objectives of the present invention is to more accurately evaluate a subject's vocabulary ability.

To achieve the above object, a vocabulary assessment device according to one aspect of the present invention includes a word collection means for collecting words contained in a subject's utterance via a terminal connected for communication, an environment acquisition means for acquiring the environment in which the utterance was made based on information from the terminal, and an analysis means for analyzing the subject's word acquisition status for each environment.

The word collection means may include a voice receiving means for acquiring a sound source including the speech collected by the terminal, and the vocabulary evaluation device may further include a voice recognition means for performing voice recognition on the sound source and extracting the words included in the sound source.

The system may further include a storage means for storing information for identifying the voice of the subject in advance, and a voiceprint identification means for referring to the storage means and identifying speech by the subject from the sound source acquired by the voice receiving means.

The word collection means may include an input information receiving means for receiving the words input via the terminal.

The environment acquisition means may acquire identification information of the terminal, and the analysis means may analyze the acquisition status of the words for each identification information of the terminal, assuming that the words collected on the same terminal were spoken in the same environment.

The environment acquisition means may acquire location information of the terminal, and the analysis means may analyze the acquisition status of the words for each piece of location information, assuming that the words collected at the same location were spoken in the same environment based on the location information.

The word collection means collects the words contained in the speech of one of the subjects via a plurality of the terminals, and further includes a storage means for storing at least the identification information of the terminal in association with the owner of the terminal, the environment acquisition means acquires the identification information and location information of the terminal, and the analysis means may analyze the learning status of the words acquired via the terminal for each identification information of the terminal if the owner of the terminal that acquired the words is not the subject, and may analyze the learning status of the words acquired via the terminal for each location information of the terminal if the owner of the terminal that acquired the words is the subject.

The device may further include a display control means for displaying on the terminal the learning status of the word for each environment analyzed by the analysis means.

To achieve the above object, a vocabulary assessment method according to another aspect of the present invention includes a word collection step of collecting words contained in a subject's utterance via a terminal connected for communication, an environment acquisition step of acquiring the environment in which the utterance was made based on information from the terminal, and an analysis step of analyzing the subject's word acquisition status for each environment.

To achieve the above object, a vocabulary assessment program according to one aspect of the present invention causes a computer to execute, via a communicatively connected terminal, a word collection command to collect words contained in a subject's speech, an environment acquisition command to acquire the environment in which the speech was made based on information from the terminal, and an analysis command to analyze the subject's word acquisition status for each environment.

Computer programs can be provided by downloading via a network such as the Internet, or by recording them on various computer-readable recording media.

The present invention allows for a more accurate assessment of a subject's vocabulary.

1 is a conceptual diagram showing an overview of an environment in which a vocabulary evaluation device according to an embodiment of the present invention is utilized. 1 is a functional block diagram illustrating an embodiment of a vocabulary evaluation device according to an embodiment of the present invention. 1A to 1D are diagrams showing examples of screens displayed on a terminal by a vocabulary evaluation device according to an embodiment of the present invention, where (a) is a first example, (b) is a second example, (c) is a third example, and (d) is a fourth example. 11A to 11C are diagrams showing other examples of screens displayed on a terminal by the vocabulary evaluation device according to the embodiment of the present invention, where (a) is a fifth example, (b) is a sixth example, and (c) is a seventh example. FIG. 2 is a sequence diagram showing a first example of a processing flow of the vocabulary evaluation device according to the embodiment of the present invention. FIG. 11 is a sequence diagram showing a second example of the processing flow of the vocabulary evaluation device according to the embodiment of the present invention.

An embodiment of a vocabulary evaluation device according to the present invention will be described with reference to the drawings.

Overview of the Vocabulary Evaluation Device The vocabulary evaluation device 1 shown in FIG. 1 is a device for evaluating the language acquisition status of a subject. The subject is, for example, a child, and in particular a child with a language development disorder is assumed, but the presence or absence of a developmental disorder and the degree of the developmental disorder are not important. The subject may also be a dementia patient or an elderly person. In this case, the vocabulary evaluation device 1 may evaluate vocabulary loss. Note that the subject is not limited to those with the above-mentioned characteristics or diseases, and can be applied to anyone of any age.

The vocabulary evaluation device 1 communicates with multiple terminals 2 to link multiple collaborators or collaborating facilities surrounding the subject. The collaborators are, for example, medical professionals or speech-language-hearing therapists 101, guardians 102, school teachers, and staff at therapeutic facilities. The collaborating facilities can be therapeutic facilities 103, schools 104, and workshops where disabled people work. The vocabulary evaluation device 1 is also capable of acquiring data from universities or research institutes 110 and reflecting it in the vocabulary evaluation at any time.

As shown in FIG. 2, for example, the vocabulary evaluation device 1 is configured as a vocabulary evaluation system 100 by connecting one or more terminals 2 via a network NW. Note that when collecting audio, the vocabulary evaluation system 100 can evaluate not only vocabulary but also information contained in the audio, as described below, and therefore can also be called a vocabulary and audio evaluation system.

Terminal 2 is assumed to be held by, for example, a collaborator, but may also be carried by the subject himself. Terminal 2 may be, for example, a smartphone, a tablet terminal, or a personal computer. Application software that realizes, for example, the vocabulary evaluation system 100 is installed on terminal 2, and each function related to the vocabulary evaluation system 100 is executed when the user launches the application software. Furthermore, terminal 2 may realize some or all of the functions related to the vocabulary evaluation system 100 via an internet browser instead of the application software.

The terminal 2 has unique identification information. The terminal 2 may be configured to be pre-linked to each collaborator in terms of hardware, or may be configured so that the collaborator's information is linked to the identification information of the terminal 2 when the collaborator logs in using a specific ID and password, etc. The number of terminals 2 is arbitrary.

The vocabulary evaluation system 100 may also include a sound collection device 3. The sound collection device 3 is installed in a facility used by the subject, such as a cooperating facility. The sound collection device 3 may be a device installed for the purpose of linking with the vocabulary evaluation device 1, or an appropriate device installed in advance may perform the function of the sound collection device 3. The sound collection device 3 is, for example, a microphone. The sound collection device 3 may also be a video camera that can also capture video. The sound collection device 3 is not limited to being installed in a fixed manner, and may be in any manner that collects the speech of the subject. The number of sound collection devices 3 may be one or more, and the configuration may not include a sound collection device 3.

In this embodiment, it is assumed that there may be multiple client terminals equipped with the vocabulary evaluation device 1, and these may be located in distant or nearby locations, and the client terminals may communicate with each other. Also, there may be a server terminal equipped with the vocabulary evaluation device 1, and multiple users or one user may use the vocabulary evaluation device 1 by accessing the server terminal.

The vocabulary evaluation device 1 may be configured in one or more distributed hardware components, and some or all of its functions may be configured on a cloud computer. The specific configuration of the vocabulary evaluation device 1 will be described later.

Each of the multiple terminals 2 is connected to one vocabulary evaluation device 1 .
The terminal 2 includes, for example, a sound pickup unit 21, an input unit 22, a display unit 23, and a position information acquisition unit 24.

The sound collection unit 21 is realized by, for example, a microphone, and collects human voices as voice data. The voice data is transmitted to the vocabulary evaluation device 1 by an appropriate communication processing unit. The terminal 2 may have a voiceprint identification function, and may use this function to store only the subject's voice in advance, or to perform individual identification by voiceprint authentication in post-processing on the cloud, etc., and transmit the data to the vocabulary evaluation device 1. With this configuration, speech by people other than the subject can be excluded, thereby ensuring privacy. The terminal 2 may also have a voice recognition function, and may be configured to transcribe the voice data into text data and transmit the data to the vocabulary evaluation device 1.

The input unit 22 is a functional unit that accepts input from a user, for example, via a touch panel display or a keyboard. The user is, for example, the owner of the terminal 2. When recording a conversation with a target person, the owner of the terminal 2 may input a command to start recording via the input unit 22.

The input unit 22 also accepts input of words used by the subject, input by the user. The user listens to the speech of the subject, identifies the words used, and inputs them into the terminal 2. In this case, the input unit 22 may accept free input from the user, or may be configured to accept selected input from words displayed on the display unit 23. In addition to the input of words, the input unit 22 may also accept input of time information, location information, environmental information, etc., when the words were used.

The display unit 23 is a functional unit that mainly displays the language acquisition status of the subject. The language acquisition status is received from the vocabulary evaluation device 1 via an appropriate communication processing unit. The acquisition status of the same subject can be displayed on each display unit 23 of the multiple terminals 2, that is, the acquisition status of the subject can be shared among collaborators.
The display unit 23 is realized by, for example, a liquid crystal display, etc. Examples of screens displayed on the display unit 23 will be described later.

The location information acquisition unit 24 is a functional unit that acquires location information uttered by the subject. The location information acquisition unit 24 acquires location information of the terminal 2 when sound is collected by the sound collection unit 21, for example. The location information acquisition unit 24 may specify the location using a GNSS (Global Navigation Satellite System) receiver, or may specify the location using other means, such as latitude and longitude information acquired by a separate communication processing unit or an IP address owned by the terminal 2. The location information acquisition unit 24 may also acquire location information of the terminal 2 when a word is input into the input unit 22 of the terminal 2. Furthermore, the location information acquisition unit 24 may acquire information related to location when the information is input by the user via the input unit 22.

Functional configuration of the vocabulary evaluation device 1 The vocabulary evaluation device 1 is composed of a storage medium such as a memory, a processor, a communication module, an input/output interface, etc., and the processor executes a computer program recorded in the storage medium to realize the functional blocks shown in Fig. 2. The storage medium is a computer-readable recording medium, and may include storage devices such as RAM (random access memory), ROM (read only memory), disk drives, SSDs (solid state drives), and flash memories. Here, non-transitory storage devices such as ROM, disk drives, SSDs, and flash memories may be included in the vocabulary evaluation device 1 as separate storage devices distinct from the memory.

The vocabulary evaluation device 1 has the above-mentioned hardware configuration, and mainly includes a word collection unit 11, a voice processing unit 12, an environment acquisition unit 13, an analysis unit 14, a display control unit 15, and a storage unit 50. Note that part or all of the configuration of the vocabulary evaluation device 1 may be realized by the hardware configuration of the terminal 2.

The vocabulary evaluation device 1 may be a computer program that is incorporated into a specific application, or may be provided as an API (application programming interface) that can be applied to a specific application.

The storage unit 50 is a functional unit that stores information necessary for analyzing the subject's vocabulary acquisition status. The storage unit 50 stores, for example, a database of the language and speech to be analyzed. The database stores the parts of speech and categories of the words together. The database may also store the difficulty level for each word. The database may also store the standard age or target age for acquiring each word. The storage unit 50 may also store information on words that are closely related to the word, so-called collocations. The database may also store information necessary for evaluating the accuracy of grammar in the collected conversation of the subject. This database is configured to be appropriately updated according to information from, for example, a university or research institute 110. The storage unit 50 also stores information necessary for evaluating the amount of speech, pronunciation, frequency, volume, and intonation. This information is, for example, information for comparing with data collected from the subject, such as standard numerical values for each item.

In addition, instead of being stored in the storage unit 50, the above-mentioned database may be stored in an appropriate external device, and the vocabulary evaluation device 1 may acquire information by accessing the external device. The external device may be, for example, a server managed by a university or research institute 110.

The storage unit 50 may also store in advance information for identifying the voice of the subject. This information is, for example, voiceprint information. The voiceprint identification unit 121, which will be described later, refers to this information and extracts the subject's voice from the collected sound source.

The storage unit 50 may also store the identification information of the terminal 2 in association with the owner of the terminal 2. In addition, when the sound collection device 3 is fixed to a predetermined location, the storage unit 50 may store the identification information of the sound collection device 3 in association with the location information of the sound collection device 3.

The memory unit 50 may record information acquired by the word collection unit 11, the voice processing unit 12, or the environment acquisition unit 13. The memory unit 50 may also record the results of analysis by the analysis unit 14 as history. The analysis results include, for example, the amount of speech, pronunciation, frequency, volume, intonation, etc.

The word collection unit 11 is a functional unit that collects words contained in the target person's utterance via the terminal 2.
The word collection unit 11 includes a voice receiving unit 111 and an input information receiving unit 112 .

The voice receiving unit 111 acquires a sound source including the speech of the target person. When only the speech of the target person is extracted by the terminal 2, the voice receiving unit 111 may receive only the speech of the target person. The voice receiving unit 111 may also receive data in which the sound source has been converted into text data by the terminal 2.

The input information receiving unit 112 receives information input via the input unit 22 of the terminal 2. The input information receiving unit 112 receives, for example, the words used by the subject and the time of use input by the user.

The voice processing unit 12 is a functional unit that processes the sound source acquired by the voice receiving unit 111 and extracts words used by the subject. The voice processing unit 12 includes, for example, a voiceprint identification unit 121 and a voice recognition unit 122.

The voiceprint identification unit 121 refers to the memory unit 50 and identifies the speech of the subject from the sound source acquired by the voice receiving unit 111. With this configuration, it is possible to reliably analyze only the content of the speech of the subject.

The voice recognition unit 122 recognizes the sound source acquired by the voice receiving unit 111 and extracts words contained in the sound source. The words extracted by the voice recognition unit 122 are used by the analysis unit 14 to analyze the learning status.

The environment acquisition unit 13 is a functional unit that acquires the environment in which the utterance was made based on information from the terminal 2. For example, the environment acquisition unit 13 acquires identification information of the terminal 2 that picked up the utterance. The terminal 2 is held, for example, by a person who is with the subject. Therefore, acquiring the identification information of the terminal 2 that picked up the utterance can provide information on who the subject was with when the utterance was made. Children, who are assumed to be one example of subjects, are likely not to act alone but to act with a limited number of adults. Therefore, by having a limited number of specific adults carry the terminal 2 in advance and having the terminal 2 with clear identification information pick up the sound, it is possible to easily determine which of the limited number of adults the conversation was with.

The environment acquisition unit 13 may acquire location information of the terminal 2. By acquiring location information of the terminal 2 that picked up the speech, it is possible to determine the location where the subject spoke. In addition, the environment acquisition unit 13 may acquire identification information or location information of the sound collection device 3.

The analysis unit 14 is a functional unit that analyzes the word acquisition status of a subject or information contained in the speech.
The analysis unit 14, for example, determines the acquisition status of each word. For example, the analysis unit 14 determines that the more times a word is used, the higher the acquisition level of the word. The analysis unit 14 may also take into consideration whether the word is used well in determining the acquisition level of the word by analyzing the usage status of the concatenated phrase of the word and the speech content before and after the word and determining the accuracy of the grammar. The analysis unit 14 may determine whether each word has been acquired. The analysis unit 14 may determine that a word with an acquisition level of a predetermined level or higher has been acquired.

The analysis unit 14 analyzes the subject's word acquisition status for each environment in which the word was collected.
The analysis unit 14 assumes that words collected on the same terminal 2 were spoken in the same environment, and analyzes the word acquisition status for each piece of identification information of the terminal 2. It can be assumed that utterances collected on the same terminal 2 were made when the user was with the same collaborator, or in a conversation with the collaborator. In other words, the identification information of the terminal 2 that collected the word is an example of information that specifies the environment in which the utterance occurred.

The vocabulary used by the subject may differ depending on who they are with or who they are talking to. For example, a subject may be relaxed and use a wide variety of words when talking to their parents, but may be nervous and use only a limited number of words when talking to a teacher. In addition, a subject may use the words they learned that day when talking to a speech-language-hearing therapist, but may not use those words when talking to their parents because they have not yet acquired those words sufficiently. The above is just one example, and speech patterns vary greatly depending on the characteristics of the subject. In this regard, the word acquisition status can be evaluated more accurately by analyzing the acquisition status for each identification information of terminal 2.

The analysis unit 14 may compare sound sources collected by multiple terminals 2 for the same subject and determine whether the sound sources are from the same conversation. When the same conversation is collected redundantly via different terminals 2, the analysis unit 14 may not analyze the sound source from one terminal 2. This configuration prevents redundant evaluation and allows for more accurate evaluation. Furthermore, when the same conversation is collected by multiple terminals 2, the analysis unit 14 may analyze it separately from when it is collected by only one terminal 2.

The analysis unit 14 may analyze the word acquisition status for each piece of location information based on the location information of the terminal 2, assuming that words collected at the same place were spoken in the same environment. In other words, the location information of the terminal 2 is another example of information that identifies the environment in which the speech occurred. This is because the speech situation of the subject changes depending on the location. For example, there may be differences in the words used in speech at home and at school. Therefore, the word acquisition status can be more accurately evaluated by analyzing the acquisition status for each piece of location information.

The analysis unit 14 may distinguish between environments based on both the identification information and location information of the terminal 2, and analyze the word acquisition status for each environment. This configuration allows for a detailed analysis of who the subject was with and where they were when the speech occurred.

The analysis unit 14 may change the mode of analysis of the learning status depending on the holder of the terminal 2 that acquired the word. For example, the analysis unit 14 may change the mode of analysis depending on whether the holder of the terminal 2 that acquired the word is a target person or not. More specifically, if the holder of the terminal 2 that acquired the word is not a target person, the analysis unit 14 analyzes the learning status for each identification information of the terminal 2 for the word acquired via the terminal 2, and if the holder of the terminal 2 that acquired the word is a target person, the analysis unit 14 analyzes the learning status for each location information of the terminal 2 for the word acquired via the terminal 2.

When a subject speaks, the person who is speaking to him or her is greatly influenced by who the conversation partner is. Therefore, if the conversation partner can be identified by the identification information of the terminal 2, an analysis can be performed for each identification information of the terminal 2 to appropriately analyze each conversation partner. Furthermore, if the subject's speech is collected on the subject's own terminal 2, it is difficult to identify the conversation partner, so by performing an analysis based on location information instead of an analysis based on the conversation partner, it is possible to perform an analysis that takes the subject's environment into consideration. If the subject is an adult, such as an elderly person, it is highly likely that both collection will be performed on the subject's own terminal 2 and on the collaborator's terminal 2, and a configuration that can combine and analyze both sets of information is even more convenient.

The analysis unit 14 analyzes the acquisition status of the word based on the acquisition status of the word in multiple environments acquired as described above. That is, for example, the analysis unit 14 calculates the acquisition level of the word for each environment, and when the acquisition level in multiple environments is equal to or higher than a predetermined value, the overall acquisition level of the word is set to a predetermined value. In a simpler example, the analysis unit 14 determines that the word has been acquired when the word is used in multiple environments. The analysis unit 14 may also determine that the word has been acquired when the word is collected on all terminals 2 linked to the subject. The analysis unit 14 may also determine that the word has been acquired when the acquisition level of the word is equal to or higher than a predetermined value on all terminals 2 linked to the subject.

When a subject uses a word in multiple environments, it can be said that the subject has reliably mastered the word and is able to use it proficiently. With the above-described configuration of the analysis unit 14, it is possible to more accurately evaluate the word acquisition status.

The analysis unit 14 calculates the number of words that are determined to have been mastered, i.e., the number of acquired words. The analysis unit 14 may also extract the frequency of use of each word by the subject.

The analysis unit 14 may also determine the overall learning status of the subject. For example, the analysis unit 14 may determine the learning status according to the number of words whose learning level is equal to or higher than a predetermined level. The analysis unit 14 may also refer to the difficulty level of each word stored in a language database and determine the learning status according to the difficulty level of words whose learning level is equal to or higher than a predetermined level or that are determined to have been acquired. The analysis unit 14 may also determine the learning status of words for each part of speech. The analysis unit 14 may also refer to an appropriate database in which words and categories are associated with each other to extract categories to which the acquired words belong, and tally up the number of acquired words for each category. Categories are classified according to the type of thing that the word indicates, such as "food" or "vehicle."

The analysis unit 14 may analyze the accuracy of the grammar of the utterance. The accuracy of the grammar may be the accuracy of particles such as "te", "ni", "wo", and "wa", or the accuracy of conjunctions. The analysis unit 14 may calculate the ratio of utterances with correct grammar to the total amount of utterance. Items with many grammatical errors may be displayed on the terminal 2 by the display control unit 15 described later. Note that, since grammar can be evaluated on the premise that the utterance is a sentence of three or more words, the analysis unit 14 may determine how many words the utterance is. In this case, the analysis unit 14 may measure the blank time between the uttered words, and if the next word is uttered within a predetermined blank time, it may determine that the utterance is made as one sentence. The analysis unit 14 may also perform grammar evaluation when the collected utterance is a sentence of three words or more. The analysis unit 14 may also use the speed of speech itself as one measure of language proficiency in the analysis.

Furthermore, the analysis unit 14 may analyze the regularity of the speech of the subject. The analysis unit 14 creates a database of words that are spoken by the subject with a relationship, and analyzes the information on the relationship between words and collocations unique to the speaker. In this case, the analysis unit 14 may perform learning using a deep learning method performed by artificial intelligence, such as a Markov chain. For example, if the subject has a tendency toward autism, the subject may use grammar with low accuracy or may speak in a way that is seemingly disorganized. However, by analyzing the regularity of the speech, it is possible to find the subject's own regularity. With this configuration, the characteristics of autism can be classified in more detail. It may also be possible to screen for pragmatic disorders and evaluate or classify the degree of the disorder. In addition, it is possible to provide more appropriate support for each characteristic. In addition, analyzing the regularity of the speech may help understand the subject's thoughts, such as the causal relationships between things that the subject perceives and the boundaries of each word. As a result, support that is tailored to the subject is possible.

This type of analysis allows for an objective evaluation that is not affected by variations in the skills of the speech-language-hearing therapists performing the analysis. In addition, because vocabulary ability can be grasped without manual intervention, it can also be used for screening developmental tests.

The analysis unit 14 may analyze information included in the voice. The information included in the voice is, for example, the clarity of speech. Specifically, the analysis unit 14 may analyze, for example, the frequency characteristics of the speech of the subject. With this configuration, it is possible to grasp, for example, a tendency that a certain consonant is strong, or that the pronunciation sounds like a different consonant.
The analysis unit 14 compares the frequency characteristics of the speech of the subject with the frequency characteristics of a predetermined speech, and extracts the characteristics of the frequency characteristics of the subject. The predetermined speech may be, for example, an average of the speech of many people, or may be a predetermined model case. The analysis unit 14 may estimate the frequency bands that the subject is good at and bad at according to the frequency characteristics of the speech of the subject. In addition, the analysis unit 14 may estimate that the subject may have hearing loss when the frequency characteristics of the subject are in a predetermined state, and may display this on the terminal 2. The analysis unit 14 may estimate that the subject may have hearing loss when, for example, there is a frequency band that is spoken at a rate above a certain threshold, or a frequency band that is spoken only at a rate below another threshold. With this configuration, it is possible to screen for hearing loss even when a speech-language-hearing therapist performs training without considering the possibility of hearing loss.
Each analysis performed by the analysis unit 14, such as the grammatical accuracy of the speech content, the regularity of speech, and the clarity of pronunciation, may be performed for each environment, as in the above-mentioned analysis of the learning situation. Furthermore, the analysis may be performed for any unit, such as a period, a time period, or a place.

The display control unit 15 is a functional unit that displays the learning status of the subject analyzed by the analysis unit 14 on the terminal 2. The display control unit 15, for example, displays the learning status of the subject on the terminal 2 linked to the subject. For example, when logging in to an account linked in advance to the subject, the learning status may be displayed on the logged-in terminal 2. With this configuration, the learning status of the subject can be easily shared almost in real time between collaborators supporting the same subject.
A specific aspect of the display by the display control unit 15 will be described later.

Screen Examples Examples of screens displayed on the terminal 2 will be described with reference to FIGS.
3A to 3D are examples of screens displayed on a terminal 2 owned by a collaborator, such as a smartphone or a tablet terminal. The terminal 2 is linked to the collaborator via, for example, a predetermined login process, and obtains the detection results from the subject from the storage unit 50 and displays them on the screen.

Screen G1 shown in FIG. 3(a) is a diagram showing an example of a registration screen on which a collaborator, who is a user of terminal 2, registers words used by a subject. Screen G1 is displayed, for example, when a user logs in on terminal 2 and selects a subject as necessary. Screen G1 may also be displayed by selecting a word registration button or the like via a separate menu screen.

Screen G1 mainly displays a search field G11 and a part of speech selection field G12. The search field G11 is a field where the user inputs the word they wish to register. For example, when a word is input into the search field G11 and a search is performed, the database stored in the memory unit 50 is referenced, and identical or similar words are displayed on the screen as search results. When the user selects one word from the search results, the word's usage history is linked to the target user and registered.

Screen G2 shown in FIG. 3(b) is an example of a screen that displays the detection results monthly. Screen G2 displays a graph G21 showing the subject's vocabulary acquisition status, a list G22 of acquired words, and the like. Graph G21 displays the number of predetermined parts of speech (adjectives in FIG. 3(b)) that the subject has been determined to have acquired, the standard number of words to be acquired for the subject's age, or the target number of words, and the ratio of the number of words acquired to this number, i.e., the acquisition rate, as well as a ring that visually represents this ratio. The entire circumference of the ring represents 100%, and the acquisition rate is indicated by a thick partial ring drawn superimposed on the ring.

Screen G3 shown in Figure 3(c) displays a bar graph showing the number of words acquired by month. This display makes it clear to the participant when the subject acquired many words.

The number of words acquired is displayed in a line graph on screen G4 shown in FIG. 3(d). For example, the number of words acquired may be counted by part of speech and graphed.

Screens G1 to G4 allow collaborators to share the progress of the subject's vocabulary acquisition. In other words, those supporting the same subject can share the subject's progress accurately, making support and care for the subject more effective. For example, if the subject is having difficulty understanding something, the collaborator can properly convey the necessary information to the subject by paraphrasing it in the vocabulary that the subject has acquired, i.e., translating it.

Screen G5 shown in Figures 4(a) to (c) is another example of a screen displayed on terminal 2, and is an example of a screen that evaluates the acquisition status in multiple environments for one word and displays the acquisition status. Screen G5 shows the acquisition status of the word "thank you." This embodiment shows an example of evaluation in three environments. For example, it is assumed that the subject's utterances are collected using three terminals 2, and are analyzed and evaluated for each identification information of the collected terminals 2. A graph G51 including a circle showing the target word and the acquisition status is displayed on screen G5.

Figure 4(a) shows a screen when it was determined that "thank you" was used or learned on only one of the three terminals 2. In other words, since use or learning was confirmed in only one-third of the environments, a thick partial ring is displayed superimposed on the circle of graph G51 by one-third of the circumference (120 degrees). Figure 4(b) shows a screen when it was determined that use or learning was done on two of the three terminals 2, and a thick partial ring is displayed superimposed on two-thirds of the circumference (240 degrees). Figure 4(c) shows a screen when it was determined that use or learning was done on all three terminals 2, and a thick ring is displayed on graph G51.

●Processing flow (manual input)
An example of a processing method of the vocabulary evaluation device 1 according to this embodiment will be described with reference to the sequence diagram of FIG. 5. Here, a processing flow will be described for the case where a collaborator manually inputs words used by a subject. In the following description, for convenience, the terminal 2a is described as a terminal that accepts input, and the terminal 2b is described as a terminal that displays the analysis results from the vocabulary evaluation device 1, but the terminals 2a and 2b may be reversed, or one terminal 2a or 2b may perform both processes. In addition, before performing the following processes, a login process for authenticating each user may be performed on the terminal 2 as appropriate.

The terminal 2a transmits identification information of the terminal 2a to the vocabulary evaluation device 1 (step S11). The terminal 2a also transmits information on the subject who will register the used word (step S12). Note that if there is only one subject linked to the account logged into the terminal 2a, step S12 can be omitted. Next, the input unit 22 accepts the input of the word, and transmits it from the terminal 2a to the vocabulary evaluation device 1 (step S13). Note that steps S11 to S13 are in no particular order.

Next, the vocabulary evaluation device 1 analyzes the subject's word acquisition status using the analysis unit 14 (step S14).

The vocabulary evaluation device 1 then records the information received in steps S11 to S13 and the analysis results in the memory unit 50 (step S15).

The vocabulary evaluation device 1 transmits display information of the analysis result to the terminal 2a or to a terminal 2b different from the terminal 2a that received the input (step S16), and the terminal 2a or the terminal 2b displays it (steps S17 and S18). Note that the display process step S16 on the terminals 2a and 2b is performed in response to a display command (not shown) from each of the terminals 2a and 2b, for example.

● Processing flow (audio pickup)
6 is a diagram showing an example of a flow of extracting and analyzing words used by a subject from a collected sound source. Note that the same processes as those in FIG. 5 are given the same reference numerals and the description thereof will be omitted.

In the process flow of FIG. 6, following steps S11 and S12, the sound pickup unit 21 of the terminal 2 picks up the speech (step S23). Steps S11, S12, and S23 are in no particular order. The picked up speech is transmitted to the vocabulary evaluation device 1.

Then, the voiceprint identification unit 121 of the vocabulary evaluation device 1 extracts the speech of the subject from the sound source (step S24). Next, the voice recognition unit 122 of the vocabulary evaluation device 1 extracts words used in the speech of the subject (step S25). Next, the processes of steps S14 to S18 are executed as in FIG. 5.

<Effects of this embodiment>
According to the present invention, it is possible to accurately evaluate the vocabulary of a subject and the information contained in the speech. More specifically, it is possible to analyze the word acquisition status of the subject for each environment in which the speech is made, and accurately determine whether the subject can use the word in any environment. Furthermore, the analysis results are easily shared almost in real time with the collaborators who are supporting the subject in each environment. With this configuration, each collaborator can provide appropriate support by understanding the state of the subject in a different environment. As a result, the subject can receive comfortable and highly effective support.

<Supplementary Notes on the Present Embodiment>
While the present invention has been described above based on the embodiments, the present invention is not limited to the above-described embodiments, and it goes without saying that various modifications and applications are possible within the scope of the gist of the present invention.

In addition to the above, the vocabulary evaluation system 100 may extract the subject's voice volume and estimate the subject's physical condition and strength. When the subject's speech volume gradually decreases, the analysis unit 14 may detect a decrease in the subject's abdominal pressure and thus a decline in physical strength. The analysis unit 14 may also detect temporary poor physical condition.

In addition to the above, the vocabulary evaluation system 100 can also detect the psychological state of the subject. In this case, the analysis unit 14 of the vocabulary evaluation device 1 may detect the intonation of the collected voice, and if the magnitude of the intonation is equal to or less than a predetermined value, it may be assumed that the subject has a tendency to be depressed. This is because, when a person has a tendency to be depressed, they tend to speak in an impersonal manner, resulting in a smaller intonation of words. The width of the intonation is, for example, analyzed by frequency analysis of the voice for each predetermined length, and if the frequency bandwidth having a specific magnitude is equal to or less than a threshold value, it is assumed that the subject has a tendency to be depressed. This threshold value may be determined based on the bandwidth calculated by continuously collecting the speech of the subject and calculating the bandwidth at different points in time. The analysis unit 14 may also detect the progression of dementia based on the magnitude of the intonation.

The analysis unit 14 may also refer to a database that stores in advance the positivity of concepts represented by words, and analyze whether the subject's speech gives a positive, bright, or forward-looking impression, or a negative, gloomy, or pessimistic impression. Positivity may be stored as a numerical value, in addition to being given as a three-choice option of positive, negative, or other. The analysis unit 14 may calculate the ratio of positive words to negative words, and calculate an index of positivity for the entire speech. It can be estimated that a person whose speech contains many positive words or words with a high degree of positivity is in a positive psychological state. With this configuration, appropriate support can be provided in accordance with the psychological state.

The analysis unit 14 may analyze the recorded speech of multiple people instead of a specific subject, and estimate the positivity of the entire conversation. The more positive words are used throughout the conversation, the more likely it is that the conversation is proceeding in a positive atmosphere. For example, when group work is being carried out, the speech of the group can be analyzed and the atmosphere of the group work can be quantified. In this case, the analysis unit 14 may analyze only the volume of the speech instead of the speech, or may analyze only sounds of a specific frequency. With this configuration, the privacy of the conversation is maintained.

When analyzing the content of speech by multiple people, the analysis unit 14 may estimate the compatibility between the people participating in the conversation based on the identification information of the people participating in the conversation and the positivity of the words or the overall speech. In other words, people who converse in a positive atmosphere can be estimated to have good compatibility and be on good terms. With this configuration, compatibility can be estimated while maintaining privacy, without a person having to directly listen to the conversation and evaluate compatibility. This configuration is also useful when it is necessary to understand the psychological state of users, for example, in day care centers or special nursing homes for the elderly. Of course, it can also be applied to daycare centers, kindergartens, schools, hospitals, etc.

Furthermore, the analysis unit 14 may estimate the personality of the subject based on the content of the subject's utterances collected over a long period of time, rather than being limited to the subject's temporary psychological state. In particular, the analysis unit 14 can estimate whether the subject has a relatively positive or negative personality. In addition, by estimating the personalities of many people, the system can be applied to a matching device for marriage partners, lovers, friends, team formation, and other purposes, which extracts and presents people with compatible personalities. Note that in this embodiment, personalities are broadly classified as positive or negative, but a more detailed personality analysis may be performed by subdividing the impression expressed by each word.

Furthermore, the analysis unit 14 can calculate the positivity of sentences in books, for example, in advance and store them in a database. That is, the configuration of the analysis unit 14 allows the positivity of a book to be quantified. Furthermore, books recommended to the subject can be extracted by referring to the subject's psychological state analyzed by the analysis unit 14 and the positivity of the book. In this case, books with positivity close to the subject's psychological state may be extracted, or if the subject's psychological state is negative, books with positive content may be extracted and presented as recommendations. With this configuration, the subject can select a book that matches their own state and personality.

The above-mentioned configuration can also be applied to a configuration in which the positivity of song lyrics is calculated in advance and stored in a database, and recommended songs are extracted based on the psychological state of the subject. For example, by linking with a music distribution service, it is possible to automatically play songs that match the psychological state of the subject. In addition to books and songs, it is possible to match with any object using words, such as television programs, movies, video works, radio, and posts on Clubhouse (registered trademark) and Twitter (registered trademark), etc.

Furthermore, by collecting a large amount of information about the subjects' personalities and psychological states, it can also be used as marketing data. In this case, by analyzing this data in conjunction with region, generation, gender, and other attributes, it is possible to observe and analyze people's conditions on a macro level.

REFERENCE SIGNS LIST 100 Vocabulary evaluation system 1 Vocabulary evaluation device 11 Word collection unit 111 Voice reception unit 112 Input information reception unit 12 Voice processing unit 13 Environment acquisition unit 14 Analysis unit 15 Display control unit 50 Storage unit 2 Terminal 2a Terminal 2b Terminal 21 Sound collection unit 22 Input unit 23 Display unit 24 Position information acquisition unit 3 Sound collection device

Claims (10)

A word collecting means for collecting words included in the speech of a subject via a terminal communicably connected thereto;
an environment acquisition means for acquiring an environment in which the utterance was made based on information from the terminal;
an analysis means for analyzing the word acquisition status of the subject for each of the environments;
having
Vocabulary assessment device.
the word collecting means includes a voice receiving means for acquiring a sound source including the speech, the sound being collected by the terminal;
The vocabulary evaluation device further includes a speech recognition unit that recognizes the speech source and extracts the words included in the speech source.
The vocabulary assessment device of claim 1.
A storage means for storing in advance information for identifying the voice of the subject;
a voiceprint identification means for referring to the storage means and identifying a speech made by the subject from a sound source acquired by the voice receiving means;
Further comprising
The vocabulary assessment device according to claim 2.
the word collecting means has an input information receiving means for receiving the words input via the terminal;
The vocabulary assessment device of claim 1.
The environment acquisition means acquires identification information of the terminal,
the analysis means analyzes the acquisition status of the words collected by the same terminal as if the words were spoken in the same environment for each of the identification information of the terminal.
The vocabulary assessment device of claim 1.
The environment acquisition means acquires location information of the terminal,
the analysis means analyzes the acquisition status of the words for each piece of location information, assuming that the words collected at the same place were uttered in the same environment based on the location information;
The vocabulary assessment device of claim 1.
the word collection means collects the words included in the utterance of one of the subjects via the multiple terminals;
Further comprising a storage means for storing at least the identification information of the terminal and the owner of the terminal in association with each other,
The environment acquisition means acquires identification information and location information of the terminal,
The analysis means includes:
If the possessor of the terminal through which the word was acquired is not the target person, the acquisition status of the word acquired through the terminal is analyzed for each identification information of the terminal;
When the possessor of the terminal through which the word was acquired is the subject, the acquisition status of the word acquired through the terminal is analyzed for each location information of the terminal.
The vocabulary assessment device of claim 1.
The device further includes a display control means for displaying, on the terminal, the learning status for each of the environments of the words analyzed by the analysis means.
A vocabulary evaluation device according to any one of claims 1 to 7.
A word collection step of collecting words included in the target person's utterance via a terminal communicatively connected to the target person;
an environment acquisition step of acquiring an environment in which the utterance was made based on information from the terminal;
an analysis step of analyzing the learning status of the word in the subject for each of the environments;
including,
Vocabulary assessment methods.
A word collection command for collecting words included in the speech of the subject via a terminal communicably connected thereto;
an environment acquisition command for acquiring an environment in which the utterance was made based on information from the terminal;
An analysis command for analyzing the word acquisition status of the subject for each environment;
to cause a computer to execute
Vocabulary assessment program.

Images