WO2023152942A1

WO2023152942A1 - Language learning system

Info

Publication number: WO2023152942A1
Application number: PCT/JP2022/005560
Authority: WO
Inventors: 伸雄椿
Original assignee: 株式会社ムゴン
Priority date: 2022-02-14
Filing date: 2022-02-14
Publication date: 2023-08-17

Abstract

[Problem] To provide a language learning system which makes it possible to efficiently improve the foreign language listening and speaking abilities of a user. [Solution] This language learning system makes it possible to efficiently improve the foreign language listening and speaking abilities of a user by comprising: a translation unit 17 which translates text of a first language into text of a second language; a voice data generation unit 11 which generates second language text voice data that is voice data of the text of the second language; and a determination unit 22 which compares the second language text voice data and pronunciation data that is voice data of the user's pronunciation.

Description

language learning system

The present invention relates to a system for learning languages.

For example, Patent Literature 1 listed below discloses a system for learning a language (language learning system).

JP 2017-191151 A

With conventional language learning systems, it is difficult to improve the user's foreign language listening and speaking ability.

SUMMARY OF THE INVENTION An object of the present invention is to provide a language learning system capable of efficiently improving a user's foreign language listening and speaking ability.

According to the language learning system in the first aspect of the present invention,
a translation unit that translates sentences in a first language into sentences in a second language;
a speech data generation unit that generates second language sentence speech data, which is speech data of the sentence in the second language;
It is characterized by comprising a judgment unit for comparing the second language text speech data with pronunciation data, which is speech data of pronunciation by the user.

Preferably,
The speech data generation unit generates second-language sentence-similar speech data, which is speech data similar in pronunciation to sentences in the second language,
The determination unit compares the second-language sentence voice data and the second-language sentence-similar voice data with the pronunciation data to determine the accuracy of the pronunciation data.

Preferably,
comprising a dividing unit that divides the sentence in the second language to generate sentence fragments;
The speech data generation unit generates sentence piece speech data, which is speech data of the sentence piece,
The determination unit is characterized by comparing the sentence piece speech data and the pronunciation data.

Preferably,
The voice data generation unit generates text segment-similar voice data, which is voice data whose pronunciation is similar to the text segment,
The determination unit is characterized in that the accuracy of the pronunciation data is determined by comparing the sentence piece speech data and the sentence piece similar speech data with the pronunciation data.

Preferably,
It is characterized by comprising a voice data output instructing section for instructing the terminal to output the second language text voice data as voice.

Preferably,
The apparatus is characterized by comprising a voice data output instructing section for instructing the terminal to output the second language text voice data as voice and to output the text piece voice data as voice.

a question sentence database in which a plurality of first language sentences are stored;
a question selection unit that selects one sentence in the first language from the question sentence database;
a screen generation unit that generates a display screen that displays sentences in the first language selected by the question selection unit;
A timer that sends a timer signal (first timer signal) to the question selection unit every time a predetermined time (first predetermined time) elapses,
The question selection section performs the selection when the timer signal is input.

According to the language learning system according to the present invention, it is possible to efficiently improve the user's foreign language listening and speaking ability.

1 is a block diagram illustrating the configuration of a language learning system according to Example 1 of the present invention; FIG. 4 is a schematic diagram showing an example of a first display screen in Example 1 of the present invention; FIG. FIG. 4 is a schematic diagram showing an example of a second display screen in Example 1 of the present invention; 4 is a flowchart illustrating an example of the operation (during user learning) of the language learning system according to Example 1 of the present invention; 4 is a flowchart illustrating an example of the operation of the language learning system according to Example 1 of the present invention (at the time of linking similar words). FIG. 2 is a block diagram illustrating the configuration of a language learning system according to Example 2 of the present invention; FIG. 3 is a block diagram illustrating the configuration of a language learning system according to Example 3 of the present invention; FIG. 11 is a conceptual diagram illustrating the timing of timer signal transmission in Example 3 of the present invention;

A language learning system according to the present invention is a system used by a user to learn a foreign language.
Hereinafter, the language learning system according to the present invention will be described with reference to the drawings.

[Example 1]
FIG. 1 is a block diagram illustrating the configuration of the language learning system according to this embodiment. As shown in FIG. 1, the language learning system (language learning system 1) according to the present embodiment is provided in a server 3 that can mutually communicate with each terminal 2 by using a network line L such as the Internet. In addition, this terminal 2 refers to a terminal capable of performing communication using the network line L, such as a PC, a smart phone, a tablet, and the like.

The language learning system 1 includes a speech data generation unit 11, a communication control unit 12, a question sentence database 13, a question selection unit 14, a screen generation unit 15, a word database 16, a translation unit 17, a division unit 18, a storage unit 19, and voice data. An output instruction unit 20, a pronunciation similarity database 21, a determination unit 22, and a transmission instruction unit 23 are provided.

Of these, the communication control unit 12 transmits and receives each piece of information to and from the terminal 2 via the network line L. The description will be omitted below.

The question sentence database 13 is a database that stores one or more sentences in the first language (for example, everyday conversation (e.g. I feel like something is wrong)) (the more the better). Note that the first language is mainly assumed to be the native language of the user.

In this embodiment, it is assumed that the first language is one language, but the present invention is not limited to this, and one or more sentences are stored for each of a plurality of languages. may be

The question selection unit 14 selects one sentence from the sentences stored in the question sentence database 13 .

The screen generation unit 15 generates at least the first display screen D1 and the second display screen D2 depending on the situation. These generated display screens are transmitted to each terminal 2 . Each terminal 2 displays these display screens received from the server 3 .

For example, as shown in FIG. 2, on the first display screen D1, sentences W selected by the question selection unit 14, one first playback button P1, a plurality of second playback buttons P2, one voice input start button V , one next question change button C1, and one previous question change button C2 are displayed. The functions of these buttons will be described later.

For example, as shown in FIG. 3, the second display screen D2 allows the system administrator or the user to enter words (parts of speech) that are similar in pronunciation to each other. It should be noted that whether the pronunciations of words are similar or not is determined at the discretion of the system administrator or the user.

The word database 16 is a database in which words (parts of speech) of a plurality of languages including a first language and a second language are associated with the same meaning and stored. The word database 16 also stores voice data of each word. It should be noted that the second language is mainly assumed to be a foreign language for the user.

Based on the information stored in the word database 16, the translation unit 17 translates the sentences in the first language selected by the question selection unit 14 into sentences in the second language (the first language is English and the second language is Japanese). If it is a word, "I feel like something is wrong" → "I feel something is wrong", etc.).

Note that both the first language and the second language can be designated by the user on the first display screen D1 displayed on the terminal 2 (or separately on the third display screen). (For example, if the first language is English and the second language is Japanese, the translation unit 17 translates English sentences into Japanese sentences).

The division unit 18 divides the sentences in the second language translated by the translation unit 17. The place to delimit should be either between words, preferably all between phrases (for example, in the case of "something different", "something", "different", "I feel") , or "something different", "I feel", etc. Conversely, do not use "na", "nka difference", "u", "kisu", "ru", etc.). Each segmented portion is hereinafter referred to as a "sentence fragment". That is, the dividing unit 18 divides the sentence in the second language to generate sentence fragments.

The storage unit 19 stores the sentences in the second language translated by the translation unit 17 and the sentence fragments divided by the division unit 18 together with audio data (the audio data is stored in the audio data generation unit 11 described later). generated in ).

A first play button P1 displayed on the first display screen D1 is a button assigned to a sentence in the second language translated by the translation unit 17, and a plurality of second play buttons P2 are displayed by the division unit 18. It is a button assigned to each of the divided sentence fragments (the number of second playback buttons P2 is equal to the number of sentence fragments, so the number is changed each time). However, the sentences in the second language and the sentence fragments themselves are not displayed on the first display screen D1.

Note that if the sentence translated by the translation unit 17 is too short to be segmented (for example, only one word), the division unit 18 does not perform division processing, and the first display screen D1 generated by the screen generation unit 15 is not provided with the second play button P2 (the first play button P1 is provided).

The pronunciation similarity database 21 stores words that are similar in pronunciation to each other in association with each other for a plurality of languages including the second language. It may be that 3 or more words are associated with each other.

The information of the pronunciation similarity database 21 can be input from the second display screen D2. This second display screen D2 may allow viewing and input only from a terminal owned by the system administrator, or may allow viewing and input from a terminal owned by the system administrator and terminals owned by general users. .

The speech data generation unit 11 extracts speech data of words constituting sentences in the second language translated by the translation unit 17 (hereinafter referred to as “second language sentence words”) from the word database 16 and combines them, Speech data of this second language sentence (hereinafter referred to as "second language sentence speech data") is generated.

In addition, the speech data generation unit 11 extracts speech data of words whose speech is similar to words in the second language sentence (hereinafter referred to as “similar words in the second language sentence”) from the pronunciation similarity database 21 and combines them to obtain the above-mentioned Speech data whose pronunciation is similar to sentences in the second language (hereinafter referred to as “sentence-similar speech data in the second language”) is generated.

Further, the speech data generation unit 11 extracts from the word database 16 speech data of words (hereinafter referred to as “sentence fragment words”) constituting the sentence fragments divided by the division unit 18, and combines them to generate the sentence fragments. Speech data (hereinafter referred to as "sentence fragment speech data") is generated (same operation is performed for each of a plurality of sentence fragments).

Then, the speech data generation unit 11 extracts and combines the speech data of words whose speech is similar to the words in the sentence fragment (hereinafter referred to as “words similar to the sentence fragment”) from the pronunciation similarity database 21, so that the pronunciation is similar to the sentence fragment. Speech data (hereinafter referred to as "sentence segment-similar speech data") is generated (same operation is performed for each of a plurality of sentence segments).

By the way, the relationship (phonetic similarity) between the words in the second language sentence and the similar words in the second language sentence is, for example, if the word in the second language sentence is "great", the similar word in the second language sentence is " "Sugao" and so on.

The second-language-sentence-similar speech data includes speech data of sentences generated by combining speech data of second-language-sentence words and speech data of second-language-sentence-similar words (for example, "second-language sentence word/second language sentence similar word/second language sentence word", etc.).

Furthermore, it is conceivable that a plurality of second-language-sentence-similar words exist for one second-language-sentence word. Therefore, it is often the case that a plurality of types of second-language-sentence-similar speech data are generated for one second-language-sentence speech data.

The relationship between the second-language sentence speech data and the second-language sentence-similar speech data and the relationship between the second-language sentence words and the second-language sentence-similar words described above are described in , and the relationship between sentence fragment words and sentence fragment-similar words.

When the voice data output instruction unit 20 receives information that the first playback button P1 has been clicked (by the user) on the first display screen D1 displayed on the terminal 2, the voice data output instruction unit 20 provides instruction information for outputting the voice data of the second language text. to generate The instruction information is transmitted to the terminal 2 together with the voice data of the second language sentence, and the terminal 2 receives the voice data of the second language sentence from the sound output unit (speaker or earphone connected to the terminal 2; not shown). Output audio.

Also, the voice data output instruction unit 20 receives information that any one of the plurality of second playback buttons P2 has been clicked (by the user) on the first display screen D1 displayed on the terminal 2. Then, instruction information is generated for outputting the text piece voice data of the text piece to which the clicked second reproduction button P2 is assigned among the plurality of text pieces. The instruction information is transmitted to the terminal 2 together with the sentence fragment speech data, and the terminal 2 that receives the instruction information outputs the second language sentence speech data from the sound output unit.

That is, the voice data output instructing unit 20 instructs the terminal 2 to output the second language text voice data as voice and to output the text segment voice data as voice.

Also, when the voice input start button V displayed on the first display screen D1 is clicked, the sound input unit (microphone; not shown) of the terminal 2 can receive the user's pronunciation. In this state, the user pronounces according to the sound output from the sound output unit of the terminal 2 (imitation of the output sound). This pronunciation is input to the sound input unit, converted into audio data (hereinafter referred to as “pronunciation data”) in the terminal 2 , and the audio data is transmitted to the server 3 .

The determination unit 22 compares the second-language sentence voice data with the pronunciation data received from the terminal 2 when the first playback button P1 is clicked. If the second language sentence speech data and the pronunciation data do not match as a result of comparison, the second language sentence similar speech data and the pronunciation data are compared. Furthermore, the judging section 22 judges the accuracy of the pronunciation data from the results of these comparisons.

Then, when one second playback button P2 is clicked, the determination unit 22 compares the sentence piece speech data and the pronunciation data of the sentence piece assigned to the second playback button P2. Also, if the result of comparison between the sentence piece speech data and the pronunciation data is that they do not match, the sentence piece-similar speech data of the sentence piece assigned to the second playback button P2 is compared with the pronunciation data. Furthermore, the judging section 22 judges the accuracy of the pronunciation data from the results of these comparisons.

It is preferable that the accuracy determined by the determination unit 22 be in the following three stages.
・The second language sentence speech data (or sentence segment speech data) matches the pronunciation data → High accuracy ・The second language sentence similar speech data (or sentence fragment similar speech data) matches the pronunciation data → Accuracy medium・The second language sentence speech data and the second language sentence similar speech data (or the sentence piece speech data and the sentence piece similar speech data) do not match the pronunciation data → accuracy is low

The transmission instruction unit 23 generates instruction information for outputting the accuracy determined by the determination unit 22 . More specifically, it instructs the terminal 2 to perform one or both of reflecting on the display screen and outputting a specific sound. The terminal 2 that has received the instruction information notifies the user of the degree of accuracy by outputting according to the instruction information.

Note that the above-mentioned reflection may be, for example, reflected in the display of the first play button P1 or the second play button P2 linked to the relevant sentence or sentence fragment on the first display screen (high accuracy and medium, change one or more of its properties such as color, shape, and size; or erase the display of the button itself; Alternatively, a new display screen (fourth display screen) representing the degree of accuracy determined by the determination unit 22 may be generated by the screen generation unit 15 and displayed.

When the accuracy is equal to or higher than the threshold value (for example, when the accuracy is high or medium in the case of the three levels described above), the determination unit 22 determines that the pronunciation of the sentence or sentence fragment is In everyday conversation with native speakers of two languages), it is considered to be no problem.

By the way, the accuracy is not limited to the three levels described above, and may be calculated as a percentage, for example. In any case, the judging section 22 is provided with a threshold value, and the judging section 22 regards the sentence or document fragment as having no problem if the degree of accuracy is equal to or higher than the threshold value.

In conventional language learning, if the learner's pronunciation does not exactly match the pronunciation of the native speaker, it is judged to be a mistake. I often lost it. Therefore, in this embodiment, even if the learner's pronunciation does not completely match the native's pronunciation, it is judged that there is no problem as long as daily conversation is established.

It should be noted that, in this embodiment, the text translated by the translated text and the text fragments divided by the dividing unit 18 are not displayed in characters on the first display screen D1.

When the next question change button C1 on the first display screen D1 is clicked and the information is transmitted from the terminal 2 to the server 3, the question selection unit 14 selects a different sentence from the sentences stored in the question sentence database 13. Select sentences. After that, the translation unit 17, the division unit 18, the voice data generation unit 11, the determination unit 22, the voice data output instruction unit 20, and the transmission instruction unit 23 perform the same processing as the processing already described.

When the previous question change button C2 on the first display screen D1 is clicked and the information is transmitted from the terminal 2 to the server 3, the question selection unit 14 selects the previous question from the sentences stored in the question sentence database 13. The sentence is selected again, and then the translation unit 17, the division unit 18, the voice data generation unit 11, the determination unit 22, the voice data output instruction unit 20, and the transmission instruction unit 23 perform processing similar to the processing already described. conduct.

The above is the description of the configuration of the language learning system 1. An example of the operation of the language learning system 1 when the user learns using the language learning system 1 will be described below with reference to the flowchart of FIG. Note that steps SA1 to SA12 in FIG. 4 describe operations by the language learning system 1 provided in the server 3, and steps SB1 to SB12 describe operations by the terminal 2. FIG.

In step SA1, the question selection unit 14 selects one sentence from the first language sentences stored in the question sentence database 13. The selected text is incorporated into the first display screen D1 generated by the screen generator 15. FIG. This first display screen D1 is transmitted from the server 3 to the terminal 2 .

At step SA2, based on the information stored in the word database 16, the translation unit 17 translates the sentences in the first language selected by the question selection unit 14 into sentences in the second language. The translated second language sentence is stored in the storage unit 19 .

At step SA3, the dividing unit 18 divides the sentence in the second language. The divided text pieces are stored in the storage unit 19 .

At step SA4, the speech data generator 11 extracts speech data of the words of the second language sentence from the word database 16 and combines them to generate speech data of the second language sentence, and extracts speech data of the second language sentence words from the pronunciation similarity database 21. Second-language sentence-similar speech data is generated by extracting and combining speech data of sentence-similar words.

Further, at step SA4, the speech data generation unit 11 extracts and combines speech data of sentence fragment words from the word database 16 to generate sentence fragment speech data, and also extracts sentence fragment speech speech data from the pronunciation similarity database 21 and extracts sentence fragment speech speech data from the similar sentence fragment words. Data is extracted to generate sentence segment-similar speech data.

At step SB1, the user confirms the first display screen D1 transmitted to the terminal 2 at step SA1, and clicks the first play button P1 or the second play button P2 in the first display screen D1. When the first play button P1 is clicked, the process proceeds to step SB2, and when the second play button P2 is clicked, the process proceeds to step SB3.

At step SB2, information is sent to the server 3 indicating that the first playback button P1 has been clicked. At step SB3, information is sent to the server 3 to the effect that the second playback button P2 has been clicked.

At step SA5, information is received from the terminal 2 indicating that the play button has been clicked. When the information transmitted from the terminal 2 is received at step SB2, the process proceeds to step SA6, and when the information transmitted from the terminal 2 is received at step SB3, the process proceeds to step SA7.

At step SA6, the voice data output instruction unit 20 generates instruction information for outputting the second language text voice data. The instruction information and the second language sentence voice data generated in step SA4 are transmitted to the terminal 2. FIG.

In step SA7, the voice data output instruction unit 20 generates instruction information for outputting the text segment voice data of the text segment to which the clicked second playback button P2 is assigned. The instruction information and the sentence fragment speech data generated in step SA4 are transmitted to the terminal 2. FIG.

At step SB4, when the terminal 2 receives the instruction information transmitted from the language learning system 1 at step SA6, the sound output unit (not shown) of the terminal 2 outputs the second language sentence voice data. At step SB5, when the terminal 2 receives the instruction information transmitted from the language learning system 1 at step SA7, the sound output section (not shown) of the terminal 2 outputs sentence fragment speech data.

At step SB6, the user clicks the voice input start button V displayed on the first display screen D1, and the terminal 2 becomes ready to receive voice.

At step SB 7 , the terminal 2 , which is in a voice-acceptable state by the user's pronunciation, accepts the pronunciation.

In step SA8, the judgment unit 22 judges the accuracy of the pronunciation data by comparing each speech data generated in step SA4 with the pronunciation data received from the terminal 2.

In step SA9, the transmission instruction unit 23 generates instruction information for outputting the accuracy determined by the determination unit 22. The instruction information is transmitted from the server 3 to the terminal 2 .

At step SB8, the accuracy is notified to the user by outputting according to the instruction information generated by the transmission instruction unit 23.

At step SB9, the user clicks any button on the first display screen D1. That is, when either of the play buttons P1, P2 (the play button different from the play button clicked in step SB1 is assumed, but the same play button may be used) is clicked again, the process proceeds to step SB10. If the next question change button C1 is clicked, the process proceeds to step SB11, and if the previous question change button C2 is clicked, the process proceeds to step SB12.

At step SB10, information is sent to the server 3 to the effect that one of the playback buttons P1 and P2 has been clicked. At step SB11, information is sent to the server 3 to the effect that the next question change button C1 has been clicked. At step SB12, information is sent to the server 3 to the effect that the previous question change button C2 has been clicked.

Here, when language learning system 1 receives from terminal 2 the information transmitted from terminal 2 in step SB10, language learning system 1 resumes processing from step SA5.

At step SA10, information is received from the terminal 2 to the effect that the change next question button C1 or the change previous question button C2 has been clicked. When the information transmitted from the terminal 2 is received at step SB11, the process proceeds to step SA11, and when the information transmitted from the terminal 2 is received at step SB12, the process proceeds to step SA12.

In step SA11, the question selection unit 14 selects a sentence different from the one stored in the question sentence database 13, and then proceeds to step SA2.

At step SA12, the question selection unit 14 selects the previously selected sentence again from the sentences stored in the question sentence database 13, and then proceeds to step SA2.

The above is an explanation of an example of the operation of the language learning system 1 when the user uses the language learning system 1. An example of the operation of the language learning system 1 when the system administrator or user associates similar words with each other will be described below with reference to the flowchart of FIG. Note that steps SA21 and SA22 in FIG. 5 describe the operation of the language learning system 1 provided in the server 3, and step SB21 describes the operation of the terminal 2. FIG.

At step SA21, the screen generation unit 15 generates the second display screen D2, which is transmitted from the server 3 to the terminal 2.

In step SB21, the system administrator or the user inputs words that are similar in pronunciation from the second display screen D2 displayed on the terminal 2 so as to associate them with each other. The entered information is transmitted from the terminal 2 to the server 3 .

In step SA22, the information received from the terminal 2 and linking words with similar pronunciations is stored in the pronunciation similarity database 21.

The above is an explanation of an example of the operation of the language learning system 1 when the system administrator or user operates the second display screen D2 of the language learning system 1.

According to this embodiment, the language learning system 1 includes the translation unit 17 that translates sentences in the first language into sentences in the second language, the voice data generation unit 11 that generates voice data of the second language text, and the second Since the determination unit 22 is provided for comparing the language sentence voice data and the pronunciation data, the second language sentence voice data and the pronunciation data can be compared, and the user's speaking ability of the second language (that is, the foreign language) can be evaluated. can be improved.

According to this embodiment, the speech data generation unit 11 generates the second-language sentence-similar speech data, and the determination unit 22 compares the second-language sentence-speech speech data and the second-language sentence-similar speech data with the pronunciation data. By doing so, the accuracy of the pronunciation data is determined, so the accuracy of the pronunciation data can be grasped, and the user's ability to speak the second language can be further improved.

According to this embodiment, the language learning system 1 further comprises a segmentation unit 18 for generating sentence fragments, the speech data generation unit 11 produces sentence fragment speech data, and the determination unit 22 produces sentence fragment speech data and pronunciation data. Since the comparison is made between and, for example, sentences in the second language can be learned by segmenting them into clauses, which improves convenience during learning.

According to this embodiment, the speech data generation unit 11 generates sentence piece-similar speech data, and the determination unit 22 compares the sentence piece speech data and the sentence piece-similar speech data with the pronunciation data to obtain the pronunciation data. Since the degree of accuracy is determined, for example, sentences in the second language can be learned by segmenting them into clauses, which improves convenience during learning.

According to this embodiment, since the terminal 2 is provided with the voice data output instruction unit 20 for instructing the terminal 2 to output the voice data of the second language text, the user can read the text in the second language output from the terminal 2. After listening to the voice, it is possible to use vocalization data by imitating the voice, so that the user's ability to hear the second language can also be improved.
be able to.

According to the present embodiment, the terminal 2 is provided with the audio data output instruction section 20 for instructing the terminal 2 to output the second language sentence audio data and to output the sentence fragment audio data as audio. After listening to the speech of sentences and sentence fragments in the second language output from the terminal 2, it is possible to make speech data by imitating the speech, so that the user's hearing of the second language can be performed. Abilities can also be improved.

According to the present embodiment, the sentences are not displayed on the first display screen D1, and only the sound output from the sound output unit of the terminal 2 is transmitted to the user, so that the user can concentrate more on the sound. You can effectively improve your foreign language speaking ability.

In this embodiment, after the question selection unit 14 selects a sentence in step SA1, the translation unit 17 translates the selected sentence into a sentence in the second language in step SA2, and the division unit 18 translates the sentence in the second language in step SA3. A sentence in two languages is divided, and in step SA4, the speech data generation unit 11 generates second language sentence speech data, second language sentence similar speech data, sentence fragment speech data, and sentence fragment similar speech data. However, the present invention is not limited to this form. For example, the operations of steps SA1 to SA4 are completed in advance before step SB1 is started, and each data derived in each step is stored in the storage unit. 19 may be stored.

Also, in this embodiment, the language learning system 1 is provided on the server 3, but the present invention is not limited to such a form. can be This point also applies to the second embodiment.

[Example 2]
The language learning system according to this embodiment is obtained by partially changing the configuration of the language learning system 1 according to the first embodiment. In the following, descriptions of the same configurations as in the first embodiment will be omitted as much as possible, and different parts will be mainly described.

FIG. 6 is a block diagram explaining the configuration of the language learning system according to this embodiment. As shown in FIG. 6, the language learning system (language learning system 1A) according to the present embodiment replaces the screen generator 15 in the configuration of the language learning system 1 according to the first embodiment, and partially changes its functions. A screen generation unit 15A is provided, and an extraction unit 24 is added.

The screen generation unit 15A in this embodiment differs from the screen generation unit 15 in that it does not generate the second display screen D2. It is similar to the generation unit 15 .

Based on the speech data of each word stored in the word database 16, the extraction unit 24 links and extracts words (parts of speech) that have similar pronunciations from each word stored in the word database 16. This extracted information is then stored in the pronunciation similarity database 21 .

Thus, in this embodiment, the system administrator and the user do not need to input from the second display screen D2, and the screen generator 15 does not need to generate the second display screen D2.

[Example 3]
The language learning system according to the present embodiment is obtained by partially changing the configuration of the language learning system 1 according to the first embodiment (or the language learning system 1A according to the second embodiment). In the following, descriptions of the same configurations as those of the first and second embodiments will be omitted as much as possible, and different parts will be mainly described.

FIG. 7 is a block diagram illustrating the configuration of the language learning system according to this embodiment. As shown in FIG. 7, the language learning system (language learning system 1B) according to the present embodiment has the screen Instead of the generation unit 15 (15A), a screen generation unit 15B whose function is partially changed is provided.

Moreover, the language learning system 1B is not provided with the division unit 18 and the voice data output instruction unit 20 described in the first embodiment, but is provided with a timer 25 instead.

The screen generator 15B generates each of the display screens described in Example 1 (Example 2). It is not provided (the text W selected by the question selection unit 14 is displayed).

Accordingly, steps SB1 to SB3, SB6, and SB9 to SB12 described with reference to FIG. 4 in the first embodiment are omitted. Steps SA5-SA7, SA11 and SA12 are also omitted.

The timer 25 selects one sentence from the sentences in the first language stored in the question sentence database 13 by the question selection section 14, and after a first predetermined time elapses, a first timer signal is sent to the question selection section 14. send. After that, the first timer signal is sent to the question selection section 14 each time the first predetermined time elapses. It should be noted that the first predetermined time is preferably set to several seconds.

Further, the timer 25 first selects one sentence from the sentences in the first language stored in the question sentence database 13 by the question selection unit 14, and sets it for a second predetermined time (second predetermined time<first predetermined time). time), a second timer signal is transmitted to the terminal 2 . After that, every time the first timer signal is transmitted to the question selection unit 14, the second timer signal is transmitted to the terminal 2 after the second predetermined time has elapsed (see FIG. 8).

The question selection unit 14 selects one sentence from sentences in the first language stored in the question sentence database 13, as in the first embodiment. Further, in this embodiment, when the first timer signal is input, one sentence is selected from sentences in the first language stored in the question sentence database 13 .

The sentences in the first language selected by the question selection section 14 are incorporated (displayed) in the first display screen D1 generated by the screen generation section 15B. This first display screen D1 is then transmitted from the server 3 to the terminal 2 . This point is the same as the first (or second) embodiment.

In addition, although the first display screen D1 generated by the screen generation unit 15B is provided with a voice input start button V, the user cannot operate this from the terminal 2, and the sound input unit of the terminal 2 It is assumed that the state in which pronunciation can be accepted and the state in which pronunciation cannot be accepted are automatically switched.

The sound input unit of the terminal 2 is in a pronunciation accepting state until the terminal 2 receives the second timer signal after displaying the first display screen D1 generated by the screen generating unit 15B. After receiving the signal, the terminal 2 is in the pronunciation unacceptable state until the terminal 2 newly displays the first display screen D1 generated by the screen generation unit 15B.

Note that the screen generation unit 15B may change properties such as the color of the voice input start button V according to the state of the sound input unit of the terminal 2 so that the user can easily distinguish the state of the sound input unit. .

Then, as in the first embodiment, when the user makes a sound while the terminal 2 is in a state where the sound can be accepted, the sound data is transmitted to the determination unit 22 .

Further, the determination unit 22 that has received the pronunciation data determines the degree of accuracy (if the terminal 2 is in a state where the pronunciation can be accepted and the user does not pronounce, the degree of accuracy will be the lowest value). The determined accuracy indicates that the instruction information generated by the transmission instruction unit 23 (information instructing to perform either or both of reflecting on the display screen and outputting a specific sound) 2, it is output from the terminal 2 and notified to the user.

Note that when the instruction information is set to display the fourth display screen, the screen generation unit 15B generates the fourth display screen showing the accuracy determined by the determination unit 22. FIG.

By configuring the language learning system 1B in this way, in the present embodiment, the question selection unit 14 first selects a sentence, and the screen generation unit 15B incorporating this selects the first display screen D1. The user confirms the first display screen D1 from the terminal 2, thinks for himself the translation of the sentence W displayed on the first display screen D1, and pronounces it before the second predetermined time elapses. The user's pronunciation is transmitted from the terminal 2 to the determination unit 22 as pronunciation data.

On the other hand, sentences in the first language selected by the question selection unit 14 are translated into sentences in the second language by the translation unit 17, and speech data of the second language sentences and sentences in the second language are generated by the speech data generation unit 11. Generate similar speech data.

After receiving the pronunciation data from the terminal 2, the determination unit 22 determines the degree of accuracy based on the second-language sentence speech data and the second-language sentence-similar speech data after the second predetermined time has elapsed. The determined accuracy is instructed to output from the transmission instruction unit 23 to the terminal 2 (instruction to perform either or both of generation of the fourth display screen and output of a specific sound).

The user is notified of the accuracy output from the terminal 2 by the output instruction. During the output by the terminal 2, the sound input unit of the terminal 2 is in a pronunciation unacceptable state.

After that (after the first predetermined time has passed since the first text selection by the question selection section 14), the question selection section 14 newly selects a sentence, and the same operation is repeated thereafter.

Therefore, from the user's point of view, the pronunciation is pronounced while the first display screen is displayed on the terminal 2, the accuracy is displayed on the terminal 2 as the fourth display screen, and then automatically switches to the new first display screen. Repeat the action, or pronounce with the first display screen displayed on the terminal 2, the accuracy is output as a specific sound from the sound output unit of the terminal 2, and then automatically a new first display screen (However, the terminal 2 may display the fourth display screen and output a specific sound).

The language learning system 1B according to the present embodiment includes a question sentence database 13 that stores one or more first language sentences, and a question selection unit 14 that selects one first language sentence from the question sentence database 13. a screen generation unit 15B for generating a first display screen D1 displaying sentences in the first language selected by the question selection unit 14; A timer 25 for sending a first timer signal is provided, and the question selection section 14 makes a selection when this first timer signal is input (the translation section 17 translates sentences in the first language selected by the question selection section 14 into the first language). Translation into two-language sentences is the same as in Example 1 (Example 2).

As a result, in this embodiment, sentences are automatically given to the terminal 2 one after another, so the user does not have to press the play button. Also, in this embodiment, unlike the first and second embodiments, the second language sentence voice data is not output from the terminal 2, so the user does not imitate this, but translates into the second language document by himself/herself. is pronounced (there is no concept of division as in Examples 1 and 2). At that time, by setting each of the above predetermined times short, it is required to instantly think of the translation and pronounce it as if it were a flash mental calculation, so it is possible to further improve speaking ability.

The present invention is suitable as a language learning system.

1, 1A, 1B Language learning system 2 Terminal 3 Server 11 Voice data generation unit 12 Communication control unit 13 Question sentence database 14

Question selection unit

15, 15A, 15B Screen generation unit 16 Word database 17 Translation unit 18 Division unit 19 Storage unit 20 Voice data output instruction unit 21 Pronunciation similarity database 22 Judgment unit 23 Transmission instruction unit 24 Extraction unit 25 Timer

Claims

a translation unit that translates sentences in a first language into sentences in a second language;
a speech data generation unit that generates second language sentence speech data, which is speech data of the sentence in the second language;
A language learning system, comprising: a determination unit that compares the second language sentence voice data with pronunciation data that is voice data of a user's pronunciation.
The speech data generation unit generates second-language sentence-similar speech data, which is speech data similar in pronunciation to sentences in the second language,
2. The judgment unit judges the accuracy of the pronunciation data by comparing the second-language sentence speech data and the second-language sentence-similar speech data with the pronunciation data. The language learning system according to .
comprising a dividing unit that divides the sentence in the second language to generate sentence fragments;
The speech data generation unit generates sentence piece speech data, which is speech data of the sentence piece,
3. The language learning system according to claim 1, wherein said judgment unit compares said sentence fragment speech data with said pronunciation data.
The voice data generation unit generates text segment-similar voice data, which is voice data whose pronunciation is similar to the text segment,
4. The language according to claim 3, wherein the judging unit judges the accuracy of the pronunciation data by comparing the sentence piece speech data and the sentence piece similar speech data with the pronunciation data. learning system.
3. The language learning system according to claim 1, further comprising an audio data output instructing unit that instructs the terminal to output the second language text audio data as audio.
5. The apparatus according to claim 3, further comprising a voice data output instructing unit that instructs the terminal to output the second language text voice data and to output the text segment voice data as voice. A language learning system as described.
a question sentence database in which one or more sentences in a first language are stored;
a question selection unit that selects one sentence in the first language from the question sentence database;
a screen generation unit that generates a display screen that displays sentences in the first language selected by the question selection unit;
A timer that sends a timer signal to the question selection unit every time a predetermined time elapses,
The question selection unit makes the selection when the timer signal is input,
3. The language learning system according to claim 1, wherein the translation unit translates sentences in the first language selected by the question selection unit into sentences in the second language.