KR101438087B1 - Method for providing language training service based on consecutive and simultaneous interpretation test using speech recognition engine - Google Patents

Abstract

The present invention relates to a method for enhancing the internalization of a foreign language sentence pattern and a method of providing a withdrawal training service using the speech recognition engine proposed in the present invention, After analyzing the meaning of the sentence quickly and carrying out a 'sequential interpreting training course' that instantly transforms into a foreign language while memorizing strongly, it instantaneously analyzes, memorizes, The interpreter training process is performed so that the deep training process is performed through the progressive interpreter training process to widen the range of available sentence patterns and increase the speed of fetching, Through the interpreter training process, more than several times It is the intrinsic information processing should exert different treatment and concentration performed can be trained to achieve internalization of a foreign language sentence patterns and automatically drawn out in a moment.
According to the present invention, based on a test technique using a speech recognition technology, the speech recognition engine is operated only for a predetermined time and is automatically stopped when the silence duration exceeds a predetermined time, . In the process of attempting to translate the learner himself without the correct answer being presented, the learner can correct the error by himself and generate the correct answer. You can strengthen your memory of correct foreign language expressions, and you can quickly train your users to translate sentences in their native language into foreign language sentences.
In addition, according to the present invention, by adopting the interpreter training method in which the mother tongue and the foreign language are rapidly intersected, the learner can intuitively grasp the different system between the two languages, Thereby making it easier to form a foreign language system.
According to the present invention, the degree of difficulty of a sentence performing sequential and simultaneous interpretation training can be adjusted so that it can be utilized for learners in various learning stages.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for enhancing the internalization of a foreign language sentence pattern and a method of providing a withdrawal training service by a sequential and simultaneous interpretation test using a speech recognition engine.

The present invention relates to a method for enhancing internalization of a foreign language sentence pattern and a method for providing a withdrawal training service, and more particularly, to a method for enhancing the internalization of a foreign language sentence pattern by sequential and simultaneous interpretation tests using a speech recognition engine, will be.

Various languages such as Korean and English have very different patterns. In order to overcome the difference, early on, Korean English educators set up 'formal' in the name of 'grammar' for various patterns of English, organize numerous rules and non - regular rules that do not follow them, The learners were asked to memorize the rules. Most Korean students have learned the laws over a long period of time, more than 10 years. However, even in the case of many excellent students who receive a high score in the difficult English test, our English education shows that it is very poor in speaking English. In support of this, Korea is ranked 121st in the list of 150 countries surveyed in the nation's speaking skills in 2012. Even later, our education authorities recognized the need for practical English education and introduced a new nationally recognized English proficiency test that included evaluations of <speaking> and <writing>. However, there is still doubt as to whether or not a fundamental change will take place have. First of all, most English educators do not know how to teach English that is outside the framework of grammar. In addition, it is not easy for our students, who have to get their English scores in order to meet even more problems in life, in an intense competition of entrance examinations, There is also a concern that it may be.

The present inventors have a belief that grammar learning should be suppressed as much as possible for effective foreign language learning. According to the learning method of memorizing the grammar in a conscious and historical manner, since the template is first formed and then it is spoken, it is necessary to search the corresponding grammar in the use of the continuous pattern phenomenon by the pre-formed grammatical information, This is because the processing speed of the language is very slow due to the process of contrasting with the law, and there are some adverse effects on the language use such as the unusual expression which is not actually used. In other words, in order to use a sentence pattern of one language naturally, it is necessary to perform an internalization process through repetitive execution rather than conscious learning such as using a keyboard or swimming, Only by practicing in an inductive way through the use of can achieve true results.

On the other hand, the English learning method of the <native language acquisition method> is rapidly increasing, in which some English education programs domestically lead to natural language exposure as if acquiring the native language. However, the acquisition of a mother tongue is an intrinsic "life" of the language of the society in which the child lives in order for the child to survive. In the infant period, which is the crucial period of language acquisition, This is a possible phenomenon. After the mother language is already formed and the instinctual descriptions of language acquisition are weakened, Korean learners who are exposed to English only during limited learning hours learn it as a natural process of learning a mother tongue, It is difficult to obtain the expected results in the acquisition.

Accordingly, the present inventors have been able to overcome various disadvantages of the English learning by the <native language acquisition method> which is performed in an environment where the native language can not be acquired while excluding the contradictory grammar education which shows an adverse effect in the actual use of the foreign language, The present inventor has repeatedly developed various learning experiments to develop the most effective method for learning a foreign language in a limited time.

The present invention has been proposed in order to solve the above-mentioned problems of the previously proposed methods. The present invention rapidly analyzes the meaning of a sentence, and instantly transcribes a foreign language without delay Simultaneous Interpretation Training Process "which implements the simultaneous interpreter training process, which requires listening to the next consecutive contents while instantly analyzing, memorizing and converting the listener contents, Repetition training is performed to broaden the range of available sentence patterns and increase the speed of retrieval. Through the simultaneous interpretation training process, the extreme information processing process is performed which requires several times faster processing and concentration than the sequential interpretation process, It is possible to achieve the internalization of sentence patterns and automatically draw out instantaneously, To provide an enhanced foreign language sentence patterns internalized according to the sequential and simultaneous interpreting test with the engine and lead training service delivery for that purpose.

In addition, the present invention is based on a test technique using a speech recognition technology, in which the speech recognition engine operates only for a predetermined time and automatically stops when the silence duration exceeds a predetermined time, In the course of attempting to translate the learner's self without the correct answer being presented, the learner can correct the error by himself and generate the correct answer. A method of enhancing the internalization of foreign language sentence patterns and providing a training service by text and voice interpretation test using a speech recognition engine, which is able to strengthen memories of foreign language expressions and train users to quickly translate sentences composed in the native language into foreign language sentences Provide for other purposes .

In addition, the present invention adopts an interpreter training method in which a mother tongue and a foreign language intersect at a high speed, thereby allowing learners to intuitively grasp different systems between the two languages, It is another object of the present invention to provide a method of enhancing the internalization of a foreign language sentence pattern and a method of providing a withdrawal training service according to a character and voice interpretation test using a speech recognition engine.

It is another object of the present invention to provide a speech recognition system and a speech recognition method which can be used for learners in various learning stages by implementing the degree of difficulty of the sentence performing sequential and simultaneous interpretation training, Another object of the present invention is to provide a method for enhancing the internalization of sentence patterns and a method for providing withdrawal training services.

In accordance with an aspect of the present invention, there is provided a method for enhancing internalization of a foreign language sentence pattern and providing an outgoing training service according to a sequential and simultaneous interpretation test using a speech recognition engine, The service providing server,

(1) storing learning contents in which a plurality of sentences composed of a mother tongue and a learning target language corresponding thereto are classified into a plurality of groups and group groups;

(2) transmitting to the user terminal a single sentence composed of the mother language out of a plurality of sentences belonging to one of the group groups stored in the step (1);

(3) receiving, from the user terminal, a first user voice recorded in a learning target language corresponding to the mother tongue voice transmitted in step (2) using a speech recognition engine;

(4) calculating a score for the first user voice received in the step (3) and transmitting the score to the user terminal;

(5) The method according to any one of (2) to (4), wherein among the plurality of sentences belonging to one group group in the step (2), when the score prepared for the sentence in the step (4) Performing the same steps as the steps (2) to (4) for the remaining sentences excluding the sentence in which the sentence is performed;

(6) The method according to any one of (1) to (6), wherein, in the case where the scores set in the step (4) are not less than a predetermined score for all of the plurality of sentences belonging to the group of the step And transmits the sentences sequentially to the user terminal at a predetermined time interval one by one, and after the respective sentences are transmitted, during the predetermined time, the learning target language corresponding to each transmitted sentence Receiving a second user voice recorded with the voice recognition engine from the user terminal;

(7) Scoring the second user voice received in step (6) and transmitting the score to the user terminal;

(8) In the case where the score set in the step (6) is equal to or larger than a predetermined score, one group (first group) among a plurality of groups belonging to one group group stored in the step (1) The first sentence of the first group is transmitted and then the second sentence is transmitted while the first sentence of the first group is transmitted, Receiving a third user voice recorded in a learning target language using the speech recognition engine from the user terminal, and performing the same process on the remaining first group of sentences;

(9) calculating a score for the third user voice received in the step (8) and transmitting the score to the user terminal;

(10) The method according to any one of (8) to (9), wherein among the plurality of groups belonging to one group group in the step (8), when the score set for the group in the step (9) Performing the same steps as those in steps (8) to (9) for groups other than the group in which the group is performed;

(11) When the score determined in the step (9) is equal to or larger than a predetermined score for all the groups belonging to any one group of the step (8), a plurality And a fourth user voice recorded in a learning target language corresponding to each transmitted sentence at the same time while transmitting the voice to the user terminal, Receiving from the user terminal using a speech recognition engine;

(12) Scoring a score for the fourth user voice received in the step (11) and transmitting the score to the user terminal.

Preferably,

(2) to (4) is re-executed if the score determined in the step (4) is less than a predetermined score, and the score calculated in the steps (7), (9) Is less than a predetermined score, it is possible to implement the steps again in the same manner.

Preferably, in said step (7), step (9) and step (12)

It may be a score summed up by sentences.

Preferably,

The degree of difficulty of the learning target language is different according to the group group,

Before the step (2), the method may further include receiving a selection of a group group matching the learning level of the user from the user terminal.

Preferably, in the steps (3), (6), (8) and (11)

It can be implemented to operate only for a predetermined time.

Advantageously, in said step (4)

The total score for the sentence and the score formed for each word or phoneme constituting the sentence can be displayed in any one or more selected from the group including numbers, alphabets, and graphs.

Preferably,

When the duration of silence during operation of the speech recognition engine exceeds a predetermined silence duration time, the operation of the speech recognition engine is automatically stopped and a score of the lowest grade can be given.

Preferably,

The learning target language may be English, and the mother tongue may be Korean.

Preferably,

In the steps (2), (6), (8) and (11)

A sentence transmitted from the service providing server to the user terminal is configured as a learning target language,

The first, second, third, and fourth user voices received from the user terminal may be voices recorded in the native language corresponding to the transmitted learning target language.

According to the present invention, the method of providing the service for enhancing the internal pattern of the foreign language and the training service of the foreign language according to the sequential and simultaneous interpretation test using the speech recognition engine can quickly analyze the meaning of the received sentence, Simultaneous Interpretation Training Course ", which is a process of interpreting and translating the contents into a foreign language and then instantaneously analyzing, memorizing and converting the contents of the listener, Through continuous interpreter training process, iterative training of deep processing is performed to widen the range of available sentence patterns and increase the speed of retrieval. Through the simultaneous interpretation training process, The extreme information processing process to be performed is performed, and foreign language sentence patterns It can be trained to achieve the internalization and automatically debited at the moment. In other words, through the process of sequential interpreting and simultaneous interpretation training for the same sentences, after learning a foreign language sentence pattern, reactivate the potential language ability of the rapidly declining brain after the mother tongue is formed, It is possible to achieve internalization of complicated foreign language sentence patterns and automatically make a momentary withdrawal so that it is possible to use a foreign language automatically and unconsciously similar to the native language in the end.

According to the present invention, based on a test technique using a speech recognition technology, the speech recognition engine is operated only for a predetermined time and is automatically stopped when the silence duration exceeds a predetermined time, . In the process of attempting to translate the learner himself without the correct answer being presented, the learner can correct the error by himself and generate the correct answer. You can strengthen your memory of correct foreign language expressions, and you can quickly train your users to translate sentences in their native language into foreign language sentences. These similar effects have been proven by the Cognitive Psychologist De Winstanley (1995) in the Theory of Generation Effectiveness. They do not present memories of college student experimenters in recalling experiments, The recall rate was significantly higher than that of the recalled group.

In addition, according to the present invention, by adopting the interpreter training method in which the mother tongue and the foreign language are rapidly intersected, the learner can intuitively grasp the different system between the two languages, Thereby making it easier to form a foreign language system. This is because, while it is difficult for foreign language systems to be formed separately due to repeated intervention of the mother tongue system in foreign language learning, it must be adhered to Parreren's claim of language learning psychologist, The reason for this can be overcome through conscious blocking in the process of foreign language interpretation. This is in accordance with Juhasz 's theory. After the mother tongue is formed, the overall thinking is also done in the mother tongue. Therefore, And actively cope with the interference, it is possible to speed up the formation of a foreign language system.

According to the present invention, the degree of difficulty of a sentence performing sequential and simultaneous interpretation training can be adjusted so that it can be utilized for learners in various learning stages. Even if it is a beginner who has not internalized many sentence patterns, it is possible to perform high level information processing of simultaneous interpretation without any difficulty if sentences are presented within the range of known vocabulary and patterns. In this process Once a fast retrieval process of a foreign language is formed, it can have a significant effect not only on strengthening internalization that has already been done but also on subsequent learning. In addition, through the experience of performing simultaneous interpretation, the learner can have confidence in the foreign language and genuine interest.

Brief Description of the Drawings Fig. 1 is a block diagram illustrating a system for implementing a method for providing a service for enhancing internalization of a foreign language sentence pattern and a withdrawal training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention.
FIG. 2 and FIG. 3 illustrate a flow of a method for enhancing internalization of a foreign language sentence pattern and providing a withdrawal training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention.
FIG. 4 is a flowchart illustrating a method for enhancing internalization of a foreign language sentence pattern according to a sequential and simultaneous interpretation test using a speech recognition engine according to an exemplary embodiment of the present invention and an information processing process of a simultaneous interpretation training step Fig.
5 is a diagram illustrating an example of a timeline setting of a speech recognition engine in a method for enhancing internalization of a foreign language sentence pattern and providing a retrieval training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention.
6 shows experimental results on the combined effect of information withdrawal and rapid response;
FIG. 7 is a diagram illustrating scores of words recorded for recorded voice in a method for enhancing internalization of a foreign language sentence pattern and providing a retrieval training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention drawing.
8 is a view illustrating a screen of a user terminal to which scored scores are transmitted in a method for enhancing internalization of a foreign language sentence pattern and providing a withdrawal training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention.
FIG. 9 is a flowchart illustrating a method for enhancing internalization of a foreign language sentence pattern according to a sequential and simultaneous interpretation test using a speech recognition engine according to an exemplary embodiment of the present invention and a simultaneous interpretation training step for each group Fig.
10 is a diagram showing experimental results on learning performance according to depth of information processing;
11 is a diagram illustrating an effective execution step of a foreign language sentence pattern internalization enhancement and a withdrawal training service providing method according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention in a foreign language learning exercise.
FIG. 12 is a diagram illustrating a series of processes of a method for enhancing internalization of a foreign language sentence pattern and providing a withdrawal training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. In the following detailed description of the preferred embodiments of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. The same or similar reference numerals are used throughout the drawings for portions having similar functions and functions.

In addition, in the entire specification, when a part is referred to as being 'connected' to another part, it may be referred to as 'indirectly connected' not only with 'directly connected' . Also, to "include" an element means that it may include other elements, rather than excluding other elements, unless specifically stated otherwise.

FIG. 1 is a block diagram illustrating a system for implementing a method for enhancing internalization of a foreign language sentence pattern and providing a withdrawal training service according to sequential and simultaneous interpretation tests using a speech recognition engine according to an embodiment of the present invention. As shown in FIG. 1, a system for implementing a method for providing a service for enhancing internalization of a foreign language sentence pattern and an outgoing training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention includes a service providing server 100, and a user terminal 200. [

The service providing server (100) includes a step of storing learning contents in which a plurality of sentences composed of a mother language and a learning target language corresponding thereto are classified into a plurality of groups and a group group; And transmits the sentence to the user terminal 200. In response to the sentence sent from the user terminal 200, the user's voice is transmitted to the user terminal 200 A sentence-based sequential interpretation test step for performing all of the sentences in each group, and a plurality of sentences belonging to a group group are composed of speech and transmitted to the user terminal 200 one by one at a predetermined time interval continuously, The user's voice recorded in the learning target language corresponding to the sentence is received from the user terminal 200 and the score And transmitting the sentences to the user terminal 200. In the whole sentence successive consecutive interpretation test step, all of the mother-tongue sentences constituting one group out of the plurality of groups belonging to one group group are constituted by speech, And transmits the second sentence after transmitting the first sentence, receives the user voice recorded in the learning target language corresponding to the first sentence from the user terminal 200, And a process of calculating a score for the transmitted user voice and transmitting the score to the user terminal 200 is performed for each of a plurality of groups belonging to a group group A simultaneous interpretation test step, and a plurality of sentences belonging to a group group, The user terminal 200 receives the user voice recorded in the learning target language corresponding to each sentence transmitted at the same time as transmitting the voice and transmits the score to the user terminal 200 And a whole sentence simultaneous simultaneous interpretation test step in which the sentence and the simultaneous interpretation test are performed using the speech recognition engine provided according to the present invention, Can be performed.

On the other hand, according to the embodiment, in the case of providing a training service for translating the learning target language into the mother language, the service providing server 100 transmits a sentence composed of the learning target language to the user terminal 200, And a voice recorded in the corresponding mother tongue may be received.

The user terminal 200 is a terminal of a user who wishes to learn a foreign language by using a foreign language sentence pattern internalization enhancement and withdrawal training service based on sequential and simultaneous interpretation tests using a speech recognition engine provided by the service providing server 100. [ The user downloads and installs a program provided by the service providing server 100 to the user terminal 200 and then records and transmits the user's voice through the user terminal 200 and receives a score from the service providing server 100 And receiving a foreign language sentence pattern internalization enhancement and withdrawal training service to be provided by the present invention. The service providing server 100 and the user terminal 200 can transmit and receive information through a network including the Internet, an intranet, a wired / wireless communication network, and a mobile communication network. In the present invention, the user terminal 200 may be a personal computer (PC), but it can be used regardless of the form of a specific terminal if it is an electronic device capable of networking with the service providing server 100 and capable of recording and image execution . However, since the speech recognition test engine must be used for the speech test, a microphone and a speaker (including a headphone) must be provided like the headset shown in FIG.

FIG. 2 and FIG. 3 are diagrams showing a flow of a method for enhancing internalization of a foreign language sentence pattern and providing a withdrawal training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention. As shown in FIGS. 2 and 3, the service providing server 100 for providing a foreign language sentence pattern internalization enhancement and withdrawal training service has a plurality of sentences composed of the mother language and the corresponding learning target language, (Step S200 to S500), a whole sentence successive sequential interpretation test step (steps S600 to S700), a group-based simultaneous interpretation test step (step S100) S800 to step S1000), and a whole sentence simultaneous simultaneous interpretation test step (steps S1100 to S1200). That is, it is configured to perform the sequential interpreting training first and then to perform the simultaneous interpreting training, and each interpreting training is composed of the test by sentence (or group) and the continuous test of the whole sentence so that the unconscious internalization Internalization), and internal punctuation of foreign sentence patterns and automatic withdrawal enhancement training services that train them to quickly withdraw what they want to say.

FIG. 4 is a flowchart illustrating a method for enhancing internalization of a foreign language sentence pattern according to a sequential and simultaneous interpretation test using a speech recognition engine according to an exemplary embodiment of the present invention and an information processing process of a simultaneous interpretation training step As shown in Fig. As shown in FIG. 4, the process of interpreting is performed through sentence listening, meaning analysis, instant memorization, language conversion, and utterance. The sequential interpreting is performed by translating the sentence into the translated B language, Simultaneous interpretation, on the other hand, speaks to the B language and simultaneously hears the next sentence of the A language. In other words, sequential interpreting is a quick analysis of the meaning of a listened sentence, instantaneous strong memorization, and conversion into foreign language without delay. Simultaneous interpretation is instantaneous analysis, memorization, The following information will be heard and extreme information processing processes will occur that can not be performed unless the processing and concentration are performed several times faster than the sequential interpretation process. The present invention provides a learning service that enables internalization of a foreign language sentence pattern and automatic instantaneous withdrawal by constructing a simultaneous interpretation training process after the sequential interpreting training process using the difference of the information processing process of the sequential interpretation and the simultaneous interpretation to provide. Hereinafter, each step will be described in detail.

Step S100 is a step in which the service providing server 100 stores learning contents in which a plurality of sentences composed of the mother language and the learning target language corresponding thereto are classified into a plurality of groups and group groups. In the group-based simultaneous interpretation test step (steps S800 to S1000), since each step is performed for each group composed of a plurality of sentences, a plurality of sentences can be classified into a plurality of groups and stored. The learning target language may be English, and the mother tongue may be Korean, but the present invention is not limited thereto, and various languages such as Chinese and Japanese can be applied and applied.

According to an embodiment, the learning content includes at least two groups of groups composed of a plurality of groups, each of which has a difficulty level of a learning target language and selects one group group according to the learning level of the user Thereby providing a differentiated learning service. In this case, the user terminal 200 may further include a step of receiving a selection of a group group matching the learning level of the user before the step S200. That is, the learning target language in one group group may be configured to have different levels of word, sentence length, and grammar, and different levels of difficulty among different group groups, so that differentiated learning can be performed by user level. Therefore, it is necessary for the novice to select a group consisting of only sentences with easy vocabulary and language pattern to be provided with services, and to be provided with a service for a group consisting of sentences having increasingly difficult vocabulary and patterns as the language ability improves So that continuous and gradual learning is possible. Also, since the sequential interpretation test steps (steps S200 to S700) and the simultaneous interpretation test step (steps S800 to S1200) can be performed separately, they can be performed on the learning target language in another group group. That is, one group group in step S200 and one group group in step S800 may mean another group group.

In steps S200 to S500, the service providing server 100 selects one of the plurality of sentences belonging to one group group to be subjected to the sequential interpretation test process among the group groups stored in the step S100, (S200) of transmitting a sentence of the first user's voice to the user terminal 200 (S200). The first user's voice recorded in the learning target language corresponding to the native language voice transmitted in step S200 (S400) of determining a score for the first user voice transmitted in step S300 and transmitting the score to the user terminal 200 (S400). In step S400, Among the plurality of sentences stored in step S100, among the sentences belonging to the group to be subjected to the sequential interpretation test procedure, in steps S200 to S400 And performing the same steps as the steps S200 to S400 for the remaining sentences excluding the sentence (S500).

The service providing server 100 stores a plurality of sentences stored in the server (e.g., "There is one armchair in the living room of my house (sentence 1)", "My mother and her brother I like to read a book quietly in an armchair (Sentence 2) "," Sometimes I sit on the armchair and nap "(Sentence 3), etc.) (First user voice) interpreted in the language of the learning target language (e.g., English) and transmitted to the user terminal 200 by voice (S200) (S300). The service providing server 100 can precisely score the pronunciation of the transmitted voice, such as the accuracy of pronunciation, the accuracy of pronunciation of words, and the order of pronunciation, and transmits the score to the user terminal 200 (S400). Interpersonal training may be performed in the same manner for other sentences (sentence 2, sentence 3) when the score for sentence 1 is higher than a predetermined score (for example, C) (S500). On the other hand, according to the embodiment, the sentence transmitted to the user terminal 200 by the service providing server 100 is composed of the learning target language, not the mother language, and the voice transmitted from the user terminal 200 corresponds to the learning target language It may be a voice recorded in the corresponding mother tongue (Korean). Hereinafter, an exemplary embodiment in which the service providing server 100 transmits in the native language and interprets it in the learning target language will be described as a representative example.

Step S300 is a step of receiving a first user's voice using the speech recognition engine from the user terminal 200. The step S300 is a step of receiving a first user's voice from the user terminal 200, In step S200, the pronunciation of the learning target language corresponding to the native language voice is received and the recorded language is received. The operation of the speech recognition engine in step S300 may be implemented to operate only for a predetermined time.

In addition, when the duration of the silence during the operation of the speech recognition engine exceeds a predetermined silence duration time, the operation of the speech recognition engine may be automatically stopped and a score of the lowest grade may be given. 5 is a diagram illustrating an example of a timeline setting of a speech recognition engine in a method for enhancing internalization of a foreign language sentence pattern and providing an outward training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention. As shown in FIG. 5, according to an embodiment of the present invention, a timeline of the speech recognition engine can be set. That is, the operation time of the speech recognition engine, the vocalization time, the continuous silence period before vocalization, and the silence duration time after vocalization can be set and reflected in the score formulation. According to an embodiment, the operation of the speech recognition engine may be automatically stopped and a score of the lowest grade may be given when the duration of the silence while the speech recognition engine is operating exceeds the set silent duration time . That is, when the user does not speak, the operation of the speech recognition engine is automatically stopped, the next sentence is immediately followed, or the test is performed again from the beginning, thereby inducing more tension and increasing concentration , It is possible to induce natural repetitive learning. In addition, since the user has to perform speaking rapidly, the user (learner) can quickly analyze the character or voice information made in the native language and train the user to convert it into a corresponding foreign language sentence.

6 is a diagram showing experimental results on the combined effect of information withdrawal and quick response. The present inventors carried out experiments in which subjects were presented with 10 foreigner face photographs and names of foreign languages for each photograph to memorize the effect of quick reaction when fetching information from the human brain. Subjects in Group A were repeatedly instructed to show their names after one second, and group B repeated the training to tell the name seven seconds after the picture appeared. At this time, the number of training repetitions in both groups was the same. As a result, as shown in FIG. 6, the group A showed overwhelmingly high recall and fast reaction rate in the test after one week. In other words, it was confirmed that, in the retrieval of memorized information, as in the present invention, the speedy speaking training exhibits a remarkably high learning effect.

Step S400 is a step of setting a score for the user's voice to provide feedback, and the score may be settled by precisely measuring the accuracy of pronunciation and expression. FIG. 7 is a diagram illustrating scores of words recorded for recorded voice in a method for enhancing internalization of a foreign language sentence pattern and providing a retrieval training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention FIG. As shown in Fig. 7, the score at step S400 can be expressed by an alphabet (e.g., A-) and a graph, which is calculated for each word. In other words, the total score (sentence score) for the sentence and the score formed for each word constituting the sentence can be displayed in alphabet and graph. However, the present invention is not limited to this, and various methods can be applied, for example, scores may be expressed by numbers for more precise feedback, and scores may be expressed by numbers. Therefore, according to the present invention, it is possible to sequentially provide diverse and precise feedbacks, such as providing graded feedback (score formation and transmission), accuracy of word and phoneme pronunciation and accuracy (word selection, word order, etc.) And to increase the learning effect.

Meanwhile, in the present invention, if the score determined in step S400 is less than a predetermined score, steps S200 to S400 may be re-performed to implement the next step (training on the next sentence). That is, if the score determined at each step is equal to or higher than a predetermined score, the user is allowed to move to the next step. By passing through the instant test, the user is allowed to perform the next step so that the user can feel tension and maintain the immersed concentration state , And it is possible to improve the efficiency of learning by realizing that the user repeatedly learns naturally about the deficient portion. In addition, it does not set the number of iterations unilaterally, but it induces an appropriate amount of the iterative learning according to the difference of the learner's ability and concentration, and provides multi-level feedback of discriminative ability to the learner's voice, It can be done without the tedious process of progressing.

Steps S200 to S400 are performed for all of the plurality of sentences belonging to the group to be subjected to the sequential interpretation test procedure (S500). When the scores set in step S400 are equal to or greater than a predetermined score, A whole sentence sequential interpreting test step (steps S600 to S700) can be performed. The entire sentence successive sequential interpretation test step (steps S600 to S700) is a step in which the service providing server 100 constructs all of a plurality of mother-tongue sentences belonging to the group to be subjected to the sequential interpretation test procedure by speech, The second user's voice recorded in the learning target language corresponding to each sent sentence is transmitted from the user terminal 200 to the user terminal 200 for a predetermined period of time after transmitting the respective sentences, (S600) receiving the speech using the recognition engine, and scoring the score of the second user voice transmitted in step S600 and transmitting the score to the user terminal 200 (S700).

Steps S600 and S700 successively test all the mother-tongue sentences (the plural mother-tongue sentences belonging to the group to be subjected to the sequential interpretation test procedure), which are the learning contents stored in the learning step, stored in step S100, And the score is derived. In step S600, sentence 1 is transmitted as speech, and the process of recording and transmitting the voice (second user voice) converted by the user into the learning target language corresponding to the user using the speech recognition engine for a predetermined period of time, . &Lt; / RTI &gt; In step S700, a score may be calculated for the second user voice and transmitted to the user terminal 200 to provide feedback. FIG. 8 is a diagram illustrating a screen of a user terminal to which scored scores are transferred in a method for enhancing internalization of a foreign language sentence pattern and providing a withdrawal training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention. 8, if the reference score (for example, exceeding 80) is not satisfied in the test for the entire sentence, the end button (&quot; CLOSE &quot;) is not generated, so that the previous step can be re-executed (&quot; RETEST &quot;). When the reference score is satisfied, an end button is generated and the process can be terminated, and re-execution is possible according to the user's selection. Depending on the embodiment, the score calculated and transmitted in step S700 may be implemented so as to score a score for each sentence and to transmit a total score summed. Scores should be set for each sentence so that more precise feedback is possible. As described above with reference to step S400, when the duration of the silence exceeds the predetermined silence duration time, the operation of the speech recognition engine is automatically stopped so that the score of the lowest grade is given. That is, when the user does not utterance and after a predetermined time, the test is performed again from the beginning, thereby inducing more tension, increasing concentration, and inducing natural repetitive learning.

A series of sentence-by-sentence and sentence sequential successive interpretation test steps described above in connection with steps S200 to S700 are performed, and when the determined score reaches a certain level above a predetermined score, the simultaneous interpretation test step (S800 to S1200) Can be performed.

The simultaneous interpretation test steps (S800 to S1200) constructs all the plural mother-tongue sentences constituting one group (the first group) out of plural groups belonging to one group group among the group groups stored in step S100 as speech, The second sentence is transmitted after the first sentence of the first group is transmitted while the third sentence of the first group is transmitted and the third user voice recorded in the learning target language corresponding to the first sentence (S800) of receiving sentences of the first group by using the speech recognition engine from the user terminal 200 and scoring the scores of the received third user voice in step S800 (S900) of transmitting to the user terminal 200, and when a score determined for the group in step S900 is equal to or greater than a predetermined score, (S1000) performing the same steps as the steps S800 to S900 for groups other than the group in which the steps S800 to S900 are performed among a plurality of groups to which the group S100 belongs, and a step S800 If the score set in step S900 is equal to or greater than a predetermined score for all of a plurality of groups belonging to a group group, all of the plurality of mother-tongue sentences belonging to a group group in step S800 are constituted by speech, The fourth user's voice recorded in the learning target language corresponding to each sentence transmitted at the same time as transmitting the voice is transmitted from the user terminal 200 through the voice recognition engine The score of the fourth user voice transmitted in step S1100, and the score of the fourth user voice received in step S1100, (Step S1200). The whole sentence simultaneous simultaneous interpretation test step may be implemented by including the step S1200.

The simultaneous interpretation test steps (S800 to S1000) for each group are similar to the above-described sequential interpretation tests for each sentence. However, unlike the sequential interpretation, the nature of the simultaneous interpretation Since a plurality of sentences must be sent consecutively, each step is performed by 'group by' rather than 'by sentence'. In addition, each sentence is transmitted to the user terminal 200 continuously without a time interval, and receives a third user voice recorded in the learning target language simultaneously with the transmission of the second sentence after the first sentence of the first group Lt; / RTI &gt; FIG. 9 is a flowchart illustrating a method for enhancing internalization of a foreign language sentence pattern according to a sequential and simultaneous interpretation test using a speech recognition engine according to an exemplary embodiment of the present invention and a simultaneous interpretation training step for each group Fig. As shown in FIG. 9, the sequential interpreting for each sentence consists of a separate step of 'sentence 1 voice presentation' and 'sentence 1 sequential interpreting', and each pass can be determined for each sentence. However, 2 voice presentation 'and' sentence 1 simultaneous interpretation 'are performed simultaneously, and the same process is performed for all the remaining sentences in the same group, and simultaneous interpretation of all sentences in the group is performed. In addition, the sequential interpreting process for the same sentences is performed prior to simultaneous interpreting training requiring high processing speed and high concentration, so that it is naturally repeated step by step. As described above, the learning contents used for performing the sequential interpretation process and the learning contents used for performing the simultaneous interpretation process are different group groups, and the learning objects can be different. That is, the learning contents used for performing the simultaneous interpretation process may be a group consisting of learning languages having a higher level of difficulty than the learning contents used for performing the sequential interpretation process, or may be composed of sentences using sequential interpretation processes It may be. This is because, if the sequential interpretation process and the simultaneous interpretation process occur on the same sentence, the training effect of repeating the memorized sentence may be seen. However, the present invention is not limited thereto, and if the simultaneous interpretative testing process is performed after the sequential interpreting test process is performed, the target learning content can be applied in various applications.

10 is a diagram showing experimental results on the learning performance according to the information processing depth. As shown in Fig. 10, the greater the depth of information processing within the potential processing capability range, the higher the learning effect is. In other words, when a user has a certain level of processing ability after practicing step by step in the practice of computer keyboard typing, he sets up a timer and quickly exercises, and at the same time, performs high-level practice of typing hantas, If you have a great depth of training, you will be able to exert a superior effect compared to less difficult practice. The present inventors paid attention to such experimental results, and made efficient foreign language learning possible by constructing a training with a large depth of information processing for performing a simultaneous interpretation process after completion of the sequential interpretation process. FIG. 11 is a diagram illustrating an effective step of performing a method for providing a training service for enhancing internalization of a foreign language sentence pattern and performing a consecutive interpretation test using a speech recognition engine according to an exemplary embodiment of the present invention. As shown in FIG. 11, the present invention is configured to perform a letter-presentation type interpreting training and a voice-presentation type interpreting training prior to the sequential interpreting training and simultaneous interpretation training, thereby improving the processing capability of the user And then perform a high-level sequential and simultaneous interpretation, thereby achieving a more efficient learning effect. The letter presentation type interpreting training means training by a process in which the service providing server 100 first transmits a sentence composed in the mother language to the user terminal 200 by text and receives the corresponding speech of the learning target language, Type interpreting training means training by a process in which the service providing server 100 transmits to the user terminal 200 a sentence composed in a native language by voice and receives a corresponding speech of the target language. That is, it means a dual training in which interpreting is relatively easy while watching the presented characters without burdening the momentary memorization in the character interpretation process, and a momentary memorizing process is added in the voice interpretation process.

On the other hand, the whole sentence simultaneous simultaneous interpretation test step (steps S1100 to S1200) is a step of continuously performing simultaneous interpretation and training of sentences practiced by each group. This corresponds to the sequential interpreting and training of all sentences practiced for each sentence in the sequential interpreting test steps (steps S200 to S700) (steps S600 to S700). However, due to the characteristics of the simultaneous interpretation training, as described in connection with steps S800 to S1000, the sentences are successively transmitted to the user terminal 200 without a time interval, and after the first sentence of the first group, And receives a third user voice recorded in the learning target language at the same time of transmission.

The process of simultaneously transmitting voice between the service providing server 100 and the user terminal 200 is based on the < immersion type simultaneous interpretation training method &gt;. The present inventors have developed and applied &quot; Immersive Simultaneous Interpretation &lt; RTI ID = 0.0 &gt; Training &quot;,< / RTI > which can effectively learn a foreign language through repetitive training of intensive language processing. The <Immersion type Simultaneous Interpretation and Training Act> was developed based on the ability to simultaneously translate one language into another language, using both languages fluently, and memorizing vocabulary, memorizing basic sentence patterns, listening and speaking Shadowing, memorization of meaning, listening afterwards, answer, speech, sequential interpretation, simultaneous interpretation. The Immersion-type Simultaneous Interpretation and Training Method induces a high concentration of concentration and learning motivation by using cognitive psychological principles in repetitively practicing foreign languages, training foreign language to the level of the unconscious automation reaction, The goal is to use. In step S600, a whole sentence successive test step for continuously testing the entire sentence is constructed, and a method of training the user to recognize the sentence sent to the user and to speak the corresponding language is used. As a result of this training, learners can use foreign language by unconscious instantaneous reaction, internalization of various vocabulary and language patterns unconsciously through repeated practice, and conscious thinking such as judgment or reasoning is omitted Intuitive and immediate automation response (withdrawal) can be achieved. Also, the &apos; immersive simultaneous interpretation &amp; training method &apos; applied to the present invention can acquire the same learning effect anytime and anywhere regardless of time, place, and ability of a teacher because online learning is possible using a computer or a dedicated terminal.

In the group-based simultaneous interpretation test steps (S800 to S1000) and the entire sentence continuous simultaneous interpretation test steps (S1100 to S1200), a series of processes such as other speech recognition engines and score development are performed in the above- S500) and the whole sentence successive sequential interpretation test steps (S600 to S700), detailed description thereof will be omitted.

FIG. 12 is a diagram illustrating a series of processes of a method for enhancing internalization of a foreign language sentence pattern and providing a withdrawal training service according to a sequential and simultaneous interpretation test using a speech recognition engine according to an embodiment of the present invention. 12, the method for enhancing the internalization of the foreign language sentence pattern and providing the withdrawal training service according to the sequential and simultaneous interpretation tests using the speech recognition engine according to an embodiment of the present invention includes the steps of: And passes through a certain level or more, the service providing server 100 performs a test and pass process for each sentence to be provided by the service providing server 100, such as a sequential interpretation test for sentence 2, A continuous sequential interpretation test process can be performed on the entire sentence including the sentence. If the score for the successive sequential interpretation test for the entire sentence is greater than the predetermined score and passes, the simultaneous interpretation test process may proceed with a higher level. Once the group-by-group simultaneous interpretation test process and the entire sentence consecutive simultaneous interpretation test process are completed, a series of learning processes may be terminated. However, if the scores are not enough for each step, the test can be re-executed. The whole sentence successive test is time-consuming and is converted into a total score, so that it can be implemented to re-execute the test only for sentences where the detailed score for each sentence is less than a predetermined score. According to such a method, it is possible to re-execute only the deficient portion except for the portion already completed, thereby improving the learning efficiency and reducing the boredom due to repetition.

Meanwhile, according to the embodiment, the step of receiving the member information from the user before the step S100 and receiving the membership information can be implemented so that the learning can be performed by the level of the user, and the progress management and the score management can be easily performed have.

The present invention may be embodied in many other specific forms without departing from the spirit or essential characteristics of the invention.

100: service providing server 200: user terminal
S100: storing a learning content in which a plurality of sentences composed of a mother tongue and a learning target language corresponding thereto are classified into a plurality of groups and a group group
S200: transmitting one sentence consisting of the mother tongue from the plurality of sentences belonging to one of the group groups stored in the step S100 to the user terminal by voice
S300: Receiving the first user voice recorded in the learning target language corresponding to the mother tongue voice transmitted in step S200 using the speech recognition engine from the user terminal
S400: Scoring a score for the first user voice transmitted in step S300 and transmitting the score to the user terminal
S500: Performing the same steps as the above-described steps S200 to S400 for the remaining sentences excluding the sentences in which the above-mentioned steps S200 to S400 are performed, among the plurality of sentences belonging to the group group of the above-mentioned step S200
S600: All of the stored plural mother-tongue sentences belonging to one group group in step S200 are composed by speech and sent to the user terminal continuously one by one at a predetermined time interval, and after each sentence is transmitted, Receiving a second user voice recorded in a learning target language corresponding to each sent sentence from the user's terminal for a predetermined period of time;
S700: The score of the second user voice transmitted in step S600 is calculated for each sentence and the sum of the scores is transmitted to the user terminal
S800: All of the plural mother-tongue sentences constituting one group (the first group) out of the plurality of groups belonging to one group group in the above-mentioned step S200 are composed by speech, Receiving a third user voice recorded in a learning target language corresponding to the first sentence from the user's terminal while transmitting a second sentence after transmitting the first sentence of the first group, Performing the same process for the remaining first group sentences
Step S900: a score is calculated for each sentence with respect to the third user voice transmitted in the step S800, and the summed score is transmitted to the user terminal
S1000: Performing the same process as the steps S800 to S900 for the remaining groups excluding the group in which the steps S800 to S900 are performed among the plurality of groups belonging to one group group in the step S800
S1100: All of a plurality of mother-tongue sentences belonging to a group group of the step S800 are composed by speech, and are sequentially transmitted to the user terminal without a time interval one by one. At the same time as transmitting the voice, Receiving a fourth user voice recorded in a learning target language corresponding to a sentence from the user terminal
S1200: Scoring a sentence score for the fourth user voice transmitted in step S1100 and transmitting the summed score to the user terminal

Claims (9)

A service providing server for providing a foreign language sentence pattern internalization enhancement and withdrawal training service,
(1) storing learning contents in which a plurality of sentences composed of a mother tongue and a learning target language corresponding thereto are classified into a plurality of groups and group groups;
(2) transmitting to the user terminal a single sentence composed of the mother language out of a plurality of sentences belonging to one of the group groups stored in the step (1);
(3) receiving, from the user terminal, a first user voice recorded in a learning target language corresponding to the mother tongue voice transmitted in step (2) using a speech recognition engine;
(4) calculating a score for the first user voice received in the step (3) and transmitting the score to the user terminal;
(5) The method according to any one of (2) to (4), wherein among the plurality of sentences belonging to one group group in the step (2), when the score prepared for the sentence in the step (4) Performing the same steps as the steps (2) to (4) for the remaining sentences excluding the sentence in which the sentence is performed;
(6) The method according to any one of (1) to (6), wherein, in the case where the scores set in the step (4) are not less than a predetermined score for all of the plurality of sentences belonging to the group of the step And transmits the sentences sequentially to the user terminal at a predetermined time interval one by one, and after the respective sentences are transmitted, during the predetermined time, the learning target language corresponding to each transmitted sentence Receiving a second user voice recorded with the voice recognition engine from the user terminal;
(7) Scoring the second user voice received in step (6) and transmitting the score to the user terminal;
(8) In the case where the score set in the step (6) is equal to or larger than a predetermined score, one group (first group) among a plurality of groups belonging to one group group stored in the step (1) The first sentence of the first group is transmitted and then the second sentence is transmitted while the first sentence of the first group is transmitted, Receiving a third user voice recorded in a learning target language using the speech recognition engine from the user terminal, and performing the same process on the remaining first group of sentences;
(9) calculating a score for the third user voice received in the step (8) and transmitting the score to the user terminal;
(10) The method according to any one of (8) to (9), wherein among the plurality of groups belonging to one group group in the step (8), when the score set for the group in the step (9) Performing the same steps as those in steps (8) to (9) for groups other than the group in which the group is performed;
(11) When the score determined in the step (9) is equal to or larger than a predetermined score for all the groups belonging to any one group of the step (8), a plurality And a fourth user voice recorded in a learning target language corresponding to each transmitted sentence at the same time while transmitting the voice to the user terminal, Receiving from the user terminal using a speech recognition engine;
(12) Scoring a score for the fourth user voice received in the step (11) and transmitting the score to the user terminal. The foreign language sentence according to the sequential and simultaneous interpretation test using the speech recognition engine Pattern internalization enhancement and withdrawal training service.
The method according to claim 1,
(2) to (4) is re-executed if the score determined in the step (4) is less than a predetermined score, and the score calculated in the steps (7), (9) Wherein the steps are re-executed in the same manner even when the score is less than a predetermined score. The method of claim 1, further comprising:
2. The method of claim 1, wherein in steps (7), (9) and (12)
And a score obtained by summing up and summing up sentences. The method for enhancing the internalization of a foreign language sentence pattern according to sequential and simultaneous interpretation tests using a speech recognition engine and providing a withdrawal training service.
The method according to claim 1,
The degree of difficulty of the learning target language is different according to the group group,
The method of claim 1, further comprising the step of receiving, from the user terminal, a selection of a group group matching the learning level of the user before the step (2). Pattern internalization enhancement and withdrawal training service.
2. The method of claim 1, wherein in steps (3), (6), (8) and (11)
Wherein the method further comprises the steps of: (a) providing a training service for a foreign language;
2. The method of claim 1, wherein in step (4)
Characterized in that the total score for the sentence and the scores formed for each word or phoneme constituting the sentence are displayed in any one or more selected from the group including numbers, alphabets, and graphs. A Method for Intensifying the Internalization of Foreign Language Sentence Patterns and Providing Training Services for Extraction by Sequential and Simultaneous Interpretation Tests Using.
The method according to claim 1,
Characterized in that the operation of the speech recognition engine is automatically stopped and a score of the lowest grade is given when the duration of silence during operation of the speech recognition engine exceeds a predetermined silence duration time limit. A Method for Enhancing the Internalization of Foreign Language Sentence Patterns by Using Sequential and Simultaneous Interpretation Tests Using Recognition Engine and Providing Training Services.
The method according to claim 1,
Wherein the learning target language is English and the mother tongue is Korean. A method for enhancing internalization of a foreign language sentence pattern and providing an outgoing training service according to a sequential and simultaneous interpretation test using a speech recognition engine.
The method according to claim 1,
In the steps (2), (6), (8) and (11)
A sentence transmitted from the service providing server to the user terminal is configured as a learning target language,
Wherein the first, second, third, and fourth user voices received from the user terminal are voice recorded in a mother tongue corresponding to the transmitted learning target language. To enhance the internalization of foreign language sentence patterns and to provide training services.

Images