KR101959246B1 - Server and method for training speech disorder - Google Patents

Abstract

A language disorder training server and method are disclosed. The language disturbance training server according to an embodiment of the present invention includes a user interface unit for providing a reference word to a user terminal according to a predetermined language training cycle and receiving pronunciation data for a reference word, An analysis unit for comparing the converted text with pre-stored similar words to analyze the degree of matching, a DB management unit for accumulating the analyzed degrees of agreement to build a training result database, and a language training And the like.

Description

Technical Field [0001] The present invention relates to a server and method for training speech disorder,

The present invention relates to a language disorder training server and method, and more particularly, to a language disorder training server and method capable of analyzing a pronunciation input from a user to determine whether a language rehabilitation treatment is performed well.

The biggest difference between humans and other animals is that they use language as a tool of communication. Occasionally animals such as primates can acquire a limited range of language forms as a result of long-term training, but at a level comparable to human language in terms of their breadth, diversity and rigorous regularity.

When we look at the process of human communication, we first think about the meaning to be conveyed from the brain, and when the words and phrases expressing it are selected, the speech is made and sent to the specific agency. In this case, the part where the brain tissue is controlled is the language, and the part making the sound in the surrounding organism corresponds to the speech.

Language is a complex symbolic system for communicating thought and communication in a variety of ways. It consists of five scales: phonology of rules of sound, semantics of rules of word meaning and combination, Morphology about how it is constructed, syntax about whether a word is used syntactically in a sentence, and pragmatics about whether to properly start a topic and keep it in context. have.

On the other hand, a horse is a movement process that makes sounds by the respiratory system, larynx, oral cavity, etc. for communication. That is, the process of making a horse can be divided into a breathing process in which diaphragm and intercostal muscles, abdominal muscles, sternocleidomastoid muscles act, a vocalization process in the larynx, and an articulation process according to the movement of the tongue, chin and pharyngeal muscles.

According to statistics from developed countries, about 5% of the population is the subject of speech therapy. Individuals with language problems may experience mental pain and frustration due to limited communication and social and economic activities. Therefore, language rehabilitation through accurate diagnosis is very important for language therapy subjects.

As the number of language-related disabilities increases year by year, accurate diagnosis of language therapy subjects requires an analysis system based on extensive data on patients with similar symptoms. However, up to now, such a system is not available.

Korean Patent Laid-Open No. 10-2016-0020008 (published on February 23, 2016)

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and it is an object of the present invention to provide a method and apparatus for receiving a pronunciation of a reference word periodically from a user having a language disorder, And to provide a language disorder training server and method which can easily grasp the information about the training of the language.

The solution of the present invention is not limited to those mentioned above, and other solutions not mentioned can be clearly understood by those skilled in the art from the following description.

As a means for solving the above-mentioned technical problems, the language disturbance training server according to the embodiment of the present invention provides a reference word to a user terminal according to a predetermined language training cycle, An interface unit, a conversion unit for converting input pronunciation data into text, an analyzing unit for comparing the converted text with pre-stored similar words to analyze the matching degree, a DB management unit for accumulating the analyzed matching degree to build a training result database, And an information providing unit for providing information related to language training using the established training result database.

Preferably, the conversion unit may be a Speech to Text (STT) conversion module.

Preferably, the DB management unit may include a nonconformity word database for storing and managing a plurality of unsuitable similar words corresponding to inappropriate pronunciation for each reference word, in accordance with the degree of matching.

Also, preferably, the analyzing unit can analyze the degree of coincidence using data stored in the nonconformity word database.

Also, preferably, the information providing unit may generate information related to the language training, including the treatment process, the treatment effect, prediction information on future training results, and information on future setting and presentation of the treatment direction for the training period.

Also, preferably, the information providing unit may give a training weight value when the analyzed degree of agreement is higher than the predetermined achievement level.

Meanwhile, a method for training a language disorder according to another embodiment of the present invention includes providing a reference word to a user terminal according to a preset training period, receiving pronunciation data for a reference word from a user terminal, Comparing the converted text with previously stored similar words to analyze the degree of coincidence, building a training result database by accumulating the analyzed degrees of coincidence, and constructing a training result database by using the constructed training result database And providing information related to training.

Preferably, the converting step may use a STT (Speech to Text) conversion module.

The method may further include building an unsuitable word database for storing and managing a plurality of unsuitable similar words corresponding to inappropriate pronunciation for each reference word in a step corresponding to the degree of matching.

Also preferably, the analyzing step may analyze the degree of coincidence using data stored in the nonconforming word database.

Also preferably, the providing step may generate information related to the language training, including information about the course of treatment, the therapeutic effect, the prediction of future training results, and the setting and presentation of future treatment directions for the training period have.

Also preferably, the providing step may provide a training weight if the analyzed degree of coincidence is higher than the preset attainment level.

The language disability training server and method according to the present invention is a language disability training server and method according to the present invention, whenever a user having a language disability inputs pronunciation data for a reference word and continuously constructs a vast database by continuously analyzing the analyzed information, It is possible to provide training information.

In addition, by providing a result report to the user based on accurate language training information, the user can grasp his / her state in detail, establish a language training plan in the future, So that it is possible to provide an opportunity to strengthen the system.

The effects of the present invention are not limited to those mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the following description.

1 is a configuration diagram of a language disorder training system according to a preferred embodiment of the present invention;
2 is a block diagram of a language disorder training server according to a preferred embodiment of the present invention,
FIG. 3 is a block diagram of a DB management unit included in a language disorder training server according to a preferred embodiment of the present invention,
4 is a flowchart illustrating a method for training a language disorder according to a preferred embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The above and other objects, features, and advantages of the present invention will become more readily apparent from the following description of preferred embodiments with reference to the accompanying drawings. However, the present invention is not limited to the embodiments described herein but may be embodied in other forms. Rather, the embodiments disclosed herein are provided so that the disclosure can be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

In this specification, where an element, component, apparatus, or system is referred to as comprising a program or an element made up of software, it is to be understood that the element, component, (E.g., memory, CPU, etc.) or other program or software (e.g., drivers necessary to run an operating system or hardware, etc.) necessary to run or operate the computer system.

It is also to be understood that the elements (or elements) may be implemented in software, hardware, or any form of software and hardware, unless the context requires otherwise.

Also, terms used herein are for the purpose of illustrating embodiments and are not intended to limit the invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. The terms "comprises" and / or "comprising" used in the specification do not exclude the presence or addition of one or more other elements.

1 is a configuration diagram of a language disorder training system according to a preferred embodiment of the present invention.

Referring to FIG. 1, the language impairment training system according to the preferred embodiment of the present invention includes a language impairment training server 100 and a plurality of user terminals 200 that can be connected through various wired and wireless networks.

The language impairment training server 100 is a server for performing a main function for training a language disorder in the present invention and exchanges predetermined information with a plurality of user terminals 200 through a network.

The language disability training server 100 provides reference words to corresponding user terminals of a plurality of user terminals 200 according to a preset training period, receives pronunciation data of the reference words, analyzes the data, Keep it in database.

In addition, the language disability training server 100 provides information related to language training to the user using the database data. The operation of the language-impaired training server 100 will be described later in more detail with reference to FIG.

A plurality of user terminals 200 are terminals corresponding to a user who is connected to the language disorder training server 100 through a network and receives language training. As shown in the figure, the user terminal 200 includes a user having a terminal such as an electronic pad, .

The plurality of user terminals 200 are connected to the language obstacle training server 100 to provide the user with a reference word provided by the language obstacle training server 100 and allow the user to pronounce the reference word, And transmits the pronunciation data to the language impaired training server 100. Accordingly, the plurality of user terminals 200 can receive the language training related information from the language obstacle training server 100 and provide the information to the user.

The plurality of user terminals 200 may be a terminal device having a history already connected to the language obstacle training server 100 and may provide personal information to the language obstacle training server 100 in advance, Device.

The personal information provided by the user may include information about the degree of the individual's language impairment, including information that the client typically provides when subscribing to the server. For example, information about an individual's language impairment information may include a result of a language impairment diagnosis issued by an institution such as a hospital. Accordingly, the language impairment training server 100 can use the reference word provided for each individual as a measure to provide differently.

In the present embodiment, only the notebook having the electronic pad and the headset is shown for convenience of explanation. However, the present invention is not limited thereto, Either device can be applied.

2 is a block diagram of a language disorder training server according to a preferred embodiment of the present invention.

2, the language impairment training server 100 according to the preferred embodiment of the present invention includes a user interface unit 110, a conversion unit 120, an analysis unit 130, a DB management unit 140, (150), and a control unit (160).

The user interface unit 110 supports an interface with a user terminal connected to the language obstacle training server 100 among a plurality of user terminals 200. More specifically, the user interface unit 110 provides a reference word to the user terminal according to a predetermined training cycle, and allows the user to pronounce the reference word, and receives pronunciation data based on the user's voice.

The conversion unit 120 converts the pronunciation data of the user input through the user interface unit 110 into text. The operation of converting the pronunciation data into text in the conversion unit 120 can be performed by a STT (Speech to Text) conversion module. Since the STT conversion module is a known technology, a description thereof will be omitted.

The analysis unit 130 compares the data converted by the conversion unit 120 with the previously stored similar words and analyzes the degree of matching. At this time, the analysis unit 130 may utilize the pre-established similar word DB, and the similar word DB will be described in more detail later with reference to FIG.

The analysis unit 130 can obtain the results shown in Table 1 using the similar word DB. Also, after the result as shown in Table 1 is obtained, this data can be stored again in the similar word DB.

Reference word: sky A similar word Degree of agreement Hanul 90% Harbor 90% Han 90% ... ... Hanul 80% One day 80% ... ... Hach 70% The 70% ... ...

Referring to Table 1, when the reference word "sky" is presented to the user and the user pronounces "sky" through his or her user terminal, And the analysis unit 130 analyzes how much the user pronounced "sky" coincides with the reference word "sky ", and the degree of correspondence in Table 1 can be expressed as a percentage.

If the degree of coincidence analyzed by the analysis unit 130 is 90%, the language disorder is not severe, or the language disorder is greatly improved through the speech disorder training system. In addition, if the degree of coincidence analyzed by the analysis unit 130 is 70% or less, the language disorder can be defined as a serious level. In this case, the language training period of the user can be shortened You can set it.

The DB management unit 140 manages at least one database in order to manage information required for the operation of the language obstacle training server 100. In the present embodiment, the DB management unit 140 accumulates the degree of correspondence analyzed by the analysis unit 130 to build a training result database, which will be described later, and provides information on the training result database to the information providing unit 150 have. The operation of the DB management unit 140 will be described later in more detail with reference to FIG.

The information providing unit 150 provides information related to language training using the degree of matching analyzed by the analyzing unit 130 and the information of the training result database. In the present exemplary embodiment, the information providing unit 150 may be configured to perform a language training, such as a training process, a treatment effect, prediction information on future training results, and information on future setting and presentation of a treatment direction, Can be generated. Here, information on the setting and presentation of the treatment direction in the future may be set so as to set and present a more specific treatment direction than the previously reported other information, that is, the treatment process, the therapeutic effect, The state corresponds to the item.

In addition, the information providing unit 150 may utilize the information related to the language training that has been prepared, and may include information on the effect of the language training on the period, analysis information, statistical information, comparison information, And the user training report can be provided to the user through the user interface unit 110. The user training unit 110 can generate a training report for the user. Training reports for users can include various types of graphs so that users can easily understand the effects of language training.

In addition, the information providing unit 150 may give a training weight value when the degree of matching is higher than the predetermined achieving level. If the information providing unit 150 has given a training weight, it is also preferably included in the user training report.

The controller 160 controls the overall functions required for the operation of the language obstacle training server 100. That is, the control unit 160 controls signal input / output between the user interface unit 110, the conversion unit 120, the analysis unit 130, the DB management unit 140, and the information providing unit 150.

More specifically, when a user's voice is input from a plurality of user terminals 200, the control unit 160 converts the pronunciation data input through the user interface unit 110 by the conversion unit 120, 130, and is controlled by the DB management unit 140 so as to be provided as information required by the information providing unit 150.

3 is a block diagram of a DB management unit included in a language disorder training server according to a preferred embodiment of the present invention.

3, the DB management unit 140 shown in FIG. 2 may include a member information DB 140a, a reference word DB 140b, a similar word DB 140c, and a training result DB 140d .

The member information DB 140a is a database for storing information on members who have joined the language disorder training server 100. [ Here, the members joining the language impairment training server 100 may be users with language impairments, or guardians of users with language impairments.

The member information stored in the member information DB 140a includes simple personal information basically requested by the server in order for the clients to access the server and use the server. Further, in the case of a user having a language disorder, the degree of language disorder, for example, a result of diagnosis of a language disorder diagnosed by an institution such as a hospital, may be further included in the member information stored in the member information DB 140a. In addition, the degree of language impairment may be updated from time to time as the user goes through language training.

According to this embodiment, the degree of language impairment may be the result of diagnosing the language impairment diagnosed by an institution such as the hospital as described above, but it is not necessarily the result of the diagnosis of the language impairment. More specifically, if it is determined that language training is necessary even for a user who does not have a language trouble diagnosis result, a reference word extracted from a reference word DB 140b described later is presented and the user is pronounced , And the degree of the language disorder can be grasped by using the pronunciation result.

The reference word DB 140b is a database for storing the reference words previously set for use in language training of the user in the language disorder training server 100 in a database. Here, it is preferable that the reference words are set differently according to the degree of the language disorder of the user. Although the term "reference word" is used in this embodiment, it is not necessarily limited to words. For example, the higher the level of language training, the more likely it is that the reference word will be provided as a sentence outside the level of the word.

In addition, the reference word DB 140b is a group of relatively easy pronunciation words for users whose degree of language impairment is severe, a group of difficult pronunciation words for users whose degree of language impairment is not severe, Level word group for users who are in the middle level.

The similar word DB 140c is a database that stores similar words that are different from the correct pronunciation of the reference word based on the reference word, that is, different words that are improperly pronounced.

There are a plurality of similar words for one reference word. The reason for this is that different pronunciations are displayed for one reference word depending on the degree of the user's language disorder.

The data stored in the similar word DB 140c is shown in Table 2. Also, the similar word DB 140c can be updated every time when the pronunciation data by the user's voice is input from the user.

Reference word: sky










A similar word Degree of agreement Hanul 90% Harbor 90% Han 90% One day 90% ... ... Hanul 80% One day 80% ... ... Hach 70% The 70% ... ... Standard word: sea




Receive 90% receive 90% ... ... Baja 70% Bar 60% ... ...

The training result DB 140d is a database that stores the results of language disability training for each user. More specifically, the user inputs a pronunciation of a reference word provided from the language disorder training server 100 through the user terminal 200 according to a predetermined language training cycle, And the language impairment training result to be acquired using the pronunciation data is stored in the training result DB 140d.

The data stored in the training result DB 140d is shown in Table 3.

User A



Reference word Input value Degree of agreement 1st round sky The 70% Second round sky Hanul 80% Three times sky Hanul 90% Four times sky sky 100% User B

1st round sky The 70% Second round sky One day 90% ... ... ... ...

As shown in Table 3, the user A has undergone a total of four consecutive language training steps, and it can be seen that the user A has improved to a level of 100% matching at a level of 70% matching with the reference word "sky" . However, language training is not completed for user A. That is, the user A has perfect pronunciation for the reference word "sky ", but may not have improved for other words. Therefore, it is desirable that language training be continued until user A has perfect pronunciation for other reference words.

In the present embodiment, it is explained that the language disorder training server 100 constructs the member information DB 140a, the reference word DB 140b, the similar word DB 140c, and the training result DB 140d, But is not limited thereto. Depending on the amount and type of information, the database can be further subdivided or consolidated.

As described above, according to the speech disturbance training system according to the preferred embodiment of the present invention, various types of pronunciation data acquired from a user having a plurality of language impairments are converted into a database, and analysis result By again managing the database, it is possible to provide the user with accurate information related to the language training.

4 is a flowchart illustrating a method for training a language disorder according to a preferred embodiment of the present invention.

Hereinafter, a method for training a language disorder according to a preferred embodiment of the present invention will be described with reference to FIGS. 1 to 4. FIG.

The DB management unit 140 extracts a reference word to be provided to the user from the reference word DB 140b and outputs the extracted reference word to the corresponding user via the user interface unit 110. [ To the corresponding user terminal 200 (S301).

After confirming the presented reference word through the user terminal 200, the user directly pronounces the reference word through his or her own voice and transmits the pronunciation to the language obstacle training server 100 through the user terminal 200 (S303).

When the pronunciation data of the user's voice is input from the user terminal 200 to the user interface unit 110 in operation S305, the user interface unit 110 provides the input pronunciation data to the conversion unit 120 in operation S307 ).

The conversion unit 120 converts the pronunciation data provided from the user interface unit 110 into text by the STT module in step S309 and provides text for the pronunciation data to the analysis unit 130 in step S311.

The analysis unit 130 analyzes the similarity of the pronunciation data using the similar word related information stored in the similar word DB 140c in association with the DB management unit 140 (S313).

The analysis unit 130 provides the analysis result obtained by comparing and analyzing the similar words of the reference word and the similar word DB 140c to the DB management unit 140 at step S315 and the DB management unit 140 The analysis result is reflected in the similar word DB 140c to update the similar word DB 140c (S317).

Thereafter, the DB management unit 140 provides updated information of the similar word DB 140c to the information providing unit 150 (S319), and the information providing unit 150 uses the latest information of the similar word DB 140c And generates a training report for the user (S321).

The user training report generated by the information providing unit 150 is transmitted to the user terminal 200 through the user interface unit 110 (S323).

According to this procedure, the user can know the correct information about the pronunciation inputted by the user according to the language training cycle, and can also know whether or not the language training is being performed effectively. Moreover, user training reports can help the user actively plan for future language training.

100: language disorder training server 110: user interface section
120: conversion unit 130:
140: DB management unit 150:
160:

Claims (12)

A user interface unit for providing a reference word to a user terminal according to a predetermined language training cycle and receiving pronunciation data for the reference word;
A conversion unit for converting the input pronunciation data into text;
An analyzer for comparing the converted text with pre-stored similar words to analyze the degree of matching;
A DB management unit for accumulating the analyzed degrees of agreement to construct a training result database; And
And providing information related to language training using the established training result database,
Wherein the information providing unit generates information related to the language training including a treatment process, a treatment effect, prediction information on future training results, and information on setting and presentation of a treatment direction for a training period during the training period Language disability training server. The method according to claim 1,
Wherein the conversion unit is a Speech to Text (STT) conversion module. The method according to claim 1,
Wherein the DB management unit includes an unsuitable word database for storing and managing a plurality of unsuitable similar words corresponding to inappropriate pronunciation for each of the reference words in a step corresponding to the matching degree. The method of claim 3,
Wherein the analyzing unit analyzes the degree of matching using data stored in the nonconforming word database. delete The method according to claim 1,
Wherein the information providing unit gives a training weight value when the analyzed degree of coincidence is higher than a predetermined achievement level. Providing a reference word to a user terminal in accordance with a predetermined training period;
Receiving pronunciation data for the reference word from the user terminal;
Converting the input pronunciation data into text;
Comparing the converted text with pre-stored similar words to analyze the degree of matching;
Accumulating the analyzed degrees of agreement to construct a training result database; And
And providing information related to language training using the established training results database,
The step of providing information related to the language training may include providing information related to the language training, including information about a treatment process, a treatment effect, prediction information on future training results, Wherein the language obstacle training server generates the language disorder training using the language disorder training server. 8. The method of claim 7,
Wherein the converting step uses an STT (Speech to Text) conversion module. 8. The method of claim 7,
And constructing an unsuitable word database for storing and managing a plurality of unsuitable similar words corresponding to inappropriate pronunciation for each of the reference words in stages according to the degree of matching. Way. 10. The method of claim 9,
Wherein the analyzing step analyzes the degree of matching using data stored in the nonconforming word database. delete 8. The method of claim 7,
Wherein the providing step provides the training weight value when the analyzed degree of agreement is higher than the predetermined achievement level.

Images