KR20230000175A - Method for evaluating pronunciation based on AI, method for providing study content for coaching pronunciation, and computing system performing the same - Google Patents

Abstract

Disclosed are a method of collecting voice data which a learner pronounces on a word-by-word basis to determine the pronunciation accuracy of each phoneme, a method of providing learning content to allow learners to practice learners' weak pronunciation using the same, and a computing system performing the same. According to one aspect of the present invention, a method of evaluating phoneme-level pronunciation accuracy to calculate a phoneme-level score for each phoneme constituting the sound system of a given language, comprises the steps in which: an evaluation system acquires voice data of each of a plurality of words to be evaluated spoken by a person being evaluated; the evaluation system calculates, for each of the plurality of words to be evaluated, a word-level score corresponding to the word to be evaluated, based on the voice data of the word to be evaluated; and the evaluation system determines, for each of phonemes to be evaluated, at least part of the total phonemes constituting the sound system of the language, a phoneme-level score corresponding to the phoneme to be evaluated. In the step of determining the phoneme-level score corresponding to the phoneme to be evaluated includes a step of determining a phoneme-level score corresponding to the phoneme to be evaluated based on the word-level score of each word including the phoneme to be evaluated among the plurality of words to be evaluated spoken by the person being evaluated.

Description

Method for evaluating pronunciation based on AI, method for providing study content for coaching pronunciation, and computing system performing the same}

The present invention relates to a method for evaluating pronunciation accuracy in units of phonemes, a method for providing learning contents for pronunciation coaching, and a computing system for performing the same. More specifically, a method for determining pronunciation accuracy in phoneme units by collecting voice data that learners pronounce in word units, and a method for providing learning contents that enable learners to practice weak pronunciation using the method and performing the same It is about a computing system that

In general, modern people tend to show an increased interest in foreign languages according to the trend of industrial specialization and internationalization. In particular, various methods for evaluating the accuracy of foreign language pronunciation by non-native speakers are being developed, and software that informs the accuracy of pronunciation when a learner pronounces it is also commercially available. In the case of a foreign language pronunciation evaluation method, there is, for example, a pronunciation comparison method using voice signal processing technology. Representatively, a hidden Markov model is used to recognize the pronunciation voice of a non-native speaker, and then the native speaker's voice and an algorithm that compares .

Meanwhile, in evaluating the pronunciation of a foreign language, it is relatively easy to evaluate the accuracy of pronunciation at the word level, and various techniques have been proposed. An example of such technology may be mentioned such as Korean Patent Registration No. 10-2052031.

However, in learning the pronunciation of a foreign language, it is quite difficult to determine the accuracy of pronunciation at the phoneme level, although it is a very effective method to determine the phonemes that learners cannot pronounce accurately and practice pronunciation with a focus on those phonemes. to be.

Registered Patent No. 10-2052031

A technical problem to be achieved by the present invention is to provide a method for evaluating pronunciation accuracy in units of phonemes, a method for providing learning contents for pronunciation coaching, and a computing system for performing the same.

According to one aspect of the present invention, as a phoneme-unit pronunciation accuracy evaluation method for calculating phoneme level scores for each phoneme constituting the sound system of a predetermined language, the evaluation system comprises a plurality of evaluation targets uttered by an appraiser. acquiring voice data of each word; calculating, by the evaluation system, a word level score corresponding to the evaluation target word, based on voice data of the evaluation target word, for each of the plurality of evaluation target words; and determining, by the evaluation system, a phoneme level score corresponding to the phoneme to be evaluated for each of a plurality of phonemes to be evaluated, which are at least a part of all phonemes constituting the sound system of the language. In the step of determining a phoneme level score corresponding to the phoneme level corresponding to the phoneme to be evaluated, based on a word level score of each word including the phoneme to be evaluated among the plurality of words to be evaluated uttered by the subject, the phoneme level corresponding to the phoneme to be evaluated. A method for evaluating pronunciation accuracy in units of phonemes, including determining a score, is provided.

In an embodiment, the step of determining a phoneme level score corresponding to the phoneme to be evaluated based on a word level score of each word including the phoneme to be evaluated among the plurality of words to be evaluated uttered by the person to be evaluated, calculating an average value of word level scores of each word including the phoneme to be evaluated among the plurality of words to be evaluated; and determining a phoneme level score corresponding to the phoneme to be evaluated based on an average value of word level scores of each word including the phoneme to be evaluated among the plurality of words to be evaluated.

In an embodiment, the calculating of the word level score corresponding to the evaluation target word may include calculating a pronunciation score indicating accuracy of pronunciation of the evaluation target word uttered by the evaluation target; calculating an intonation score indicating accuracy of intonation of the word to be evaluated uttered by the subject of evaluation; calculating a volume score representing accuracy of the volume of the evaluation target word uttered by the subject; Calculating a length score representing the accuracy of the long and short length of the word to be evaluated uttered by the subject of evaluation; and calculating a stress score representing the accuracy of the stress of the word to be evaluated uttered by the person to be evaluated.

In an embodiment, the obtaining of audio data of each of a plurality of evaluation target words uttered by the evaluation subject may include acquiring audio data of a plurality of evaluation target sentences uttered by the evaluation subject; and dividing the audio data of the evaluation target sentence into words for each of the plurality of evaluation target sentences, wherein the plurality of evaluation target words include each word included in the plurality of evaluation target sentences. can

According to another aspect of the present invention, as a phoneme-unit pronunciation accuracy evaluation method for calculating a phoneme level score for each phoneme constituting the sound system of a predetermined language, the evaluation system comprises a plurality of evaluations uttered by an appraiser. obtaining voice data of each target word; and updating, by the evaluation system, a phoneme level score of the subject of evaluation based on the speech data of the word to be evaluated whenever voice data of each of the plurality of words to be evaluated is obtained. The step of updating the phoneme level score of the subject of evaluation based on the voice data of the evaluation target may include: calculating a word level score corresponding to the word to be evaluated based on the voice data of the word to be evaluated; determining at least one evaluation target phoneme constituting the evaluation target word; and for each of the phonemes to be evaluated, a word level score corresponding to the word to be evaluated is reflected in the pre-calculated phoneme level score corresponding to the phoneme to be evaluated of the subject to be evaluated. A method for evaluating phoneme-by-phoneme pronunciation accuracy, which includes updating a phoneme level score, is provided.

According to another aspect of the present invention, the learning system extracts a phoneme to be learned based on the phoneme level score for each phoneme established by the above-described method for evaluating pronunciation accuracy in units of phonemes; selecting, by the learning system, a word including the phoneme to be learned as the learning word; and providing, by the learning system, the words for learning or the contents for learning including the words for learning to a learner terminal.

In an embodiment, the learning content providing method may include selecting, by the learning system, a word including a replacement phoneme that is replaced with the learning target phoneme as a word for additional learning; and providing, by the learning system, the additional learning words or additional learning contents including the additional learning words to the learner terminal.

According to another aspect of the present invention, a computer program installed in a data processing device and recorded on a medium for performing the above method is provided.

According to another aspect of the present invention, a computer readable recording medium on which a computer program for performing the above method is recorded is provided.

According to another aspect of the present invention, a computing system comprising: a processor; and a memory storing a computer program, wherein the computer program, when executed by the processor, causes the computing system to perform the above-described method.

According to another aspect of the present invention, a phoneme-unit pronunciation accuracy evaluation system for calculating phoneme level scores for each phoneme constituting the sound system of a predetermined language, wherein each of a plurality of evaluation target words uttered by an appraiser an acquisition module for obtaining voice data; a word level evaluation module for calculating a word level score corresponding to the evaluation target word based on voice data of the evaluation target word for each of the plurality of evaluation target words; and a phoneme level evaluation module for determining a phoneme level score corresponding to the phoneme to be evaluated for each of a plurality of phonemes to be evaluated, which are at least a part of all phonemes constituting the sound system of the language, wherein the phoneme level evaluation module comprises: , In order to determine a phoneme level score corresponding to the phoneme to be evaluated, based on the word level score of each word including the phoneme to be evaluated among the plurality of words to be evaluated uttered by the subject, A phoneme-by-phoneme pronunciation accuracy evaluation system is provided that determines a corresponding phoneme level score.

In one embodiment, the phoneme level evaluation module calculates an average value of word level scores of each of the words including the phoneme to be evaluated among the plurality of words to be evaluated, uttered by the subject, and the plurality of words to be evaluated. A phoneme level score corresponding to the phoneme to be evaluated may be determined based on an average value of word level scores of each word including the phoneme to be evaluated.

In an embodiment, the word level evaluation module may include a pronunciation score indicating accuracy of pronunciation of the word to be evaluated uttered by the subject, an intonation score indicating accuracy of intonation of the word to be evaluated uttered by the subject, and the subject to be evaluated. A volume score indicating the accuracy of the volume of the word to be evaluated uttered, a length score indicating the accuracy of the length of the word to be evaluated uttered by the subject, and an accuracy of stress of the word to be evaluated uttered by the subject At least some of the indicated stress scores may be calculated.

In one embodiment, the acquisition module acquires audio data of a plurality of evaluation target sentences uttered by the person being evaluated, and divides the audio data of the evaluation target sentence by word for each of the plurality of evaluation target sentences; The plurality of evaluation target words may include each word included in the plurality of evaluation target sentences.

According to another aspect of the present invention, a phoneme-unit pronunciation accuracy evaluation system for calculating phoneme level scores for each phoneme constituting the sound system of a predetermined language, wherein each of a plurality of evaluation target words uttered by an appraiser an acquisition module for obtaining voice data; and an evaluation module configured to update a phoneme level score of the subject of evaluation based on the speech data of the evaluation target word whenever voice data of each of the plurality of evaluation target words is acquired. A word level evaluation module for calculating a word level score corresponding to the evaluation target word based on the voice data of the evaluation target word; a phoneme determination module for determining at least one evaluation target phoneme constituting the evaluation target word; and for each of the phonemes to be evaluated, a word level score corresponding to the word to be evaluated is reflected in the pre-calculated phoneme level score corresponding to the phoneme to be evaluated of the subject to be evaluated. A phoneme unit pronunciation accuracy evaluation system including a phoneme level evaluation module for updating a phoneme level score is provided.

According to another aspect of the present invention, a determination module for extracting a phoneme to be learned based on a phoneme level score for each phoneme established by the phoneme-by-phoneme pronunciation accuracy evaluation method; a selection module for selecting a word including the phoneme to be learned as the learning word; and a provision module for providing the learning words or the learning contents including the learning words to a learner terminal.

In an embodiment, the selection module further selects a word including a replacement phoneme that is replaced with the target phoneme as an additional learning word, and the providing module further selects the additional learning word or the additional learning word including the additional learning word. Content for learning may be further provided to the learner terminal.

According to the technical idea of the present invention, it is possible to determine pronunciation accuracy in phoneme units by collecting voice data that a learner pronounces in word units. That is, by calculating a score representing a score of a phoneme unit using a word-level evaluation score that is relatively easy to judge, there is an effect of easily grasping the pronunciation accuracy of the subject to be evaluated.

In addition, it is possible to provide learning contents that enable learners to practice weak pronunciation by using this.

In order to more fully understand the drawings cited in the detailed description of the present invention, a brief description of each drawing is provided.
1 is a diagram schematically illustrating an environment in which a method for evaluating pronunciation accuracy in phoneme units and a method for providing learning contents for pronunciation coaching according to the technical idea of the present invention are performed.
2 is a flowchart illustrating an example of a method for evaluating pronunciation accuracy in units of phonemes according to an embodiment of the present invention.
3 is a flowchart illustrating a specific example of step S110 of FIG. 2 .
FIG. 4A is a diagram showing word level scores of each of the eight evaluation target words W ₁ to W ₈ and phonemes constituting the corresponding words.
4B is a chart showing examples of equations for calculating phoneme level scores for each of the phonemes P ₁ to P ₅ .
4C is a chart showing an example of another equation for calculating phoneme level scores for each of the phonemes P ₁ to P ₅ .
5 is a flowchart illustrating an example of a method for evaluating pronunciation accuracy in units of phonemes according to another embodiment of the present invention.
FIG. 6 is a chart showing changes in phoneme level scores of evaluation target phones P ₁ to P ₅ updated by the method of FIG. 5 .
7 is a flowchart illustrating an example of a method of providing learning contents for pronunciation coaching according to an embodiment of the present invention.
8 is a block diagram showing a schematic configuration of an evaluation system according to an embodiment of the present invention, and FIG. 9 is a block diagram showing a schematic configuration of a learning system according to an embodiment of the present invention.

Since the present invention can apply various transformations and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, it should be understood that this is not intended to limit the present invention to specific embodiments, and includes all transformations, equivalents, and substitutes included in the spirit and scope of the present invention. In describing the present invention, if it is determined that a detailed description of related known technologies may obscure the gist of the present invention, the detailed description will be omitted.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. Terms such as first and second do not indicate a particular order, and are used only for the purpose of distinguishing one element from another.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise.

In this specification, terms such as "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other It should be understood that the presence or addition of features, numbers, steps, operations, components, parts, or combinations thereof is not precluded.

In addition, in the present specification, when one component 'transmits' data to another component, the component may directly transmit the data to the other component, or through at least one other component. It means that the data can be transmitted to the other component. Conversely, when one component 'directly transmits' data to another component, it means that the data is transmitted from the component to the other component without going through the other component.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings, focusing on embodiments of the present invention. Like reference numerals in each figure indicate like elements.

1 is a diagram schematically illustrating an environment in which a method for evaluating pronunciation accuracy in phoneme units and a method for providing learning contents for pronunciation coaching according to the technical idea of the present invention are performed.

Referring to FIG. 1 , a method for evaluating pronunciation accuracy in units of phonemes according to an embodiment of the present invention may be performed by a system for evaluating pronunciation accuracy in units of phonemes (100; hereinafter referred to as 'evaluation system'). A method of providing learning contents for pronunciation coaching according to an embodiment may be performed by a system 200 (hereinafter referred to as a 'learning system') for providing learning contents for pronunciation coaching.

The evaluation system 100 may calculate a phoneme level score for each phoneme constituting the sound system of a predetermined language.

The language may be typically English, but is not limited thereto and may be various such as Chinese, Japanese, Korean, French, Spanish, and the like. Hereinafter, description will be given mainly in English, but the technical idea of the present invention is not limited thereto.

On the other hand, a phoneme may be the smallest unit of sound recognized by a language user, which is distinguished from other sounds within the sound system of a specific language and plays an opposing function. For example, phonemes may include consonants and vowels, and vowels may include long vowels, short vowels, diphthongs, and the like. Since phonemes that can be pronounced in a specific language are already widely known, a detailed description thereof will be omitted. Conventionally, each phoneme may be represented by a phonetic symbol that is distinguished from each other, and the international phonetic alphabet is a representative example of the phonetic symbol.

The evaluation system 100 may obtain a plurality of voice data uttered by the subject from the user terminal (eg, 20) of the subject, and calculate a phoneme level score for each phoneme based on the obtained speech data. can

In an embodiment, the evaluation system 100 may generate score information (eg, a table) for each phoneme in which each phoneme constituting the sound system of the language is mapped to a phoneme level score of the corresponding phoneme. Meanwhile, the evaluation system 100 may separately configure score information for each phoneme for each user (eg, a subject to be evaluated, a student to be learned, etc.).

In one embodiment, the evaluation system 100 may store the generated score information for each phoneme in the database 300 (DB). The DB 300 may store various types of information required to implement the technical idea of the present invention in addition to score information for each phoneme. For example, the DB 300 may store various words or sentences to be used for phoneme evaluation or pronunciation learning, and may store mapping relationships between phonemes and words or sentences. In addition, the DB 300 may further store information on replacement phonemes corresponding to specific phonemes. In this specification, DB may be implemented as at least one table, and may be used to mean further including a separate database management system (DBMS) for searching, storing, and managing information stored in the DB. In addition, it can be implemented in various ways such as linked-list, tree, relational DB, etc., and includes all data storage media and data structures capable of storing information to be stored in the DB. can be used

Meanwhile, the learning system 200 uses the score information for each phoneme established by the evaluation system 100 to generate learning content that allows the user to learn weak pronunciation, and creates content for learning using the learner's user terminal (for example, , 20).

For example, the learning system 200 uses the score information for each phoneme established by the evaluation system 100 to determine a phoneme that is likely to require learning, and then sends a learning word or a learning sentence including the phoneme to the user terminal. (20) can be provided. Then, the user terminal 20 may acquire voice data for the learning word or learning sentence uttered by the user and transmit it to the learning system 200, and the learning system 200 may evaluate it.

On the other hand, the evaluation system 100 and/or the learning system 200 may be a computing system that is a data processing device having an arithmetic capability for implementing the technical concept of the present invention, and is generally connectable by a client through a network. It may include a computing device such as a personal computer or portable terminal as well as a server that is a data processing device.

The evaluation system 100 and/or the learning system 200 may be implemented with any one physical device, but a plurality of physical devices are organically combined as needed to form the evaluation system according to the technical idea of the present invention ( 100) and/or the learning system 200 can be implemented easily by an average expert in the art of the present invention.

As shown in FIG. 1 , the evaluation system 100 and the learning system 200 may be implemented as subsystems of a predetermined mother system 10 . The parent system 10 may be a server. The server 10 refers to a data processing device having computing power for implementing the technical idea of the present invention, and generally provides specific services such as personal computers and mobile terminals as well as data processing devices accessible to clients through a network. An average expert in the art of the present invention will readily infer that any device capable of performing can be defined as a server.

As shown in FIG. 1, the evaluation system 100 and the learning system 200 may be implemented in separate forms, but according to an embodiment, the evaluation system 100 and the learning system 200 It can also be implemented as a single system.

The user terminal 200 may be a computing device having a processing device. For example, the user terminal 200 may be a mobile device such as a smart phone, a mobile phone, a PDA, or a tablet PC, but is not limited thereto, and may include a laptop, a desktop, and the like. Meanwhile, the user terminal 200 may include a microphone capable of receiving a speaker's voice, a display device capable of outputting contents for learning provided by the learning system 100, and a speaker.

On the other hand, the evaluation system 100 and/or the learning system 200 are connected to the user terminal 20 through a wired/wireless network (eg, Internet), and various information required to implement the technical idea of the present invention. , data and / or signals can be transmitted and received.

2 is a flowchart illustrating an example of a method for evaluating pronunciation accuracy in units of phonemes performed by the evaluation system 100 .

Referring to FIG. 2 , the evaluation system 100 may obtain voice data of each of a plurality of evaluation target words uttered by the subject (S110).

In one embodiment, the evaluation system 100 may obtain voice data of each of the plurality of evaluation target words from the user terminal (eg, 20) of the subject of evaluation. For example, the evaluation system 100 may transmit a word to be evaluated to the user terminal 20, the user terminal 20 may display the word to be evaluated, and the subject to be evaluated utters the word to the user terminal 20. It can be input through a microphone provided in the terminal 20 . Then, the user terminal 20 may transmit voice data corresponding to the voice input through the microphone to the evaluation system 100 .

The evaluation system 100 may acquire voice data input in units of sentences, an example of which is shown in FIG. 3 . 3 is a flowchart illustrating a specific example of step S110 of FIG. 2 .

Referring to FIG. 3 , the evaluation system 100 may provide an evaluation target sentence including a plurality of words to the user terminal 20 (S111), and the user terminal 20 outputs the evaluation target sentence. and the subject to be evaluated can utter the sentence to be evaluated and input it through a microphone. Then, the user terminal may transmit voice data corresponding to the voice of the evaluation target sentence uttered by the subject to the evaluation system 100 (S112).

Thereafter, the evaluation system 100 may divide the speech data of the evaluation target sentence by word (S113). A technology for classifying voice data corresponding to a sentence into words may be implemented through waveform analysis or machine learning technology. Since the technology itself of classifying voice data corresponding to a sentence into words is widely known, a detailed description thereof will be omitted in the present specification.

The evaluation system 100 may perform steps S111 to S113 for each of a plurality of evaluation target sentences. Accordingly, in the embodiment of FIG. 3 , the plurality of evaluation target words may be composed of each word included in the plurality of evaluation target sentences.

For example, the plurality of evaluation target sentences are T ₁ , T ₂ , and T ₃ , T ₁ is composed of words W ₁ , W ₂ , and W ₃ , and T ₂ is composed of words W ₄ , W ₅ , and W ₆ . , and T ₃ is composed of words W ₇ , W ₈ , W ₉ , and W ₁₀ , the words to be evaluated may be W ₁ to W ₁₀ .

Referring back to FIG. 2 , the evaluation system 100 may obtain word level scores corresponding to each of a plurality of words to be evaluated (S120). In other words, the evaluation system 100 may calculate a word level score corresponding to the evaluation target word based on voice data of the evaluation target word for each of the plurality of evaluation target words.

According to embodiments, the word level score may include scores of detailed items. For example, the word level score of the word to be evaluated includes a pronunciation score indicating accuracy of pronunciation of the word to be evaluated uttered by the subject, an intonation score indicating accuracy of intonation of the word to be evaluated uttered by the subject, the A volume score indicating the accuracy of the volume of the word to be evaluated uttered by the subject, a length score indicating the accuracy of the length of the word to be evaluated uttered by the subject, and an accuracy of stress of the word to be evaluated uttered by the subject It may include at least some of the stress scores representing , and the evaluation system 100 may separately perform a process of calculating scores for each subitem.

Alternatively, the word level score of the word to be evaluated may be a value calculated as a weighted average of at least some of pronunciation scores, intonation scores, volume scores, length scores, and stress scores.

A method in which the evaluation system 100 calculates a word level score corresponding to a specific word using voice data of a specific word may be various. For example, the evaluation system 100 may compare the degree of agreement by analyzing the waveform of voice data of the corresponding word and comparing it with the waveform of predetermined answer voice data. In addition to this, there are various methods that can calculate word level scores. Since the technology of determining word-level accuracy or score based on voice data is not a key feature of the present invention and is a well-known technology, a detailed description thereof will be omitted herein.

Thereafter, the evaluation system 100 may determine a phoneme level score corresponding to the phoneme to be evaluated for each of a plurality of phonemes to be evaluated, which are all phonemes or a part of all phonemes constituting the sound system of the language (S130).

More specifically, the evaluation system 100 determines the phoneme level score corresponding to the phoneme to be evaluated based on the word level score of each word including the phoneme to be evaluated among the plurality of words to be evaluated uttered by the person to be evaluated. It can be determined (S140).

In an embodiment, the evaluation system 100 may calculate an average value of word level scores of each word including the phoneme to be evaluated among the plurality of words to be evaluated, and the calculated average value may be used as the phoneme to be evaluated. It can be determined as a phoneme level score corresponding to , which will be described in more detail with reference to FIGS. 4A and 4B. For convenience of explanation, it is assumed in FIGS. 4A and 4B that all phonemes constituting the sound system of language are _P1 to P5.

FIG. 4A is a table showing word level scores of each of eight evaluation target words W ₁ to W ₈ and phonemes constituting the word, and FIG. 4B is an equation for calculating phoneme level scores of each of phonemes P ₁ to P ₅ . is a table showing

As shown in FIG. 4A , each evaluation target word W ₁ to W ₈ may be composed of phonemes or phonemes, and the phonemes or phonemes constituting the evaluation target word are predefined. Therefore, the evaluation system 100 may store phonemes constituting the candidate word in advance for each candidate word that may be an evaluation target word, and the mapping relationship between the candidate word and the phonemes constituting the candidate word may be determined in the DB. (300) may be stored in advance.

Referring to FIG. 4B , since the phoneme P ₁ is included in W ₁ , W ₂ , and W ₄ of words W ₁ to W ₈ to be evaluated, the evaluation system 100 determines the word level score S ₁ , W of W ₁ . The average value of the word-level score S ₂ of ₂ and the word-level score S ₄ of W ₄ (ie, (S ₁ +S ₂ +S ₄ )/3) can be calculated and determined as the phoneme level score of P ₁ . can

Since the phoneme P ₂ is included in the words W ₁ , W ₅ , W ₆ , and W ₇ to be evaluated, the evaluation system 100 calculates the average value of their word level scores [(S ₁ +S ₅ +S ₆ +S ₇ )/4] can be determined as the phoneme level score of P ₂ .

Since the phoneme P ₃ is included in the words W ₃ , W ₄ , and W ₈ to be evaluated, the evaluation system 100 calculates [(S ₃ +S ₄ +S ₈ )/3], the average value of their word level scores, as P . It can be determined with a phoneme level score of ₃ .

Since the phoneme P ₄ is included in the words W ₂ , W ₃ , W ₅ , and W ₇ to be evaluated, the evaluation system 100 calculates the average value of their word level scores [(S ₂ +S ₃ +S ₅ +S ₇ )/4] can be determined as the phoneme level score of P ₄ .

Since the phoneme P ₅ is included in the words W ₂ , W ₃ , W ₅ , W ₆ , and W ₈ to be evaluated, the evaluation system 100 calculates the average value of their word level scores [(S ₂ +S ₃ +S ₅ +S ₆ +S ₈ )/5] can be determined as the phoneme level score of P ₅ .

Meanwhile, as shown in FIG. 4B, the evaluation system 100 may determine the average value of word level scores of each word including the phoneme to be evaluated as the phoneme level score corresponding to the phoneme to be evaluated. According to an example, a phoneme level score corresponding to the phoneme to be evaluated may be determined based on an average value of word level scores of each word including the phoneme to be evaluated among the plurality of words to be evaluated. For example, if the average value of word level scores of each word including the phoneme to be evaluated belongs to a specific numerical range, the evaluation system 100 assigns a score corresponding to the corresponding numerical interval to the phoneme level corresponding to the phoneme to be evaluated. It can also be decided by score. In addition to this, of course, there may be various methods for determining a phoneme level score corresponding to the phoneme to be evaluated.

In addition, depending on the embodiment, instead of calculating the average value of word level scores of each word including the phoneme to be evaluated, various values that can be calculated based on the word level score of each word including the phoneme to be evaluated are calculated. Based on this, a phoneme level score corresponding to the phoneme to be evaluated may be determined. For example, the evaluation system 100 calculates a weighted average in which the reciprocal of the number of phonemes included in each word including the phoneme to be evaluated is used as a weight, and based on this, a phoneme level score corresponding to the phoneme to be evaluated is calculated. It can also be determined, an example of which is shown in Figure 4c. The phoneme level score of FIG. 4C shows the case where the evaluation target words and their word level scores are in FIG. 4A, except that a weighted average is calculated with the reciprocal of the number of phonemes included in each word as a weight. Since it is the same as, detailed description will be omitted.

5 is a flowchart illustrating an example of a method for evaluating pronunciation accuracy in units of phonemes according to another embodiment of the present invention. In the description of FIG. 5 , detailed descriptions of contents that are redundant with the above-described contents or that can be obviously derived from the above-described contents will be omitted.

Referring to FIG. 5 , the evaluation system 100 may perform steps S210 to S240 for each of a plurality of words to be evaluated (S200).

The evaluation system 100 may acquire voice data of the word to be evaluated uttered by the person to be evaluated (S210).

The evaluation system 100 may update the phoneme level score of the subject of evaluation based on the speech data of each word to be evaluated whenever voice data of each word to be evaluated is obtained (S220 to S240).

More specifically, the evaluation system 210 may calculate a word level score corresponding to the evaluation target word based on voice data of the evaluation target word (S220).

In addition, the evaluation system 210 may determine at least one evaluation target phoneme constituting the evaluation target word (S230).

Thereafter, the evaluation system 210 calculates the phoneme level score corresponding to the phoneme to be evaluated for each of the at least one phoneme to be evaluated constituting the word to be evaluated, and the word corresponding to the word to be evaluated. By reflecting the level score, the phoneme level score corresponding to the evaluation target phoneme of the subject can be updated (S230 and S240). For example, the evaluation system 100 may calculate a cumulative average value of word level scores of each word including the phoneme to be evaluated among the words to be evaluated that have been input so far, and the phoneme level corresponding to the phoneme to be evaluated. The score can be updated with the calculated cumulative average value. Meanwhile, the initial phoneme level score of each phoneme to be evaluated may be set to 0.

When the voice data of each evaluation target word W ₁ to W ₈ in FIG. 4A is obtained sequentially, FIG. 6 shows changes in phoneme level scores of evaluation target phones P ₁ to P ₅ updated by the method of FIG. 5 .

Referring to FIGS. 4A and 6 , voice data of the word W ₁ to be evaluated is obtained in a state in which the phoneme level scores of the phonemes to be evaluated P ₁ to P ₅ are set to a predetermined initial value (eg, 0). (S210), the evaluation system 100 may determine the word level score of the evaluation target word W ₁ as S ₁ (S220), and at least one evaluation target phoneme P ₁ and P ₂ constituting the evaluation target word It is possible to determine (S230).

Then, the evaluation system 100 calculates the phoneme level score of the phoneme P ₁ to be evaluated as the word level of words (ie, W ₁ ) including the phoneme to be evaluated P ₁ among words to be evaluated (ie, W ₁ ) input so far. The cumulative average value of the score S ₁ = S ₁ /1 may be updated (S240). Similarly, the evaluation system 100 calculates the phoneme level score of the phoneme P ₂ to be evaluated as the word level of words (ie, W ₁ ) including the phoneme to be evaluated P ₂ among words to be evaluated (ie, W ₂ ) input so far. The cumulative average value of the score S ₁ = S ₁ /1 may be updated (S240).

Next, when voice data of the word W ₂ to be evaluated is obtained (S210), the evaluation system 100 may determine the word level score of the word W ₂ to be evaluated as S ₂ (S220), and configure the word to be evaluated. At least one evaluation target phoneme P ₁ , P ₄ , and P ₅ may be determined (S230).

Then, the evaluation system 100 assigns the phoneme level score of the phoneme to be evaluated P ₁ to words including _the phoneme to be evaluated P ₁ (ie, W _{1 ,} W ₂ ) may be updated to the cumulative average value of word level scores (S _{1 +} S ₂ )/2 (S240). In addition, the evaluation system 100 assigns the phoneme level score of the phoneme P ₄ to be evaluated to words including the phoneme P ₄ to be evaluated (ie, W ₂ ) among words to be evaluated (ie, W ₁ and W ₂ ) input so far. The cumulative average value of word level scores of S ₂ = S ₂ /1 may be updated (S240). In addition, the evaluation system 100 assigns the phoneme level score of the phoneme P ₅ to be evaluated to words including the phoneme P ₅ to be evaluated (ie, W ₂ ) among words to be evaluated (ie, W ₁ and W ₂ ) input so far. The cumulative average value of word level scores of S ₂ = S ₂ /1 may be updated (S240).

Next, when voice data of the word W ₃ to be evaluated is obtained (S210), the evaluation system 100 may determine the word level score of the word W ₃ to be evaluated as S ₃ (S220), and configure the word to be evaluated. At least one evaluation target phoneme P ₃ , P ₄ , and P ₅ may be determined (S230).

Then, the evaluation system 100 assigns the phoneme level score of the phoneme P ₃ to be evaluated to words including the phoneme P ₃ to be evaluated (ie, W _{1 ,} W _{2 ,} W ₃ ) among words to be evaluated (ie, W 1 , W 2 , and W 3 ) input so far. The cumulative average value of word level scores of W ₃ ) may be updated to S ₃ = S ₃ /1 (S240). In addition, the evaluation system 100 assigns the phoneme level score of the phoneme P ₄ to be evaluated to words including the phoneme P ₄ to be evaluated (ie, W ₁ , W _{2 ,} W ₃ ) among words to be evaluated (ie, W 1 , W 2 , and W 3 ) input so far. The cumulative average value of word level scores of W _{2 and} W ₃ ) may be updated to (S ₂ +S ₃ )/2 (S240). In addition, the evaluation system 100 assigns the phoneme level score of the phoneme P ₅ to be evaluated to words including the phoneme P ₅ to be evaluated (ie, W ₁ , W _{2 ,} W ₃ ) among words to be evaluated (ie, W 1 , W 2 , and W 3 ) input so far. The cumulative average value of word level scores of W _{2 and} W ₃ ) may be updated to (S ₂ +S ₃ )/2 (S240).

In the same way, the evaluation system 100 repeats the above process whenever evaluation target words W ₄ , W ₅ , W ₆ , W _{7 ,} and W ₈ are sequentially entered, and as shown in FIG. 6 , each evaluation target phoneme P ₁ The phoneme level scores of through P ₅ may be updated.

7 is a flowchart illustrating an example of a method of providing learning contents for pronunciation coaching performed by the learning system 200 .

Referring to FIG. 7 , the learning system 200 may extract one or more phonemes to be learned based on the phoneme level score for each phoneme generated by the evaluation system 100 (S310).

For example, the learning system 200 may determine a phoneme having the lowest phoneme level score as a learning target phoneme. Alternatively, the learning system 200 may determine one or more phonemes having a phoneme level score equal to or less than a predetermined reference value as a learning target phoneme.

Thereafter, the learning system 200 may select a word including the phoneme to be learned as a word for learning (S320). The learning system 200 may, for example, store a plurality of candidate words for learning in advance in the DB 300, and the learning system 200 selects a word including the phoneme to be learned from among the stored candidate words for learning. can be chosen by word. Depending on the embodiment, if the number of phonemes to be learned is two or more, the learning system 200 may select a word including as many phonemes to be learned as a learning word.

Depending on the embodiment, the learning system 200 may select learning content including the learning word. For example, the contents for learning may be sentences or paragraphs including the words for learning, or may be audio or video data corresponding to the words for learning. Alternatively, it may be audio data or video data corresponding to a sentence or paragraph including the learning word.

All contents for learning that can be selected by the learning system 200 may be created in advance and stored in the DB 300, for example. If there are many contents including the words for learning among the pre-stored contents, the learning system 200 may select the contents for learning by referring to the learner's previous learning history.

Thereafter, the learning system 200 may provide the learning word or the learning content including the learning word to the learner terminal (eg, 20) (S330). Then, the learner terminal 20 may perform a listening test or a pronunciation test based on the provided learning word or learning contents including the learning word.

Meanwhile, according to an embodiment, the learning system 20 may further provide learning content for a replacement phoneme that can be compared with the target phoneme to the learner terminal so that learning of the target phoneme can be performed more effectively. . The replacement phoneme corresponding to each phoneme may be designated in advance, for example, the replacement phoneme of the phoneme 'r' may be designated as 'l', and the replacement phoneme of the phoneme 'f' may be designated as 'p' may have been

With continued reference to FIG. 7 , the learning system 20 may select a word for additional learning that includes a replacement phoneme that is replaced with a phoneme to be learned (S340), and the word for additional learning or the word for additional learning is included. Additional learning content may be further provided to the learner terminal (S350).

8 is a block diagram showing a schematic configuration of an evaluation system 100 according to an embodiment of the present invention, and FIG. 9 shows a schematic configuration of a learning system 200 according to an embodiment of the present invention. It is a block diagram.

The evaluation system 100 and the learning system 200 may refer to a logical configuration having hardware resources and/or software required to implement the technical idea of the present invention, and must be one physical It does not mean a component or a device. That is, the evaluation system 100 and the learning system 200 may refer to a logical combination of hardware and/or software provided to implement the technical idea of the present invention, and if necessary, may be installed in devices spaced apart from each other. It may be implemented as a set of logical components for implementing the technical idea of the present invention by being installed and performing each function. In addition, the evaluation system 100 and the learning system 200 may refer to a set of components implemented separately for each function or role to implement the technical idea of the present invention. Each component of the evaluation system 100 and the learning system 200 may be located in different physical devices or in the same physical device. In addition, depending on the implementation example, the combination of software and/or hardware constituting each of the components of the evaluation system 100 and the learning system 200 is also located in different physical devices, and components located in different physical devices Each of the modules may be organically combined with each other.

Also, in this specification, a module may mean a functional and structural combination of hardware for implementing the technical concept of the present invention and software for driving the hardware. For example, the module may mean a logical unit of a predetermined code and a hardware resource for executing the predetermined code, and does not necessarily mean a physically connected code or one type of hardware. can be easily deduced to an average expert in the art of the present invention.

Referring to FIG. 8 , the evaluation system 100 may include an acquisition module 110 and an evaluation module 115, and the evaluation module 115 includes a word level evaluation module 120 and a phoneme determination module ( 130), and a phoneme level evaluation module 140. Depending on the embodiment of the present invention, some of the above-described components may not necessarily correspond to the components essential to the implementation of the present invention, and according to the embodiment, the evaluation system 100 may It goes without saying that more components may be included. For example, the evaluation system 100 includes a communication module (not shown) for communicating with an external device, a control module (not shown) for controlling components and resources of the evaluation system 100, and various information and data. A storage module (not shown) for storage may be further included.

According to an embodiment of the present invention, the acquisition module 110 may acquire voice data of each of a plurality of evaluation target words uttered by a person to be evaluated, and the word level evaluation module 120 may obtain the plurality of evaluation target words. For each word, a word level score corresponding to the word to be evaluated may be calculated based on the voice data of the word to be evaluated, and the phoneme level evaluation module 140 may calculate a word level score among all phonemes constituting the sound system of the language. A phoneme level score corresponding to the phoneme to be evaluated may be determined for each of at least some of the plurality of phonemes to be evaluated. In particular, in order to determine a phoneme level score corresponding to the phoneme to be evaluated, the phoneme level evaluation module 140 determines the word level of each word including the phoneme to be evaluated among the plurality of words to be evaluated uttered by the subject to be evaluated. Based on the score, a phoneme level score corresponding to the phoneme to be evaluated may be determined.

In an embodiment, the phoneme level evaluation module 140 calculates an average value of word level scores of each word including the phoneme to be evaluated among the plurality of words to be evaluated uttered by the person to be evaluated, and the plurality of evaluation targets. A phoneme level score corresponding to the phoneme to be evaluated may be determined based on an average value of word level scores of each word including the phoneme to be evaluated among target words.

In an embodiment, the word level evaluation module may include a pronunciation score indicating accuracy of pronunciation of the word to be evaluated uttered by the subject, an intonation score indicating accuracy of intonation of the word to be evaluated uttered by the subject, and the subject to be evaluated. A volume score indicating the accuracy of the volume of the word to be evaluated uttered, a length score indicating the accuracy of the length of the word to be evaluated uttered by the subject, and an accuracy of stress of the word to be evaluated uttered by the subject At least some of the indicated stress scores may be calculated.

In one embodiment, the acquisition module acquires voice data of a plurality of evaluation target sentences uttered by the subject, and divides the voice data of the evaluation target sentence by word for each of the plurality of evaluation target sentences; The plurality of evaluation target words may include each word included in the plurality of evaluation target sentences.

According to another embodiment of the present invention, the evaluation module 115 is configured so that, whenever voice data of each of the plurality of words to be evaluated is obtained, the evaluation system determines the person to be evaluated based on the voice data of the word to be evaluated. As described above, the evaluation module 115 may include a word level evaluation module 120, a phoneme determination module 130, and a phoneme level evaluation module 140. The word level evaluation module 120 calculates a word level score corresponding to the evaluation target word based on the voice data of the evaluation target word, and the phoneme determination module 130 calculates at least one word constituting the evaluation target word. The phoneme level evaluation module 140 determines the phoneme to be evaluated, and the phoneme level evaluation module 140 determines the phoneme level score corresponding to the phoneme to be evaluated for each of the phonemes to be evaluated, and the word corresponding to the word to be evaluated is calculated in advance. A phoneme level score corresponding to the phoneme to be evaluated may be updated by reflecting the level score.

9 is a block diagram showing the configuration of a learning system 200 according to an embodiment of the present invention.

Referring to FIG. 9 , the learning system 200 may include a judgment module 210 , a selection module 220 and a provision module 230 . Depending on the embodiment of the present invention, some of the above-mentioned components may not necessarily correspond to the components necessary for the implementation of the present invention, and according to the embodiment, the learning system 200 has more It goes without saying that more components may be included. For example, the learning system 200 includes a communication module (not shown) for communicating with an external device, a control module (not shown) for controlling components and resources of the long frame estimation system 100, various data and information It may further include a storage module (not shown) capable of storing .

In addition, according to embodiments, the learning system 200 may be implemented in an integrated form with the evaluation construction system 100. In this case, the learning system 200 is a component included in the evaluation system 100 described above. may further include.

The determination module 210 may extract a phoneme to be learned based on the phoneme level score for each phoneme established by the evaluation system 100, and the selection module 220 may extract a word including the phoneme to be learned. can be selected as the learning word. The provision module 230 may provide the learning word or the learning content including the learning word to a learner terminal (eg, 20).

In an embodiment, the selection module 220 further selects a word including a replacement phoneme that is replaced with the target phoneme as a word for additional learning, and the providing module 230 further selects the word for additional learning or the word for additional learning. It is possible to further provide additional content for learning, which includes, to the learner terminal 20.

Meanwhile, according to an implementation example, the evaluation system 100 and the learning system 200 may include a processor and a memory storing a program executed by the processor. The processor may include a single-core CPU or a multi-core CPU. The memory may include high-speed random access memory and may also include non-volatile memory such as one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid-state memory devices. Access to memory by processors and other components may be controlled by a memory controller.

On the other hand, the method according to the embodiment of the present invention may be implemented in the form of computer-readable program instructions and stored in a computer-readable recording medium, and the control program and target program according to the embodiment of the present invention are also computer-readable. It can be stored on a readable recording medium. The computer-readable recording medium includes all types of recording devices in which data that can be read by a computer system is stored.

Program commands recorded on the recording medium may be specially designed and configured for the present invention, or may be known and usable to those skilled in the software field.

Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, floptical disks and hardware devices specially configured to store and execute program instructions, such as magneto-optical media and ROM, RAM, flash memory, and the like. In addition, the computer-readable recording medium is distributed in computer systems connected through a network, so that computer-readable codes can be stored and executed in a distributed manner.

Examples of program instructions include high-level language codes that can be executed by a device that electronically processes information using an interpreter, for example, a computer, as well as machine language codes generated by a compiler.

The hardware device described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

The above description of the present invention is for illustrative purposes, and those skilled in the art can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, the embodiments described above should be understood as illustrative in all respects and not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and equivalent concepts thereof should be construed as being included in the scope of the present invention. .

Claims (17)

A phoneme-unit pronunciation accuracy evaluation method for calculating a phoneme level score for each phoneme constituting the sound system of a predetermined language,
acquiring voice data of each of a plurality of evaluation target words uttered by an appraiser, by an evaluation system;
The evaluation system, for each of the plurality of evaluation target words,
Calculating a word level score corresponding to the evaluation target word based on voice data of the evaluation target word; and
determining, by the evaluation system, a phoneme level score corresponding to the phoneme to be evaluated, for each of a plurality of phonemes to be evaluated, which are at least a part of all phonemes constituting the sound system of the language;
The step of determining a phoneme level score corresponding to the phoneme to be evaluated,
and determining a phoneme level score corresponding to the phoneme to be evaluated based on a word level score of each word including the phoneme to be evaluated among the plurality of words to be evaluated. Assessment Methods.
According to claim 1,
The step of determining a phoneme level score corresponding to the phoneme to be evaluated based on a word level score of each word including the phoneme to be evaluated among the plurality of words to be evaluated uttered by the subject,
calculating an average value of word level scores of each word including the phoneme to be evaluated among the plurality of words to be evaluated; and
and determining a phoneme level score corresponding to the phoneme to be evaluated based on an average value of word level scores of each word including the phoneme to be evaluated among the plurality of words to be evaluated. .
According to claim 1,
Calculating a word level score corresponding to the word to be evaluated,
Calculating a pronunciation score representing accuracy of pronunciation of the word to be evaluated uttered by the person to be evaluated;
calculating an intonation score indicating accuracy of intonation of the word to be evaluated uttered by the subject of evaluation;
calculating a volume score representing accuracy of the volume of the evaluation target word uttered by the subject;
Calculating a length score representing the accuracy of the long and short length of the word to be evaluated uttered by the subject of evaluation; and
The method of evaluating pronunciation accuracy in units of phonemes, comprising at least a part of calculating a stress score representing the accuracy of the stress of the word to be evaluated uttered by the person to be evaluated.
According to claim 1,
Acquiring voice data of each of a plurality of words to be evaluated uttered by the person to be evaluated includes:
obtaining voice data of a plurality of evaluation target sentences uttered by the subject of evaluation; and
For each of the plurality of evaluation target sentences, dividing the audio data of the evaluation target sentence by word,
The plurality of evaluation target words include each word included in the plurality of evaluation target sentences.
A phoneme-unit pronunciation accuracy evaluation method for calculating a phoneme level score for each phoneme constituting the sound system of a predetermined language,
acquiring voice data of each of a plurality of evaluation target words uttered by an appraiser, by an evaluation system; and
Whenever voice data of each of the plurality of words to be evaluated is obtained, the evaluation system updating a phoneme level score of the subject based on the voice data of the word to be evaluated;
Updating the phoneme level score of the subject based on the voice data of the word to be evaluated,
Calculating a word level score corresponding to the evaluation target word based on voice data of the evaluation target word;
determining at least one evaluation target phoneme constituting the evaluation target word; and
For each of the phonemes to be evaluated,
and updating a phoneme level score corresponding to the phoneme to be evaluated by reflecting a word level score corresponding to the word to be evaluated in the pre-calculated phoneme level score corresponding to the phoneme to be evaluated of the evaluated person. A method for evaluating phoneme-unit pronunciation accuracy.
extracting, by a learning system, a phoneme to be learned based on a phoneme level score for each phoneme constructed by the method according to claim 1 or 5;
selecting, by the learning system, a word including the phoneme to be learned as the learning word; and
The learning content providing method comprising providing, by the learning system, the learning word or the learning content including the learning word to a learner terminal.
According to claim 6,
selecting, by the learning system, a word including a replacement phoneme that is replaced with the learning target phoneme as a word for additional learning; and
The learning content providing method further comprising providing, by the learning system, the additional learning words or additional learning contents including the additional learning words to the learner terminal.
A computer program installed in a data processing device and recorded on a medium for performing the method according to any one of claims 1 to 5.
A computer readable recording medium on which a computer program for performing the method according to any one of claims 1 to 5 is recorded.
As a computing system,
processor; and a memory for storing a computer program;
The computer program, when executed by the processor, causes the computing system to perform the method according to any one of claims 1 to 5.
A phoneme-unit pronunciation accuracy evaluation system for calculating a phoneme level score for each phoneme constituting the sound system of a predetermined language,
an acquisition module for obtaining voice data of each of a plurality of evaluation target words uttered by an appraiser;
a word level evaluation module for calculating a word level score corresponding to the evaluation target word based on voice data of the evaluation target word for each of the plurality of evaluation target words; and
A phoneme level evaluation module for determining a phoneme level score corresponding to the phoneme to be evaluated for each of a plurality of phonemes to be evaluated, which are at least a part of all phonemes constituting the sound system of the language,
The phoneme level evaluation module determines a phoneme level score corresponding to the phoneme to be evaluated,
The phoneme-unit pronunciation accuracy evaluation system for determining a phoneme level score corresponding to the phoneme to be evaluated based on a word level score of each word including the phoneme to be evaluated among the plurality of words to be evaluated.
According to claim 11,
The phoneme level evaluation module,
Calculating an average value of word level scores of each word including the phoneme to be evaluated among the plurality of words to be evaluated uttered by the subject;
A phoneme-unit pronunciation accuracy evaluation system for determining a phoneme level score corresponding to the phoneme to be evaluated based on an average value of word level scores of each word including the phoneme to be evaluated among the plurality of words to be evaluated.
According to claim 1,
The word level evaluation module,
A pronunciation score indicating the accuracy of pronunciation of the word to be evaluated uttered by the subject, an intonation score indicating accuracy of intonation of the word to be evaluated uttered by the subject, and accuracy of volume of the word to be evaluated uttered by the subject Phoneme unit pronunciation that calculates at least a part of a volume score representing a volume score, a length score representing accuracy of length of the word to be evaluated uttered by the subject, and a stress score representing accuracy of stress of the word to be evaluated uttered by the subject. Accuracy evaluation system.
According to claim 11,
The acquisition module,
Acquiring audio data of a plurality of evaluation target sentences uttered by the subject, and dividing the audio data of the evaluation target sentence by word for each of the plurality of evaluation target sentences;
The plurality of evaluation target words include phoneme unit pronunciation accuracy evaluation system including each word included in the plurality of evaluation target sentences.
A phoneme-unit pronunciation accuracy evaluation system for calculating a phoneme level score for each phoneme constituting the sound system of a predetermined language,
an acquisition module for obtaining voice data of each of a plurality of evaluation target words uttered by an appraiser; and
Whenever voice data of each of the plurality of words to be evaluated is acquired, the evaluation system includes an evaluation module for updating a phoneme level score of the subject of evaluation based on the voice data of the word to be evaluated;
The evaluation module,
a word level evaluation module for calculating a word level score corresponding to the evaluation target word based on the voice data of the evaluation target word;
a phoneme determination module for determining at least one evaluation target phoneme constituting the evaluation target word; and
For each of the phonemes to be evaluated,
Phoneme level evaluation to update the phoneme level score corresponding to the phoneme to be evaluated by reflecting the word level score corresponding to the word to be evaluated to the pre-calculated phoneme level score corresponding to the phoneme to be evaluated A phoneme unit pronunciation accuracy evaluation system including a module.
a decision module for extracting a phoneme to be learned based on the phoneme level score for each phoneme constructed by the method according to claim 1 or 5;
a selection module for selecting a word including the phoneme to be learned as the learning word; and
A learning content providing system comprising a provision module for providing the learning word or the learning content including the learning word to a learner terminal.
According to claim 6,
The selection module,
Further selecting a word including a replacement phoneme that is replaced with the learning target phoneme as a word for additional learning,
The providing module,
A learning content providing system for further providing the additional learning words or additional learning contents including the additional learning words to the learner terminal.

Images