KR101944675B1 - Voice training system through breathing measurement - Google Patents

Abstract

The present invention relates to a voice training system through breathing measurement for providing a customizing voice training in accordance with a physical condition of a user. The voice training system through breathing measurement comprises: a voice measurement device including a breathing measurer; a voice data measurement module including a condition input module which receives user information including a target breathing condition from a user device, a gender, an age, a weight and a job of the user; a voice data input unit which receives breathing pressure including expiratory pressure and inspiratory pressure from the voice measurement device as voice data; and a main server including a result analysis module which includes a data comparative and analysis unit calculating result data compared to the target breathing condition and the voice data and an information supply module which provides training information to the user through the user device in accordance with the result data.

Description

{VOICE TRAINING SYSTEM THROUGH BREATHING MEASUREMENT}

The present invention relates to a vocal training system through breath measurement. More specifically, the present invention relates to a vocal training system for training breathing, To a vocal training system through measurement.

Generally speaking, vocalization refers to giving a voice, which is accompanied by breathing every time a voice is made in the body structure. Therefore, vocal training is one of the most important issues for those who have a vocal voice like a vocalist or lecturer, and training breathing techniques is very helpful in making voices bigger and more widespread.

The conventional breathing training method was a method of training by changing the height of the air in the cylinder, or a training machine was developed and used so that the user can visually recognize the amount of breathing.

However, there is a disadvantage that such a breathing trainer must be moved to the location of the device.

In order to solve such a problem, Korean Patent Registration No. 10-1620992 entitled " Breathing Training System Using User Breath Pattern Analysis ", and a method of providing breathing training contents using the same) measures a user's inspiration and expiration, And the respiration training target pattern is compared with the respiration training target pattern set in the breathing training setting module and the respiration training target pattern is reset by reflecting the similarity derived from the respiration training target pattern The present invention proposes a respiration training system through user breathing pattern analysis that can provide optimized breathing training to a user by reflecting sophisticated patterns of breathing in real time and a method of providing breathing training contents using the same.

The present invention specifically relates to a respiration measurement module for measuring respiration of a user; A breathing pattern extracting module for extracting a breathing pattern by analyzing breathing measured by the breathing measuring module; A breathing training setting module for setting and storing a breathing training target pattern corresponding to the user; A breathing pattern analysis module for comparing the breathing training target pattern set through the breathing training setting module and the breathing pattern extracted by the breathing pattern extracting module to derive the similarity; And a feedback module for resetting the breathing training target pattern by reflecting the similarity derived by the breathing pattern analysis module.

However, the present invention has a problem in that it can not perform effective respiration training because it does not consider the user's physical condition, age and occupation by simply setting a target breathing training pattern.

Therefore, in order to solve the above-mentioned problems, there is a need to develop a vocal training system through respiration measurement that provides user-customized vocal training in consideration of the user's gender, age, weight, and occupation as well as target breathing training conditions .

The main object of the present invention is to overcome the problem of the above-described technique, and to provide a vocal training result of a user after inputting vocal data through respiratory pressure measurement and comparing with a target condition inputted by a user.

Another object of the present invention is to generate and provide a question and a sentence for instructing a specific action of a user in order to measure a respiration of a user.

According to an aspect of the present invention, there is provided a speech training system for breathing measurement comprising: a speech measurement device including a breathing meter; A condition input module for inputting user information including a target breathing condition and a user's sex, age, weight, and occupation from a user terminal; and a breathing apparatus for inputting breathing pressure including breath pressure and inspiration pressure as speech data A result analyzing module including a utterance data measuring module including a utterance data input unit and a data comparing and analyzing unit for comparing the target breathing condition with the utterance data and calculating result data; And a main server including an output module.

The vocalization data measurement module may include an item database storing items including directives and songs for instructing the user to perform a specific action for respiration measurement of the user, and providing an item to the user terminal Wherein the speech data input unit receives the speech data as the user performs the action indicated by the item.

Further, the result analysis module may further include a similarity calculation unit for calculating a similarity based on the difference between the target breathing condition and the resultant data with respect to the target breathing condition, and a determination unit for determining whether or not the speech data is re- Wherein the result output module outputs the result data, the degree of similarity, and the re-measurement to the user terminal.

In addition, the vocalization data measuring module may include a sentence database storing a sentence included in an article including an existing article, a story, and a novel, and a sentence sentence designating unit designating any one of the sentence as a presented sentence, And the item providing unit provides the item and the presented sentence to the user terminal.

A vocal training system based on breath measurement according to the present invention comprises:

1) It is possible to provide a custom voice training method according to a user's physical condition,

2) It provides effects of inputting respiratory pressure (that is, speech data) of the user through various presentation sentences and obtaining correct result.

delete

1 is a conceptual diagram showing a schematic configuration of a system of the present invention;
2 is a block diagram showing the overall configuration of the system of the present invention;
3 is a conceptual diagram showing an embodiment in which a report of the present invention is generated;

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The accompanying drawings are not drawn to scale and wherein like reference numerals in the various drawings refer to like elements.

1 is a conceptual diagram showing a schematic configuration of a system of the present invention.

Referring to FIG. 1, a vocal training system based on breath measurement of the present invention compares training data generated by measuring respiration and speech of a user through a vocal measurement device 20 after receiving training conditions from a user, And provides the result to the user. The voice training provides a function of effectively accumulating vocal and speech basics.

Here, the term " user " means an individual who wishes to perform vocal training through the system of the present invention. The term " user " refers to an individual who intends to perform vocal training through the system of the present invention. A person who wants to give a lecture or performance in front of you.

Generally speaking, the term 'vocalization' refers to an act of giving a voice, but since breathing is the most important factor in a song or speech, it is understood as a concept including both breathing and voice generated while voicing.

The system of the present invention basically comprises a user terminal 30, a main server 10, and a utterance measurement device 20. [

The user terminal 30 can be a PC, a smart phone, a tablet PC, or the like as a terminal carried by a user who wants to train a method of vocalizing. At this time, the user terminal 30 is not limited to an application capable of implementing the system of the present invention or a terminal provided with an environment capable of accessing an Internet web page. Through the user terminal 30, the user can use the system of the present invention and perform speech training and practice.

The main server 10 plays a role of implementing a vocal training system based on breath measurement of the present invention and is provided with input from a user based on a physical condition (including height, weight, age, sex, etc.) The target condition and the speech data inputted from the user (which may include the respiratory pressure of the user, and a detailed description thereof will be described with reference to FIG. 2) And provides the user with appropriate vocal training information.

The main server 10 plays a role of managing an Internet web page, an application, or a computer program, which is provided for measuring voice data of a user through a vocal breathing apparatus. Furthermore, the main server 10 stores the existing vocalization method as data, and performs a function of providing an appropriate vocal training method to the user according to the result of analyzing the user's vocal data. At this time, the applications managed by the main server 10 may be the same or different from the applications provided in the user terminal 30.

The vocal measurement apparatus 20 measures the utterance of a user, and specifically refers to a device that numerically measures the utterance of a user by measuring the respiration of the user. The vocal measurement apparatus 20 may be basically a respiration measuring instrument such as an inspiration meter, and can measure a voice of a user when a microphone or a chest band is included, thereby contributing to more accurately grasp a user's utterance ability.

The system according to the present invention configured as described above performs a function of contributing to perform vocal training repeatedly until a user reaches a target value by comparing and analyzing a target vocal ability and an actual user's vocal ability. The system of the present invention allows a user to efficiently learn a method of controlling breathing or making a voice.

In the description of the present invention, a server includes software for implementing a communication unit, a RAM, a ROM, a CPU, a hardware and a program, and a module of the system of the present invention to be described later is a combination of such a server or a server, (Not shown).

The overall structure for implementing the system of the present invention to perform these functions will be described with reference to FIG. 2 to be described later.

2 is a block diagram showing the overall configuration of the main server of the present invention.

2, the main server 10 of the present invention basically comprises a condition input module 100, a voicing data measurement module 200, a result analysis module 300, a training information providing module 400, (600).

The condition input module 100 receives the user information including the target breathing conditions and the sex, age, weight, and occupation of the user from the user terminal 30, and includes user information including the physical information of the user and the user, And provides a function of receiving target respiration conditions to be achieved through the system of the present invention.

Here, the target breathing condition refers to the breathing condition that the user wishes to achieve, and in particular, the breathing pressure that the user wishes to achieve through the system of the present invention (the breathing pressure is the pressure generated during breathing, And the intake pressure, which is a pressure to be applied to the engine.

The voicing data measurement module 200 receives voicing data from the voices of the user measured through the voicing measurement device 20 and receives the voicing data. The voicing data measurement module 200 may specifically include a voicing data input unit 210.

The utterance data input unit 210 receives the breathing pressure including the atmospheric pressure and the inspiration pressure from the utterance measurement apparatus 20 as speech data. For example, when the user performs breathing while wearing the breathing apparatus And performs a function of receiving the measured atmospheric pressure and inspiratory pressure as user's utterance data.

The result analysis module 300 may include a data comparison and analysis unit 310 for analyzing the result of speech training performed by the user by calculating result data obtained by comparing target breathing conditions and speech data.

The data comparison and analysis unit 310 compares the target breathing condition with the utterance data to calculate result data. Specifically, the data comparison and analysis unit 310 can generate the result data through the difference between the target breathing condition and the utterance data.

For example, each received target respiratory condition input by the user (i.e., a target respiratory pressure) The heading pressure 40mmH ₂ O and the intake pressure 60mmH ₂ O No. of measured user via the respiratory measuring device when the atmospheric pressure and the intake pressure 38mmH ₂ O, 59 mm H ₂ O, the resulting data for the atmospheric pressure can be ₂ mm H ₂ O, and the resulting data for the inspiratory pressure can be ₁ mm H ₂ O.

The training information providing module 400 provides the training information to the user through the user terminal 30 according to the result data. The training information providing module 400 generates training information such as a Beltangato method using a nasal cavity resonance, a side breath method, (The explanation of the side breathing method will be described later.)

It is needless to say that the system administrator of the present invention can determine which training information should be provided to the user according to the result of the result data.

The training module (600) has a function of activating the respiratory tightening muscles by repeating the 'side' action once or several times with the fingers on the multi-lips and the function of Hi, Hiya, Cha, Tcha, Ste, Sta, fit, Whouit, and a breathing tightening motion with one hand in front of mouth and one hand in front of waist line 10cm to 30cm in front of each other, And performs a function of indicating at least one of the functions to be performed.

Through the training module 600 performing this function, a user using the system of the present invention can efficiently learn a method of maintaining a longer breathing, a larger volume and a higher pressure as compared with the conventional vocal training.

The voicing data measurement module 200 may provide a question for instructing a specific action to the user in order to receive the user's utterance data. For this purpose, the voicing data measurement module 200 may include an item database 220 and an item provider 230 have.

The item database 220 stores a question including an instruction and a song to indicate a specific action to the user in order to measure respiration of the user. The question provider 230 stores at least one of the items in question And provides it to the terminal 30.

Accordingly, the utterance data input unit 210 performs a function of receiving utterance data generated as the user performs the action indicated by the item.

In addition, the result analysis module 300 can determine whether the user should perform speech training again by calculating the similarity by comparing the user's speech data with the target breathing condition. For this, the result analysis module 300 may include a similarity calculation unit 320 and a remeasurement determination unit 330.

The degree of similarity calculation unit 320 calculates the degree of similarity based on the difference between the target breathing condition and the resultant data in comparison with the target breathing condition. It can be understood that the degree of similarity is calculated as the difference between the target breathing condition and the result data is small.

The re-measurement determination unit 330 determines whether or not the voice data is re-measured according to the degree of similarity. If the degree of similarity is high, the re-measurement determination unit 330 determines that the user is performing voice training properly, And if the degree of similarity is low, it is determined that the user does not have a vocalization ability as much as the target value, and the vocalization data is judged to be re-measured once more.

Accordingly, the training information providing module 400 outputs the similarity and re-measurement to the user terminal 30, thereby allowing the user to perform breathing training based on the training information provided.

Furthermore, the utterance data measurement module 200 may include a sentence database 240 and a sentence designation unit 250 for receiving a lot of user's utterance data and calculating an accurate result by providing various presenting sentences to the user .

The sentence database 240 stores sentences included in articles including existing articles, fairy tale, and novels, and is a place where the user can store various sentences that can be pronounced as data.

The presentation sentence designating unit 250 designates any one of the sentences as a presented sentence, and performs a function of arbitrarily designating any one of sentences stored in the sentence database 240 as a presented sentence. However, in order to select a sentence in which the user can produce a more diverse pronunciation, the presentation sentence designation unit 250 specifically includes a candidate sentence selection part 251, a sentence designation part 252, a sentence analysis part 253, A sequence generating part 254, a sequence matrix generating part 255, a coordinate generating part 256, a pronunciation similarity calculating part 257, and a presentation sentence selecting part 258.

First, the candidate sentence selection part 251 provides a function of selecting a plurality of sentences including a consonant for producing a plosive sound as candidate sentences. The sentence designation part 252 designates one of the candidate sentences as a reference sentence, The sentence analyzing part 253 separates the reference sentence and the comparison sentence into morpheme units, and then the reference sentence and the comparison sentence are included in the reference sentence and the comparison sentence And recognizes the respective nouns.

The character string generation part 254 generates a character string by segmenting the reference sentence and a noun included in the comparison sentence into syllables, segmenting the syllable into phonemes, and then generating the phonemes. For example, in the case of ' It generates a string such as a, a, b, a, c, a.

The sequence matrix generation part 255 generates a sequence matrix corresponding to each of the strings through a sequence sorting algorithm. Here, the sequence sort algorithm refers to a sequence of genes or proteins using a dynamic program-based computer string sorting algorithm . That is, the sequence matrix generation part 255 performs a function of generating a sequence matrix from a string by focusing on the arrangement of the nucleotide sequence of the gene or protein.

The coordinate generation part 256 generates coordinates of a noun corresponding to a character string based on the components of the sequence matrix, which generates coordinates of a noun based on a multidimensional quantization algorithm. In this case, the multidimensional quantization algorithm calculates numerical values for each phoneme. For example, when the value derived from the language laboratory of Hankuk University of Foreign Studies is used, the coordinates of 'a' become (-1.2, 1.1, 1.0) The coordinates become (-1.4, -0.3, -0.6), and the coordinates of 'a' become (-1.2, 1.1, 1.0, -1.4, -0.3, -0.6).

The pronunciation similarity calculating part 257 calculates pronunciation similarity between the reference sentence and the comparison sentence based on a distance between coordinates of a noun, and the pronunciation similarity is calculated through the following equation (1).

Equation 1

는 기준 문장 A 및 비교 문장 B의 발음 유사도,

은 기준 문장 A에 포함된 명사의 좌표,

은 비교 문장 B에 포함된 명사의 좌표, n은 상기 명사의 좌표 성분의 개수)(here,

Is the pronunciation similarity of the reference sentence A and the comparison sentence B,

The coordinates of a noun included in the reference sentence A,

Is the coordinate of the noun included in the comparison sentence B, and n is the number of coordinate components of the noun)

Equation 1 is a formula for calculating the pronunciation similarity of a noun included in a sentence based on the coordinates of a noun, and provides a function of calculating the pronunciation similarity between the reference sentence and the comparison sentence from the distance between the coordinates. In other words, Equation (1) performs a function of numerically indicating pronunciation similarity, which is an index for determining whether to select a presentation sentence according to the calculated degree of pronunciation similarity.

In order to apply Equation (1), the number of nouns included in the reference sentence and the comparison sentence should be the same.

The presentation sentence selection part 258 can provide the user with various presentation sentences by performing the function of selecting the comparison sentence having the pronunciation similarity lower than the predetermined reference value as the presented sentence.

When the presented sentence is selected, the question provider 230 provides the question and the presented sentence to the user terminal 30, and the user's utterance data can be measured through the presented sentence.

3 is a conceptual diagram showing an embodiment in which a report of the present invention is generated.

3, the condition input module 100 receives the target score from the user terminal 30, and the utterance data may include a breathing time from the time when the user starts to take the air to the time when the user finishes the breathing, The main server 10 includes a vocalization data analysis module 500 including a vocalization data and a result score calculation unit 510 for calculating a score obtained by digitizing the breathing ability of the user based on the user information, Can be quantified.

The result score is calculated by the following equation (2).

(2)

는 결과 점수,

는 호기압(cmH₂O),

는 흡기압(cmH₂O),

는 상기 호흡 시간,

는 성별 가중치,

는 상기 유저의 체중(kg),

는 직업 가중치를 의미한다.here,

The result score,

The call pressure (cmH ₂ O),

(CmH ₂ O),

The respiration time,

Gender weight,

(Kg) of the user,

Means a work weight.

Equation 2 is a formula for calculating a result score based on the atmospheric pressure, the inspiratory pressure, the breathing time, and the user information. It is not merely calculating the result score based on the speech pressure and the inspiratory pressure, By reflecting the profession more accurately, it is possible to calculate a more accurate result score (the user's respiratory ability differs depending on whether the user is male or female, how much weight is going, whether the user is a singer or other profession, The result score by 2 can be calculated differently.)

Accordingly, the data comparison and analysis unit 310 can calculate the result data by comparing the target score and the result score, and output the result data to the user terminal 30. Therefore, the user can be provided with more accurate training method.

Further, the utterance data input unit 210 receives the breath amount calculated through the utterance measurement apparatus 20 during the breathing time, and the utterance data analysis module 500 calculates the correction score By including the score calculating unit 520, it is possible to provide the user with a more accurate speech training result.

At this time, the correction score can be calculated by the following equation (3).

(3)

는 보정 점수,

는 호기압(cmH₂O),

는 흡기압(cmH₂O),

는 상기 호흡 시간,

는 성별 가중치,

는 상기 유저의 체중(kg),

는 직업 가중치,

는 호흡량(L)을 의미한다.here,

A correction score,

The call pressure (cmH ₂ O),

(CmH ₂ O),

The respiration time,

Gender weight,

(Kg) of the user,

The work weight,

Means the volume of breath (L).

Equation (3) is a formula for calculating the correction score by further reflecting the breathing amount in the above Equation (2). The breathing ability of the user is more accurately reflected by reflecting not only the breathing pressure but also the breathing amount which is called the user's lung capacity.

Accordingly, the data comparison and analysis unit 310 compares the target score and the correction score to calculate the resultant data.

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

delete

As described above, the construction and operation of the vocal training system through breath measurement according to the present invention are described in the above description and drawings, but the present invention is not limited to the above description and drawings, And various changes and modifications may be made without departing from the technical spirit of the present invention.

10: main server 20: vocal measurement device
30: user terminal 100: condition input module
200: vocalizing data measuring module 210: vocalizing data inputting part
220: item database 230:
240: sentence database 250:
251: Candidate sentence selection part 252: Sentence designation part
253: Sentence Analysis Part 254: String Generation Part
255: Create sequence matrix Part 256: Create coordinate sequence
257: pronunciation similarity degree calculation part 258: present sentence selection part
300: Result analysis module 310: Data comparison analysis part
320: degree-of-similarity calculation unit 330: re-
400: training information providing module 500: vocalizing data analysis module
510: Result score calculating unit 520: Correction score calculating unit
600: Training Module

Claims (15)

As a vocal training system through breath measurement,
A vocal measurement device including a breath meter;
A condition input module for inputting user information including a target breathing condition and a user's gender, age, weight, and occupation from a user terminal and a target score; and a respiratory-pressure- Wherein the speech data includes a speech data input unit for inputting a breath volume calculated through the speech measurement apparatus during the breathing time in a state including a breathing time from the time when the user starts the inspiration to the time when the breathing is completed, And a data comparison / analysis module for comparing the target breathing condition with the utterance data to calculate result data, and a data analysis module for analyzing the target breathing condition and the utterance data to provide training information to the user through the user terminal according to the result data. An information providing module, and a storage unit for storing the utterance data and the user information A resultant score calculating unit for calculating a score of the breathing ability of the user, and a correction score calculating unit for calculating a correction score by further reflecting the breathing amount on the result score. , ≪ / RTI &
Wherein the data comparison / analysis unit compares the target score with the result score, and compares the target score and the correction score to calculate result data. The method according to claim 1,
The system comprises:
Hia, Hiya, Cha, Tcha, Ste, Sta, fit, and Whouit sounds combined by analyzing the characteristics of consonants and vowels. And a function of performing a motion in which the one hand is in front of the mouth and the other hand is in front of the waist line in front of the waist line and 10 to 30 cm in front of the waist line when the respiratory tightening muscle is activated, Wherein said training module comprises at least one of a training module and a training module. The method according to claim 1,
The voicing data measurement module includes:
An item database storing items including directives and songs for instructing the user to perform a specific action for respiration measurement of the user,
And an item provision unit for providing the item to the user terminal,
The speech data input unit may include:
Wherein the speech data is input as the user performs the action indicated in the item. The method according to claim 1,
The result analysis module,
A similarity degree calculating unit for calculating a degree of similarity based on a difference between the target breathing condition and the resultant data with respect to the target breathing condition;
And a remeasurement judgment unit for judging whether the voice data is re-measured according to the degree of the similarity,
The training information providing module includes:
And outputting the result data, the degree of similarity, and the re-measurement to the user terminal. The method of claim 3,
The voicing data measurement module includes:
A sentence database storing a sentence included in an article including an existing article, a story, and a novel,
And a presentation sentence designation unit designating any one of the sentences as a presentation sentence,
The item provider,
And providing the item and the presented sentence to the user terminal. 6. The method of claim 5,
The presentation sentence designation unit,
A candidate sentence selection part for selecting a plurality of sentences including a consonant to generate a plosive sound as a candidate sentence among the sentences,
A sentence designation part designating any one of the candidate sentences as a reference sentence and a candidate sentence excluding the reference sentence as a comparison sentence,
A sentence analysis part for separating the reference sentence and the comparison sentence into morpheme units and recognizing the nouns included in the reference sentence and the comparison sentence,
A string generating part for segmenting the nouns included in the reference sentence and the comparison sentence into syllables, segmenting the syllable into phonemes, and generating a string listing the phonemes,
A sequence matrix generating part for generating a sequence matrix corresponding to each of the strings through a sequence alignment algorithm;
A coordinate generating part for generating coordinates of the noun corresponding to the character string based on the components of the sequence matrix;
A pronunciation similarity degree calculating section for calculating a pronunciation similarity degree between the reference sentence and the comparison sentence based on a distance between coordinates of the noun;
And a presentation sentence selection part for selecting the comparison sentence having pronunciation similarity less than a preset reference value as a presented sentence. The method according to claim 6,
The pronunciation similarity degree may be,
Wherein the speech training is performed through the following equation (1).
Equation 1

(here,

Is the pronunciation similarity of the reference sentence A and the comparison sentence B,

The coordinates of a noun included in the reference sentence A,

Is the coordinate of the noun included in the comparison sentence B, and n is the number of coordinate components of the noun) delete The method according to claim 1,
The result score,
(2) is calculated by the following equation (2).
(2)

(here,

The result score,

The call pressure (cmH ₂ O),

(CmH ₂ O),

The respiration time,

Gender weight,

(Kg) of the user,

Job weighting) delete The method according to claim 1,
The above-
Wherein the speech training is performed through the following equation (3).
(3)

(here,

A correction score,

The call pressure (cmH ₂ O),

(CmH ₂ O),

The respiration time,

Gender weight,

(Kg) of the user,

The work weight,

(L))
delete delete delete delete

Images