KR102093369B1 - Control method, device and program of hearing aid system for optimal amplification for extended threshold level - Google Patents

Abstract

Provided are a method, an apparatus, and a program for controlling a hearing aid system for optimal amplification. The method for controlling a hearing aid system is performed by a server and a hearing aid, and comprises the steps of: allowing a server to obtain a plurality of frequency bands on the basis of Korean voice data; allowing the server to obtain a first frequency gain value and a first maximum output value for each of the plurality of frequency bands; allowing a hearing aid to perform primary correction thereof on the basis of the first frequency gain value and the first maximum output value for each of the obtained frequency bands; allowing the server to obtain data on a hearing condition of a hearing aid user and voice data of the hearing aid user; and allowing the hearing aid to perform secondary correction thereof having been primarily corrected on the basis of the data on the hearing condition of the hearing aid user and the voice data of the hearing aid user. According to the present invention, a hearing aid can be corrected in accordance with a suitable formula of the hearing aid optimized for Koreans.

Description

Control method, device and program of hearing aid system for optimal amplification for extended threshold level}

The present invention relates to a control method, apparatus and program for a hearing aid system for optimal amplification for an extended threshold level.

Hearing aids were developed to help hearing impaired people who are deaf or deaf. This hearing aid is a device that amplifies the voice data coming into the microphone and outputs it to a receiver so that a hearing-impaired person can hear the inaudible sound.

The purpose of this is to improve the speech recognition ability by obtaining the optimum gain according to the hearing and input sound pressure from each channel divided according to the frequency band, and then amplifying the speech data according to the optimum gain. . At this time, it is important to improve the performance of the hearing aid by finding a suitable formula for obtaining each frequency band and maximum gain.

However, it is a reality that most suitable formulas used in current hearing aids are developed based on the characteristics and cognitive abilities of English sounds. Accordingly, the necessity for a hearing aid fit formula based on the Korean anatomy and the hearing anatomical characteristics of Koreans and the cognitive abilities of Korean speech is increasing, and furthermore, a method of correcting a hearing aid optimized for each hearing aid user is becoming important.

Registered Patent Publication No. 10-0956167, 2010.04.27

The problem to be solved by the present invention is to provide a control method, apparatus and program of a hearing aid system for optimal amplification for an extended threshold level.

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

A control method of a hearing aid system according to an aspect of the present invention for solving the above-mentioned problems is a control method of a hearing aid system performed by a server and a hearing aid, wherein the server, based on the Korean voice data, a plurality of frequency bands Obtaining; Obtaining, by the server, a first frequency gain value and a first maximum output value for each of the plurality of frequency bands; The hearing aid performing a primary correction of the hearing aid based on a first frequency gain value and a first maximum output value for each of the obtained plurality of frequency bands; Acquiring data about the hearing state of the hearing aid user and voice data of the hearing aid user by the server; And the hearing aid performing a second correction of the first corrected hearing aid based on data on the hearing state of the hearing aid user and voice data of the hearing aid user. It includes.

At this time, obtaining a first frequency gain value and a first maximum output value for each of the plurality of frequency bands includes: classifying the Korean speech data into eight frequency bands; Obtaining a positive response rate for each of the classified eight frequency bands; Obtaining respective weights for each frequency band based on the obtained positive response rate, and determining the first frequency gain value and the first maximum output value based on the weights; It may include.

At this time, obtaining a first frequency gain value and a first maximum output value for each of the plurality of frequency bands may include: determining a loudness of the conversational sound of the Korean speech data; When the conversation sound volume is within a preset range, a first parameter is determined based on the compression threshold, the compression ratio, the extension threshold, the extension ratio, and the weight The method may further include determining the first frequency gain value and the first maximum output value based on the determined first parameter.

At this time, the step of acquiring a plurality of frequency bands based on the Korean voice data, the text included in the Korean standard sentence list for adults (KS-SL-A) is generated inside the soundproof booth, but the speaker of the Korean voice data is soundproofed Located in the center of the booth, the microphone is located in front of the speaker 1M, obtaining the first voice data through the microphone; Learning the artificial intelligence model to extract feature values for Korean speech data by inputting the first speech data into the artificial intelligence model; Inputting second voice data including noise into the learned artificial intelligence model, extracting feature values for the second voice data, and correcting the second voice data based on the extracted feature values; And obtaining a plurality of frequency bands based on the Korean voice data including the first voice data and the corrected second voice data.

At this time, the step of acquiring data about the hearing state of the hearing aid user and the audio data of the hearing aid user is based on the data on the hearing state of the hearing aid user, the frequency gain value for each of a plurality of frequency bands and the Correcting the maximum output value to obtain a second frequency gain value and a second maximum output value; And extracting a feature value for correcting the waveform of the sound source sensed by the hearing aid based on the voice data of the hearing aid user, wherein the second correction step comprises: the second frequency gain value, the second And secondly correcting the primary corrected hearing aid based on a maximum output value and a feature value for correcting the waveform of the sound source.

At this time, the control method, the hearing aid, transmitting the voice data input from the hearing aid to the electronic device; Transmitting, by the electronic device, the received voice data to the server; If the server satisfies a preset condition, acquiring a third frequency gain value and a third maximum output value for the third correction of the hearing aid based on the received voice data; And thirdly correcting the second-corrected hearing aid based on the third frequency gain value and the third maximum output value. It may include.

At this time, the control method may include: collecting background information including information about background noise in the living space of the hearing aid user, information about the speech speed of the hearing aid user, and information about the average speech speed of the general public; Determining a background noise and an utterance rate to be included in the hearing training program based on the collected background information; Generating a hearing training program based on the determined background noise, speech recognition level, and speech rate; Receiving result data for the hearing training program and hearing aid wearing from the hearing aid user; And fourthly correcting the hearing aid based on the result data. Including, and the fourth correction step, obtaining a frequency interval having a wrong answer rate above a predetermined ratio among the result data; Comprising the step of correcting the frequency gain value of the obtained frequency section to the third frequency gain value or more, The generated hearing training program includes the visual cues, phonemes, words and sentences according to the background information of the hearing aid user Difficulty, sentence length, environmental sound and whether to include music can be determined.

At this time, in the step of generating the hearing training program, a plurality of first data including background information of a plurality of hearing aid users, answer information on a questionnaire question, and detailed information obtained from the answer information are stored for each of the plurality of users To do; Storing a plurality of second data including the hearing aid training program of the plurality of hearing aid users and result data on wearing the hearing aid for each of the plurality of hearing aid users; Clustering the plurality of first data to obtain a plurality of first clusters; Clustering the plurality of second data to obtain a plurality of second clusters; Obtaining background information and response information for a new first hearing aid user; Determining a first-first cluster related to background information and answer information about the acquired first hearing aid user among the plurality of first clusters; Determining a 2-1 cluster existing within a predetermined distance from the 1-1 cluster among the plurality of second clusters; And generating a hearing training program for the first hearing aid user based on the data included in the 2-1 cluster. It may include.

Other specific matters of the present invention are included in the detailed description and drawings.

According to various embodiments of the present invention described above, it is possible to correct the hearing aid according to a suitable formula of a hearing aid optimized for a Korean.

Furthermore, a method of calibrating a hearing aid suitable for each individual hearing aid user may be provided.

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

1 is a diagram illustrating a system according to an embodiment of the present invention.
2 is a flowchart illustrating a hearing aid correction method according to an embodiment of the present invention.
3 is a flowchart illustrating a method of classifying Korean voice data into a plurality of frequency bands according to an embodiment of the present invention.
4 is a flowchart for explaining a method of acquiring parameters according to the size of a conversation sound according to an embodiment of the present invention.
5 is a flowchart illustrating a method of acquiring a frequency band using an artificial intelligence model according to an embodiment of the present invention.
6 is a flowchart for explaining a method of secondaryly correcting a hearing aid according to an embodiment of the present invention.
7 is a flowchart for explaining a method of third-order correction of a hearing aid according to an embodiment of the present invention.
8 is a flowchart illustrating a method of fourth-order correction of a hearing aid according to an embodiment of the present invention.
9 is a flowchart illustrating a method of obtaining usage data for a hearing aid according to an embodiment of the present invention.
10 to 14 are graphs for explaining experimental data according to various embodiments of the present invention.
15 is a configuration diagram of an apparatus according to an embodiment of the present invention.

Advantages and features of the present invention, and methods for achieving them will be clarified with reference to embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only the present embodiments allow the disclosure of the present invention to be complete, and are common in the technical field to which the present invention pertains. It is provided to fully inform the skilled person of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terminology used herein is for describing the embodiments and is not intended to limit the present invention. In the present specification, the singular form also includes the plural form unless otherwise specified in the phrase. As used herein, “comprises” and / or “comprising” does not exclude the presence or addition of one or more other components other than the components mentioned. Throughout the specification, the same reference numerals refer to the same components, and “and / or” includes each and every combination of one or more of the components mentioned. Although "first", "second", etc. are used to describe various components, it goes without saying that these components are not limited by these terms. These terms are only used to distinguish one component from another component. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used as meanings commonly understood by those skilled in the art to which the present invention pertains. In addition, terms defined in the commonly used dictionary are not ideally or excessively interpreted unless explicitly defined.

The term "part" or "module" as used in the specification refers to a hardware component such as software, FPGA, or ASIC, and "part" or "module" performs certain roles. However, "part" or "module" is not meant to be limited to software or hardware. The "unit" or "module" may be configured to be in an addressable storage medium or may be configured to reproduce one or more processors. Thus, as an example, "part" or "module" means components, processes, functions, attributes, such as software components, object-oriented software components, class components and task components. Includes procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, database, data structures, tables, arrays and variables. The functionality provided within components and "parts" or "modules" can be combined into a smaller number of components and "parts" or "modules" or into additional components and "parts" or "modules" Can be further separated.

The spatially relative terms “below”, “beneath”, “lower”, “above”, “upper”, etc., are as shown in the figure. It can be used to easily describe a correlation between a component and other components. Spatially relative terms should be understood as terms including different directions of components in use or operation in addition to the directions shown in the drawings. For example, if a component shown in the drawing is turned over, a component described as "below" or "beneath" of another component will be placed "above" the other component. Can be. Accordingly, the exemplary term “below” can include both the directions below and above. The component can also be oriented in other directions, so that spatially relative terms can be interpreted according to the orientation.

In the present specification, the computer means all kinds of hardware devices including at least one processor, and may be understood as a meaning encompassing software configurations operating in the corresponding hardware device according to embodiments. For example, a computer may be understood as meaning including, but not limited to, a smart phone, a tablet PC, a desktop, a laptop, and a user client and application running on each device.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Each step described in this specification is described as being performed by a computer, but the subject of each step is not limited thereto, and according to an embodiment, at least some of the steps may be performed in different devices.

Meanwhile, in the present specification, hearing means a process of detecting, discriminating, confirming, and recognizing sound. Hearing impaired means a person with hearing impairment in one or both ears. Hearing aid refers to an electroacoustic wear device designed for processing sound signals for the purpose of compensating for hearing loss. A hearing aid system means an individual customized system composed of related components such as one or both hearing aids, earplugs, remote control devices, and connectors. Hearing aid professional refers to a person who has hearing ability to provide hearing aid fitness management services including hearing aid selection, adjustment, verification, and hearing training for hearing impaired persons. . A client or a hearing aid user means a hearing aid expert or a person receiving a hearing aid conformity management service provided by a hearing aid conformance management system. Hearing aid fitting management is a comprehensive management system that includes an evaluation module, a fit module, and a follow-up module. Consultation, hearing test, hearing aid test wear, hearing aid selection, ear impression collection, hearing aid adjustment, verification, customer training It refers to a systematic process that helps elevate hearing aids wearing hearing aids, improve speech perception, and continuously use and manage hearing aids by performing 11 components of hearing training, result measurement, and comprehensive report preparation. Auditory rehabilitation refers to a systematic and comprehensive process of improving hearing skills and communication skills through hearing aid conditioning, counseling, education and various training. The assessment module refers to the process of hearing aid conformance management that includes the five components of all stages of hearing aid fitness: counseling, hearing testing, hearing aid trial wearing, hearing aid selection and ear impression collection and procedures between components. The fitting module refers to a module for performing the process of hearing aid fit management including the three components of the hearing aid fit phase, including hearing aid adjustment, verification, and customer training and procedures between components. The follow-up module means a module for performing the process of hearing aid fitness management including the three components of hearing aid fitness, hearing training, outcome measurement and overall report preparation and procedures between components. . Counseling refers to the process of providing and supporting comprehensive information about a customer's or their family's hearing loss status, difficulties caused by hearing loss, hearing aid fit, and hearing training. Hearing evaluation refers to the process of grasping the hearing loss type and degree and the degree of communication by measuring the hearing threshold and speech recognition ability before selecting a hearing aid. Hearing aid trial refers to the process of actually experiencing a hearing aid adjusted to the hearing condition of the customer in all stages of hearing aid selection in order to select a hearing aid suitable for the hearing loss condition of the customer. Hearing aid selection refers to the process of determining the type of hearing aid suitable for a customer based on hearing aid test results and hearing background information. Ear impression refers to the process of replicating the shape of the ear canal and the instep to create the shape of an earplug or hearing aid. Hearing aid adjustment refers to the hearing aid's external characteristics and wearing status for physical comfort when wearing it, and changes it if necessary. Also, it is based on hearing evaluation results and prescriptions in consideration of speech recognition ability improvement and acoustic comfort. This refers to the process of optimizing the electroacoustic characteristics of the hearing aid. Verification for hearing aid adjustment refers to the process of evaluating hearing aid adjustment results through coupler measurement, real ear measurement, or sound field measurement and questionnaire response of electroacoustic and psychoacoustic aspects. Client education refers to the process of providing information related to the use of hearing aids and information necessary to establish realistic expectations of customers for hearing aids, such as how to operate hearing aids, how to inspect hearing aids, and how to solve problems that may occur when using hearing aids. Auditory training (auditory training) refers to a learning process that includes intensive and repetitive listening exercises and communication strategies over a period of time to improve the communication skills of deaf people. Outcome measurements are the process of multi-dimensional evaluation of hearing aid users 'wearing results in terms of acoustic and psychosocial aspects. Generally speaking, hearing aid users' perception of speech, objective benefits, quality of sound, and improvement in hearing effort It includes both objective and subjective assessments such as frequency of hearing aid use and satisfaction. Comprehensive report or result report means a report that reflects key information arising from the overall process of hearing aid compliance management. An ear mold refers to a mechanically-acoustic connection device between a hearing aid and an external auditory meatus individually manufactured or selected. The client profile (client profile) refers to comprehensive information related to hearing, including not only the auditory background information of the customer, but also hearing function, social situation, activity opportunities, hearing aid needs, and expectations. A fitting system refers to a set of equipment such as computers and suitable software used to control a hearing aid. Pre-setting of hearing aids refers to hearing aid adjustments based on conformity formulas and relevant hearing information. Fine-tuning refers to the tuning of a hearing aid system that best meets the needs and preferences of individual hearing loss.

1 is a view for explaining a system according to an embodiment of the present invention.

The system according to the present invention may consist of a server 10, a hearing aid 20 and an electronic device 30.

The server 10 is configured to receive and analyze various data from the electronic device 30. The server 10 may collect various data for calibrating the hearing aid, and obtain parameters for hearing aids based on the collected data.

The electronic device 30 is configured to transmit various data related to the hearing aid user's use of the hearing aid to the server 10. The electronic device 30 may be implemented as a smart phone and transmit various data about the hearing aid user to the server 10, but is not limited thereto. For example, the electronic device 30 is implemented as a hearing aid, and it is needless to say that the data of the hearing aid user can be transmitted to the server 10 according to user interaction.

Furthermore, according to various embodiments of the present invention, the electronic device 30 includes a tablet personal computer (PC), a mobile phone, a video phone, an e-book reader, and a desktop PC. , Laptop PC, netbook computer, workstation, server, personal digital assistant (PDA), portable multimedia player (PMP), MP3 player, mobile medical device, camera, or wearable device ( wearable device). According to various embodiments, the wearable device may be an accessory type (for example, a watch, ring, bracelet, anklet, necklace, glasses, contact lens, or head-mounted device (HMD)), a fabric, or a garment type ( For example, it may include at least one of an electronic garment, a body attachment type (for example, a skin pad or a tattoo), or a living body implantation type (for example, an implantable circuit).

As another embodiment, the electronic device 30 may be a home appliance. Household appliances include, for example, televisions, DVD players (Digital Video Disk players), audio, refrigerators, air conditioners, vacuum cleaners, ovens, microwave ovens, washing machines, air cleaners, set-top boxes, and home automation. Home automation control panel, security control panel, TV box (eg Samsung HomeSync ™, Apple TV ™, or Google TV ™), game console (eg Xbox ™, PlayStation ™), electronics It may include at least one of a dictionary, an electronic key, a camcorder, or an electronic picture frame.

In another embodiment, the electronic device 30 includes various medical devices (eg, various portable medical measurement devices (such as a blood glucose meter, heart rate monitor, blood pressure meter, or body temperature meter), magnetic resonance angiography (MRA), and magnetic resonance (MRI). imaging), CT (computed tomography), imager, or ultrasound, etc., Internet of things (e.g. light bulbs, various sensors, electricity or gas meters, sprinkler devices, fire alarms, thermostats, street lights) , Toaster (toaster), exercise equipment, hot water tank, heater, boiler, etc.).

2 is a flowchart illustrating a hearing aid correction method according to an embodiment of the present invention.

In step S100, the server 10 may acquire a plurality of frequency bands based on Korean voice data.

In step S200, the server 10 may obtain a first frequency gain value and a first maximum output value for each of the plurality of frequency bands.

In step S300, the hearing aid 20 may first correct the hearing aid 20 based on the first frequency gain value and the first maximum output value for each of the obtained plurality of frequency bands.

In one embodiment, the first maximum output value and the various maximum output values described below may be set such that the sound amplified based on the maximum output value is not greater than the discomfort level, and the normal-sized dialogue sound is within the comfort level range. Therefore, when the sound amplified by the frequency gain value is greater than or equal to the maximum output value, the server 10 may change the frequency gain value.

In step S400, the server 10 may acquire data about the hearing state of the hearing aid user and voice data of the hearing aid user.

In step S500, the hearing aid 20 may perform a second correction on the first corrected hearing aid 20 based on the acquired hearing aid user's hearing state data and the hearing aid user's voice data.

That is, the server 10 first corrects the hearing aid 20 based on the Korean voice data to correct the hearing aid 20 to be suitable for Koreans, and the hearing aid to be suitable for the hearing aid user based on various data about the hearing aid user ( 20) can be calibrated second.

3 is a flowchart illustrating a method of classifying Korean voice data into a plurality of frequency bands according to an embodiment of the present invention.

In step S210, the server 10 may classify Korean voice data into eight frequency bands.

At this time, the frequency band can be determined experimentally. For example, the frequency unit may be classified based on Korean speech data recorded by a normal hearing person aged 19 to 26 years in a soundproof booth. In one embodiment, the frequency band is 224 Hz or less, 224 Hz to 447 Hz, 447 Hz to 891 Hz, 891 Hz to 1413 Hz, 1413 Hz to 2239 Hz, 2239 Hz to 3548 Hz, 3548 Hz to 5603 Hz, 5623 Hz to 11000 Hz, 11000 Hz or more, based on 9 frequency sections. Acquiring the number of center frequencies, and may be a frequency band for each of the obtained 8 center frequencies. In this specification, the eight center frequencies may be 250 Hz, 500 Hz, 800 Hz, 1000 Hz, 1600 Hz, 2500 Hz, 4000 Hz, 6000 Hz, and the eight frequency bands may be frequency bands within a preset range from the center frequency.

In one embodiment, the eight frequency bands may be 0 Hz to 400 Hz, 350 Hz to 750 Hz, 700 Hz to 950 HZ, 900 HZ to 1500 Hz, 1400 HZ to 2300 Hz, 2200 Hz to 3600 Hz, 3500 Hz to 5500 HZ, 5000 HZ to 11000 Hz.

In step S220, the server 10 may obtain a positive response rate for each of the 8 frequency bands classified.

In one embodiment, the Korean voice data sentences are high-pass or low-pass cutoff based on the eight frequency bands described above, and these eight frequency bands may be composed of ten sentences each. In addition, Korean voice data sentences were recorded by confirming the correct response to 40 key words out of 10 sentences. The sentence is presented in the size of 45 dB, and the noise level is different from the sentence to obtain the positive response rate according to the frequency band. Can be.

Specifically, FIG. 10 is a graph showing the positive response rate for the low-pass filter, FIG. 11 is a graph showing the positive response rate for the high-pass filter, and the graphs shown in FIGS. 10 and 11 are the same as the contents of the following table.

Low-pass High-pass Average SNR
Hz 0 +5 0 +5 0 +5 Less than 224 0.00 0.00 2.05 0.23 1.02 0.11 224-447 4.00 17.50 3.86 1.14 3.93 9.32 447 ~ 891 30.50 54.33 9.77 2.05 20.14 28.19 891 ~ 1413 23.83 16.83 13.18 2.95 18.51 9.89 1413 ~ 2239 16.50 6.17 21.14 15.45 18.82 10.81 2239 ~ 3548 13.00 3.33 37.27 54.09 25.14 28.71 3548 ~ 5623 5.50 1.17 11.36 21.82 8.43 11.49 5623 ~ 11000 2.00 -0.50 1.36 2.27 1.68 0.89 11000 or more 4.67 1.17 0.00 0.00 2.33 0.58 Sum 100.00 100.00 100.00 100.00 100.00 100.00

In step S230, the server 10 may obtain respective weights for each frequency band based on the obtained positive response rate. In one embodiment, the server 10, based on Table 1 above, 2239 Hz ~ Frequency band of 3548Hz, frequency band of 447Hz ~ 891Hz, frequency band of 3548Hz ~ 5623Hz, frequency band of 1413Hz ~ 2239Hz, frequency band of 891Hz ~ 1413Hz, frequency band of 224Hz ~ 447Hz, frequency band of 5623Hz ~ 11000Hz, frequency above 11000Hz The magnitude of the weights can be determined in the order of bands and frequency bands below 224 Hz. In one embodiment, the weight may be determined based on a band-importance function (BIF). The band-importance function (BIF) is a value for importance of different frequencies for speech intelligibility and can be determined as a value between 0 and 1. At this time, the BIF can be determined based on the SII (Speech Intelligibility Index). SII is an index for predicting speech intelligibility and may be determined by Equation 1 below.

At this time, S means a percent correct score, and may be a value related to a positive response rate for speech data. P means the proficiency value of the speaker and listener, and 1 in the case of normal people. Q and n may mean fitting constants.

In step S240, the server 10 may determine the first frequency gain value and the first maximum output value based on the weight.

4 is a flowchart for explaining a method of acquiring parameters according to the size of a conversation sound according to an embodiment of the present invention.

In step S260, the server 10 may determine the loudness of the dialogue sound of the Korean voice data.

At this time, the loudness of the dialogue sound can be obtained based on the experimental data measured inside the soundproof booth. The experimental data at this time may be first voice data to be described later.

Specifically, after the subject undergoes a basic hearing test, the subject is seated in the center of the soundproofing room and a Sound Level Meter (SLM) is installed 1 m from the face of the subject. While the subject is reading the sentence used for the general sentence test, the tester uses the SLM to measure the speaker's speech volume, and the subject speaks each with three intensities (speaking quietly, speaking loudly, speaking loudly). And the LZeq value corresponding to the frequency of the 1/3 octave band can be measured. The measured values for this are shown in Table 2 below, and a graph showing a DB according to frequency and sound intensity is shown in FIG. 12.

Vocal Effort frequency LZeq 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500 3150 4000 5000 6300 8000 Average

Casual 27.45 29.30 39.05 41.22 30.75 25.00 31.58 26.52 29.03 22.42 21.25 24.70 20.53 18.95 18.62 19.83 19.45 17.02 17.17 52.93 Normal 32.02 34.10 50.80 54.85 46.22 44.12 47.72 44.65 45.67 39.15 33.25 36.23 32.90 29.05 27.62 24.82 25.53 24.17 24.67 59.28 Raised 31.87 36.18 51.92 59.57 54.98 50.32 56.73 53.98 55.40 50.32 45.47 48.67 43.78 40.57 39.22 35.27 32.70 29.32 29.60 65.75 Standard
Deviation Casual 2.10 5.31 7.35 11.22 8.60 6.16 9.28 6.93 6.58 3.88 8.11 7.56 4.30 4.77 4.85 5.44 4.81 5.68 5.05 2.37 Normal 11.19 5.62 4.30 5.16 5.90 6.94 5.44 3.20 5.87 3.61 7.46 6.80 5.72 4.06 6.11 5.52 4.46 6.01 4.91 4.55 Raised 10.56 12.02 4.02 5.03 8.29 8.66 8.44 5.63 6.70 5.95 10.04 10.02 8.86 6.87 9.48 7.87 4.87 7.61 6.27 6.58

In step S270, the server 10, if the conversation volume is within a preset range, compression threshold, compression ratio, extension threshold, extension ratio and the weight A first parameter may be determined based on, and a first frequency gain value and a first maximum output value may be determined based on the determined first parameter. At this time, the compression threshold, compression ratio, expansion threshold, and expansion ratio are nonlinear amplification methods. It is used in hearing aids using. It is used to change the high-intensity conversation sound less and the low-intensity conversation sound higher. That is, when amplifying conversational sounds of all intensities through the same method, the very small conversational sound is not sufficiently amplified, or the very large conversational sound is excessively amplified, so that distortion of sound may occur, so that the server 10 has conversational sound Depending on the size of the (depending on the compression threshold), you can adjust the degree of the middle of the dialogue sound (using the compression ratio). Specifically, when the volume of the dialogue sound is normal (normal), The first frequency gain value and the first maximum output value may be determined based on the individual hearing threshold of the hearing aid user and the weight determined above.

However, the server 10, if the size of the conversation sound is normal (normal) or less than (Casual) or loud (Raised), the hearing threshold of the individual hearing threshold of the user and the weight determined above, as well as the compression threshold (Compression Threshold) , A first frequency gain value and a first maximum output value may be determined by further considering a compression ratio. For example, when the LZeq in the 2000 Hz band is 45, the server 10 determines that the conversation sound is a conversation sound larger than a normal conversation sound, and considers the compression threshold and compression ratio for this. Thus, the first frequency gain value and the first maximum output value can be determined.

Through the above-described method, the server 10 may obtain a correction method of the hearing aid 20 suitable for Koreans. As an embodiment, FIGS. 13 and 14 show graphs of frequency gain values in each frequency band according to input sound pressure level (input) and compression threshold level (CT) for 50 dB flat hearing loss adults. 13 and 14, the frequency gain value for the compression threshold for CT31 can be increased to 30db and decreased after 30db. The frequency gain value for the compression threshold for CT41 can increase to 40db and decrease after 40db. The frequency gain value for the compression threshold for CT51 remains constant until around 50db and can decrease thereafter. The frequency gain value for the compression threshold for CT61 remains constant until around 60db and may decrease thereafter. The frequency gain value for the compression threshold for CT71 remains constant until around 70db and may decrease thereafter. The frequency gain value for each compression threshold can be obtained by the embodiment of FIG. 3 described above.

Meanwhile, according to an embodiment of the present invention, the first parameter may be calculated by applying the following formula according to each condition. For convenience of explanation, X1 is defined as the HTL (Hearing Threshold Level), X2 is the volume of the input sound, and Y (XX) is defined as the first parameter in the condition XX. In addition, it is assumed that when the calculated Y (XX) <0, Y (XX) = 0. In addition, the eight frequency bands described above are the first frequency band for the center frequency of 250 Hz, the second frequency band for the center frequency of 500 Hz, the third frequency band for the center frequency of 800 Hz, the fourth frequency band for the center frequency of 1000 Hz, and the center. It is assumed that the fifth frequency band for the frequency 1600 Hz, the sixth frequency band for the central frequency 2500 Hz, the seventh frequency band for the central frequency 4000 Hz, and the eighth frequency band for the central frequency 6000 Hz.

In the first embodiment, the first parameter Y (ET30) for each frequency band in the extended threshold level (ET) is ET30, X2 = <30dB SPL (Sound Pressure Level), Y (ET30) in the first frequency band = 0.714X1 + 0.041X2-37.243, in the second frequency band Y (ET30) = 0.630X1 + 0.201X2-20.364, in the third frequency band Y (ET30) = 0.598X1 + 0.251X2-13.206, in the fourth frequency band Y (ET30) = 0.566X1 + 0.288X2-3.016, Y (ET30) = 0.548X1 + 0.289X2-0.327 in the 5th frequency band, Y (ET30) = 0.513X1 + 0.289X2 + 2.934, the 6th frequency band In the 7th frequency band, Y (ET30) = 0.302X1 + 0.167X2 + 4.271, and in the 8th frequency band, Y (ET30) = 0.240X1 + 0.128X2 + 1.266.

In the second embodiment, the first parameter Y (ET40) for each frequency band at the extended threshold level ET is ET40, X2 = <40dB SPL, Y (ET40) = 0.714X1 + 0.016 in the first frequency band. X2-37.243, Y (ET40) = 0.630X1 + 0.075X2-20.364 in the second frequency band, Y (ET40) = 0.598X1 + 0.094X2-13.206 in the third frequency band, Y (ET40) = in the fourth frequency band 0.566X1 + 0.108X2-3.016, Y (ET40) = 0.548X1 + 0.108X2-0.327 in 5th frequency band, Y (ET40) = 0.513X1 + 0.108X2 + 2.934 in 6th frequency band, Y in 7th frequency band (ET40) = 0.302X1 + 0.063X2 + 4.271, in the eighth frequency band, Y (ET40) = 0.240X1 + 0.048X2 + 1.266.

In a third embodiment, X2 = <40dB SPL, and in a hearing aid using a linear amplification method, the first parameter Y (Lin1) for each frequency band is

In the first frequency band Y (Lin1) = 0.714X1-37.243, in the second frequency band Y (Lin1) = 0.630X1-20.364, in the third frequency band Y (Lin1) = 0.598X1-13.206, in the fourth frequency band Y (Lin1) = 0.566X1-3.016, in the fifth frequency band Y (Lin1) = 0.548X1-0.327, in the sixth frequency band Y (Lin1) = 0.513X1 + 2.934, in the seventh frequency band Y (Lin1) = It may be 0.302X1 + 4.271, Y (Lin1) = 0.240X1 + 1.266 in the eighth frequency band.

In a fourth embodiment, X2 = <50dB SPL, and the first parameter Y (Lin2) for each frequency band in a hearing aid using a linear amplification method is Y (Lin2) = 0.714X1-37.243 in the first frequency band. In the second frequency band Y (Lin2) = 0.630X1-20.364, in the third frequency band Y (Lin2) = 0.598X1-13.206, in the fourth frequency band Y (Lin2) = 0.566X1-3.016, in the fifth frequency band Y (Lin2) = 0.548X1-0.327, Y (Lin2) = 0.513X1 + 2.934 in 6th frequency band, Y (Lin2) = 0.302X1 + 4.271 in 7th frequency band, Y (Lin2) = in 8th frequency band 0.240X1 + 1.266.

In a fifth embodiment, X2 = <60dB SPL, and the first parameter Y (Lin3) for each frequency band in a hearing aid using a linear amplification method is Y (Lin3) = 0.714X1-37.243 in the first frequency band. In the second frequency band Y (Lin3) = 0.630X1-20.364, in the third frequency band Y (Lin3) = 0.598X1-13.206, in the fourth frequency band Y (Lin3) = 0.566X1-3.016, in the fifth frequency band Y (Lin3) = 0.548X1-0.327, Y (Lin3) = 0.513X1 + 2.934 in 6th frequency band, Y (Lin3) = 0.302X1 + 4.271 in 7th frequency band, Y (Lin3) = in 8th frequency band 0.240X1 + 1.266.

In the sixth embodiment, the first parameter Y (CT31) for each frequency band at a compression threshold level (CT) CT31, X2> = 31dB SPL is Y (CT31) = 0.714X1-0.062 in the first frequency band. X2-34.143, Y (CT31) = 0.630X1-0.301X2-5.314 in the second frequency band, Y (CT31) = 0.598X1-0.377X2 + 5.644 in the third frequency band, Y (CT31) = in the fourth frequency band 0.566X1-0.432X2 + 18.584, Y (CT31) = 0.548X1-0.433X2 + 21.323 in 5th frequency band, Y (CT31) = 0.513X1-0.433X2 + 24.584 in 6th frequency band, Y in 7th frequency band (CT31) = 0.302X1-0.250X2 + 16.771, Y (CT31) = 0.240X1-0.192X2 + 10.866 in the eighth frequency band.

In the seventh embodiment, the first parameter Y (CT41) for each frequency band at a compression threshold level (CT) of CT41, X2> = 41dB SPL is Y (CT41) = 0.714X1-0.062 in the first frequency band. X2-34.143, Y (CT41) = 0.630X1-0.301X2-5.314 in the second frequency band, Y (CT41) = 0.598X1-0.377X2 + 5.644 in the third frequency band, Y (CT41) = in the fourth frequency band 0.566X1-0.432X2 + 18.584, Y (CT41) = 0.548X1-0.433X2 + 21.323 in 5th frequency band, Y (CT41) = 0.513X1-0.433X2 + 24.584 in 6th frequency band, Y in 7th frequency band (CT41) = 0.302X1-0.250X2 + 16.771, Y (CT41) = 0.240X1-0.192X2 + 10.866 in the eighth frequency band.

In the eighth embodiment, when the compression threshold level CT is CT51, X2> = 51 dB SPL, the first parameter Y (CT51) for each frequency band is Y (CT51) = 0.714X1-0.062 in the first frequency band. X2-34.143, Y (CT51) = 0.630X1-0.301X2-5.314 in the second frequency band, Y (CT51) = 0.598X1-0.377X2 + 5.644 in the third frequency band, Y (CT51) = in the fourth frequency band 0.566X1-0.432X2 + 18.584, Y (CT51) = 0.548X1-0.433X2 + 21.323 in 5th frequency band, Y (CT51) = 0.513X1-0.433X2 + 24.584 in 6th frequency band, Y in 7th frequency band (CT51) = 0.302X1-0.250X2 + 16.771, Y (CT51) = 0.240X1-0.192X2 + 10.866 in the eighth frequency band.

In the ninth embodiment, the first parameter Y (CT61) for each frequency band in which the compression threshold level (CT) is CT61, X2> = 61dB SPL, Y (CT61) = 0.714X1-0.078 in the first frequency band X2-33.507, Y (CT61) = 0.630X1-0.37625X2 +2.211 in the second frequency band, Y (CT61) = 0.598X1-0.47125X2 + 15.069 in the third frequency band, Y (CT61) = in the fourth frequency band 0.566X1-0.540X2 + 29.384, Y (CT61) = 0.548X1-0.54125X2 + 32.148 in 5th frequency band, Y (CT61) = 0.513X1-0.54125X2 + 35.409 in 6th frequency band, Y in 7th frequency band (CT61) = 0.302X1-0.3125X2 + 23.021, Y (CT61) = 0.240X1-0.240X2 + 15.666 in the eighth frequency band.

In a tenth embodiment, the first parameter Y (CT71) for each frequency band in which the compression threshold level (CT) is CT71, X2> = 71dB SPL, Y (CT71) = 0.714X1-0.103333 in the first frequency band X2-30.0097, Y (CT71) = 0.630X1-0.501667X2 +14.75267 in the second frequency band, Y (CT71) = 0.598X1-0.628333X2 + 30.77733 in the third frequency band, Y (CT71) = in the fourth frequency band 0.566X1-0.720X2 + 47.384, Y (CT71) = 0.548X1-0.7217X2 + 50.192 in 5th frequency band, Y (CT71) = 0.513X1-0.7217X2 + 53.453 in 6th frequency band, Y in 7th frequency band (CT71) = 0.302X1-0.41667X2 + 33.43767, Y (CT71) = 0.240X1-0.320X2 + 23.666 in the eighth frequency band.

5 is a flowchart illustrating a method of acquiring a frequency band using an artificial intelligence model according to an embodiment of the present invention.

In step S211, the server 10 may acquire the first voice data. Specifically, the server 10 generates the text included in the Korean standard sentence list for adults (KS-SL-A) inside the soundproof booth, but the speaker of the Korean speech data is located in the center of the soundproof booth, and the microphone is placed in front of the speaker 1M. Located, it is possible to obtain the first voice data through the microphone.

The first voice data obtained in step S211 may be used as data for obtaining the first frequency gain value and the first maximum output value in FIGS. 3 and 4 described above. However, since determining the first frequency gain value and the first maximum output value based only on text included in the Korean standard sentence list for adults (KS-SL-A) does not include various words and sentences, the server 10 Various Korean voice data that can be collected in everyday life can be collected and converted and used as data for determining a first frequency gain value and a first maximum output value.

To this end, in step S212, the server 10 may train the artificial intelligence model to extract feature values for Korean speech data by inputting the first speech data into the artificial intelligence model.

* In step S213, the server 10 inputs the second voice data including noise into the trained artificial intelligence model to extract feature values for the second voice data, and based on the extracted feature values, the second voice data Can be corrected. The second voice data is voice data including Korean, and may be voice data that can be collected in various environments. For example, audio data output from video content, audio data collected from a distance may be various.

That is, the server 10 may input the second voice data into the artificial intelligence model learned based on the first voice data, and correct the second voice data in a format of first voice data suitable for hearing aid correction. Specifically, the server 10 extracts feature values of the first voice data, extracts feature values of the second voice data, and corrects feature values of the second voice data based on the feature values of the first voice data. , The second audio data may be corrected based on the characteristic values of the corrected second audio data.

In step S214, the server 10 may acquire a plurality of frequency bands based on Korean voice data including first voice data and corrected second voice data.

As described above, the server 10 may acquire various Korean voice data, thereby determining a correction method of the hearing aid 20 that is more suitable for Koreans, and further, various words according to times, such as new words, sweet words, and abbreviations There is an effect that can reflect the sentence.

6 is a flowchart for explaining a method of secondaryly correcting a hearing aid according to an embodiment of the present invention.

In FIGS. 3 to 5, the hearing aid 20 is first corrected according to the overall Korean voice data used by Koreans, but a method of secondly correcting the hearing aid suitable for each hearing aid user may be required.

Accordingly, in step S410, the server 10 corrects the frequency gain value and the maximum output value for each of the plurality of frequency bands based on the data on the hearing state of the hearing aid user, thereby obtaining the second frequency gain value and the second maximum output value. Can be obtained. At this time, the user

In step S420, the server 10 may extract a feature value for correcting the waveform of the sound source detected by the hearing aid 20 based on the voice data of the hearing aid user.

Specifically, when the voice data generated by the hearing aid user passes through the hearing aid 20 and is corrected, the actual voice of the hearing aid user and the voice heard by the hearing aid user's ear may be different. Therefore, the server 10 may correct the voice data by acquiring a feature value for the waveform of the sound source to amplify the size of the voice data as well as to correct the waveform of the voice data similar to the actual voice data.

In step S510, the hearing aid 20 may second-compensate the first-corrected hearing aid 20 based on the second frequency gain value, the second maximum output value, and feature values for correcting the waveform of the sound source.

7 is a flowchart for explaining a method of third-order correction of a hearing aid according to an embodiment of the present invention.

In step S610, the hearing aid 20 may transmit voice data input to the hearing aid 20 to the electronic device 30.

Specifically, the hearing aid 20 may transmit both the voice data sensed through the microphone and the voice data output through the receiver to the electronic device 30. To this end, the hearing aid 20 may further include a communication unit and a memory. The hearing aid 20 and the electronic device 30 may be connected via Bluetooth. In addition, the memory of the hearing aid 20 may store the minimum data for miniaturization of the hearing aid 20, and the voice data transmitted to the electronic device 30 after storage may be immediately deleted to store the next voice data. The hearing aid 20 may determine a voice data transmission cycle based on the storage space of the memory. That is, the hearing aid 20 may transmit voice data to the electronic device 30 only when the memory storage space fills more than a preset ratio. Through the above-described method, the number of times the hearing aid 20 transmits voice data to the electronic device 30 is minimized to prevent unnecessary power consumption of the hearing aid 20.

In another embodiment, the hearing aid 20 performs a communication connection with the electronic device 30 when the memory storage space fills more than a preset ratio, and cuts off the communication connection after transmitting the voice data to consume unnecessary power of the hearing aid 20 Can be prevented.

In another embodiment, when it is determined that a communication connection with the electronic device 30 is impossible, the hearing aid 20 may not store the voice data obtained through the microphone or receiver in the memory.

In step S620, the electronic device 30 may transmit the received voice data to the server 10.

In step S630, the server 10 may obtain a third frequency gain value and a third maximum output value for the third correction of the hearing aid 20 based on the received voice data when the preset condition is satisfied.

In this case, the preset condition may be a condition based on the amount of voice data received. That is, when the amount of the received voice data is greater than or equal to a preset amount, the server 10 obtains a third frequency gain value and a third maximum output value for the third correction of the hearing aid 20 based on the received voice data Can be.

In step S640, the hearing aid 20 may perform a third correction on the second-corrected hearing aid 20 based on the third frequency gain value and the third maximum output value.

8 is a flowchart illustrating a method of fourth-order correction of a hearing aid according to an embodiment of the present invention.

In step S710, the server 10 may collect background information including information about the background noise in the living space of the hearing aid user, information about the speech speed of the hearing aid user, and information about the average speech speed of the general public. .

In step S720, the server 10 may determine the background noise and the speech rate to be included in the hearing training program based on the collected background information.

In step S730, the server 10 may generate an auditory training program based on the determined background noise and utterance speed. At this time, the generated hearing training program may determine whether to include visual cues, phonemes, word and sentence difficulty, sentence length, environmental sounds, and music according to the background information of the hearing aid user.

In one embodiment, when the speech rate of the hearing aid user is slower than the average speech rate of the general public, the server 10 determines the speech rate of the hearing training program based on the speech rate of the hearing aid user, and gradually the speech of the hearing training program The speed can be increased to the average ignition rate of the general public. Alternatively, when the speech rate of the hearing aid user is faster than the average speech rate of the general public, the server 10 determines the speech rate of the hearing training program based on the average speech rate of the public, and gradually increases the speech rate of the hearing training program. It can be increased at the user's ignition speed.

In another embodiment, the server 10 may generate a hearing training program based on information about the surrounding environment of the hearing aid user. For example, the server 10 acquires background noise information in the home of a hearing aid user, background noise information in the workplace, and background noise information in a moving path, and generates an auditory training program based on the acquired background noise information can do.

In step S740, the server 10 may receive result data for the hearing training program and hearing aid wearing from the hearing aid user.

In step S750, the server 10 may fourth-correct the third-corrected hearing aid 20 based on the result data.

Specifically, in step S751, the server 10 may obtain a frequency section having an incorrect answer rate of a preset ratio or more among result data for the hearing training program.

In step S752, the server 10 may correct the frequency gain value of the obtained frequency section to be greater than or equal to the third frequency gain value.

That is, the server 10 may acquire a frequency band for voice data in which a false answer rate is greater than or equal to a preset rate among result data for the hearing training program, and amplify the sound of the obtained frequency band. Specifically, the server 10 may determine the fourth frequency gain value over the third frequency gain value previously corrected. However, it goes without saying that the fourth frequency gain value may be determined such that the amplification value of the audio data by the third frequency gain value is equal to or less than the third maximum output value.

9 is a flowchart illustrating a method of obtaining usage data for a hearing aid according to an embodiment of the present invention.

In step S731, the server 10 may store a plurality of first data for each of a plurality of hearing aid users. At this time, the plurality of first data may include background information, answer information for a questionnaire question, and detailed information obtained from the answer information.

Furthermore, the server 10 uses all of the background information, the basic information obtained based on the background information, the questionnaire generated based on the basic information, the response information for the questionnaire question, and the result information generated from the answer information as the first data. Of course it can be saved. At this time, the server 10 may store the acquired background information, basic information, questionnaire question, answer information, and result information for each user. That is, the stored first data can be grouped into background information, basic information, questionnaire questions, answer information and result information for the first user, background information, basic information, questionnaire questions, answer information, and result information information for the second user. Can be.

In step S732, the server 10 may store a plurality of second data for a plurality of hearing aid users for each of the plurality of users. In this case, the second data may include result data for the hearing training program and hearing aid wearing.

In step S733, the server 10 may obtain a plurality of first clusters by clustering a plurality of first data.

That is, the server 10 may group data for a plurality of hearing aid users into similar data.

However, the present invention is not limited thereto, and in one embodiment, the server 10 may perform clustering for each of data on background information, basic information, questionnaire question, answer information, and result information. In this case, the server 10 may acquire a plurality of background information clusters, basic information clusters, questionnaire question clusters, answer information clusters, and result information clusters, respectively. The server 10 may match and store each background information cluster, basic information cluster, questionnaire question cluster, answer information cluster, and result information cluster. Specifically, the server 10 includes a first basic information cluster associated with the first background information cluster, a first questionnaire cluster associated with the first basic information cluster, a first answer information cluster and a first answer associated with the first questionnaire cluster The first result information cluster associated with the information cluster may be matched to obtain one first group. In the same way, the server 10 can be grouped by matching all clusters for each background information cluster, basic information cluster, questionnaire question cluster, answer information cluster, and result information cluster. The server 10 may obtain a plurality of first clusters by clustering data for a plurality of grouped groups.

That is, the server 10 classifies background information, basic information, questionnaire question, answer information, and result information included in the first data, and is based on the categorized results, and is not virtual data (ideal). Data about the user can be obtained. For example, when clustering is performed based on the first data for an actual user, there is an error in data classification and analysis when the actual user incorrectly inputs the answer information for the questionnaire and does not obtain accurate detailed information. Can be. Accordingly, when creating a group for background information, basic information, questionnaire question, answer information, and result information based only on information obtained by each user, there is an effect of reducing the above-described error.

In step S734, the server 10 may obtain a plurality of second clusters by clustering a plurality of second data.

In step S735, the server 10 may acquire background information and answer information for the new first hearing aid user.

Furthermore, of course, the server 10 can receive all background information, basic information, questionnaire questions, answer information, and result information from the first hearing aid user.

In step S736, the server 10 may determine a 1-1 cluster related to background information and answer information for the first hearing aid user among the plurality of first clusters.

In step S737, the server 10 may determine a 2-1 cluster existing within a predetermined distance from the 1-1 cluster among the plurality of second clusters.

In step S738, the server 10 may generate an aural training program suitable for the first hearing aid user based on the data included in the 2-1 cluster.

That is, the first hearing aid user will have background information similar to the hearing aid user belonging to the first-first cluster. Accordingly, the server 10 acquires the 2-1 cluster associated with the user of the 1-1 cluster, obtains the 2-1 cluster, and based on the data included in the acquired 2-1 cluster, the first Audience training programs suitable for hearing aid users can be created.

However, according to various embodiments of the present invention, the server 10 may store a plurality of hearing aid users' test wear data and cluster them to generate a plurality of third clusters. At this time, the test wear data is data obtained by a plurality of hearing aid users using the first corrected hearing aid, data used for the second correction described in FIG. 6 or data used for the third correction described in FIG. 7. Can mean The server 10 determines a 3-1 cluster existing within a predetermined distance from the 1-1 cluster among a plurality of third clusters, and exists within a preset distance from the 3-1 cluster among the plurality of second clusters It goes without saying that the 2-1 cluster may be determined. Alternatively, the server 10 may determine the 2-1 cluster such that the 2-1 cluster is within a predetermined distance from both the 1-1 cluster and the 3-1 cluster.

15 is a configuration diagram of an apparatus according to an embodiment.

The processor 102 may include one or more cores (not shown) and a connection passage (for example, a bus) for transmitting and receiving signals to and from a graphic processing unit (not shown) and / or other components. .

The processor 102 according to an embodiment performs the methods described with respect to FIGS. 1 to 9 by executing one or more instructions stored in the memory 104.

For example, the processor 102 acquires new learning data by executing one or more instructions stored in memory, and uses the trained model to perform tests on the acquired new learning data, and the test results and labeling. Extracts the first learning data obtained by the obtained information with an accuracy equal to or greater than a predetermined first reference value, deletes the extracted first learning data from the new learning data, and extracts the new learning data from which the extracted learning data is deleted. By using it, the trained model can be trained again.

On the other hand, the processor 102 is a RAM (Random Access Memory, not shown) and a ROM (Read-Only Memory) that temporarily and / or permanently stores signals (or data) processed inside the processor 102. , Not shown). Further, the processor 102 may be implemented in the form of a system on chip (SoC) including at least one of a graphic processing unit, RAM, and ROM.

Programs (one or more instructions) for processing and controlling the processor 102 may be stored in the memory 104. Programs stored in the memory 104 may be divided into a plurality of modules according to functions.

The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, a software module executed by hardware, or a combination thereof. The software modules may include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer readable recording medium well known in the art.

The components of the present invention may be implemented as a program (or application) to be executed in combination with a hardware computer, and stored in a medium. The components of the present invention can be implemented in software programming or software components, and similarly, embodiments include C, C ++, including various algorithms implemented in a combination of data structures, processes, routines or other programming components. , Can be implemented in programming or scripting languages such as Java, assembler, etc. Functional aspects can be implemented with algorithms running on one or more processors.

The embodiments of the present invention have been described above with reference to the accompanying drawings, but a person skilled in the art to which the present invention pertains may implement the present invention in other specific forms without changing its technical spirit or essential features. You will understand. Therefore, it should be understood that the above-described embodiments are illustrative in all respects and not restrictive.

10: Server
20: hearing aid
30: electronic device

Claims (1)

In the control method of the hearing aid system performed by the server and the hearing aid,
Obtaining, by the server, a plurality of frequency bands based on Korean voice data;
Obtaining, by the server, a first frequency gain value and a first maximum output value for each of the plurality of frequency bands;
The hearing aid performing a primary correction of the hearing aid based on a first frequency gain value and a first maximum output value for each of the obtained plurality of frequency bands;
Acquiring data about the hearing state of the hearing aid user and voice data of the hearing aid user by the server; And
The hearing aid performing a second correction of the first corrected hearing aid based on data on the hearing state of the hearing aid user and voice data of the hearing aid user; Including,
The step of acquiring data about the hearing state of the hearing aid user and the voice data of the hearing aid user may include:
Obtaining a second frequency gain value and a second maximum output value by correcting the frequency gain value and the maximum output value for each of a plurality of frequency bands based on data on the hearing state of the hearing aid user;
And extracting a feature value for correcting a waveform of a sound source sensed by the hearing aid based on the voice data of the hearing aid user.
The second correction step,
And a second correction of the first corrected hearing aid based on the second frequency gain value, the second maximum output value, and a feature value for correcting the waveform of the sound source.
The control method,
The hearing aid transmitting the voice data input from the hearing aid to an electronic device;
The electronic device transmitting the received voice data to the server;
If the server satisfies a preset condition, acquiring a third frequency gain value and a third maximum output value for thirdly correcting the hearing aid based on the received voice data; And
The hearing aid performing a third correction of the second-corrected hearing aid based on the third frequency gain value and the third maximum output value; Including,
The control method,
Collecting, by the server, background information including information about background noise in the living space of the hearing aid user, information about the speech speed of the hearing aid user, and information about the average speech rate of the general public;
Determining, by the server, background noise and speech rate to be included in the hearing training program based on the collected background information;
Generating, by the server, a hearing training program based on the determined background noise and utterance speed;
Receiving, by the server, result data for the hearing training program and hearing aid wearing from the hearing aid user; And
The hearing aid, the fourth step of correcting the hearing aid based on the result data; Including,
The fourth correction step,
Obtaining a frequency section having a wrong answer rate of a preset ratio or higher among the result data;
Comprising the step of correcting the frequency gain value of the obtained frequency interval to more than the third frequency gain value; includes,
The generated hearing training program determines whether visual cues, phonemes, word and sentence difficulty, sentence length, environment sounds, and music are included according to the background information of the hearing aid user,
The step of generating the hearing training program,
Storing a plurality of first data for each of the plurality of users including background information of a plurality of hearing aid users, answer information on a questionnaire question, and detailed information obtained from the answer information;
Storing a plurality of second data including a hearing training program of the plurality of hearing aid users and result data on wearing the hearing aid for each of the plurality of hearing aid users;
Clustering the plurality of first data to obtain a plurality of first clusters;
Clustering the plurality of second data to obtain a plurality of second clusters;
Obtaining background information and response information for a new first hearing aid user;
Determining a first-first cluster related to background information and answer information about the acquired first hearing aid user among the plurality of first clusters;
Determining a 2-1 cluster existing within a predetermined distance from the 1-1 cluster among the plurality of second clusters; And
Generating a hearing training program for the first hearing aid user based on the data included in the 2-1 cluster; Including,
Obtaining a first frequency gain value and a first maximum output value for each of the plurality of frequency bands,
Classifying the Korean speech data into eight frequency bands;
Obtaining a positive response rate for each of the classified eight frequency bands;
Obtaining respective weights for each frequency band based on the obtained positive response rate, and
Determining the first frequency gain value and the first maximum output value based on the weight; Including,
Obtaining a first frequency gain value and a first maximum output value for each of the plurality of frequency bands,
Determining a conversation volume of the Korean voice data;
When the conversation sound volume is within a predetermined range, a first parameter is determined based on the compression threshold, compression ratio, extension threshold, extension ratio, and weight And determining the first frequency gain value and the first maximum output value based on the determined first parameter.
The first parameter,
The extended threshold level (ET) is the first parameter Y (ET30) for each frequency band at ET30, X2 = <30dB SPL,
The Y (ET30) is Y (ET30) = 0.714X1 + 0.041X2-37.243 in the first frequency band, Y (ET30) = 0.630X1 + 0.201X2-20.364 in the second frequency band, Y (ET30) in the third frequency band ) = 0.598X1 + 0.251X2-13.206, Y (ET30) = 0.566X1 + 0.288X2-3.016 in 4th frequency band, Y (ET30) = 0.548X1 + 0.289X2-0.327, 6th frequency band in 5th frequency band At Y (ET30) = 0.513X1 + 0.289X2 + 2.934, in the 7th frequency band Y (ET30) = 0.302X1 + 0.167X2 + 4.271, in the 8th frequency band Y (ET30) = 0.240X1 + 0.128X2 + 1.266 Determined by the formula,
X1 is HTL, X2 is the control method characterized in that the volume of the input sound.

Images