Language training hearing aid system carrying AI speech anomaly evaluation algorithm

Abstract

The invention discloses a speech training hearing aid system carrying an AI speech anomaly assessment algorithm, and relates to the AI speech anomaly assessment algorithm, which is characterized in that comprehensive hearing assessment is carried out by collecting a composite signal and formulating an age adaptation rule base, hearing threshold dispersion and signal-to-noise ratio attenuation are obtained, and then a hearing level assessment coefficient is obtained; daily speech evaluation is carried out by collecting speech feature data and formulating a standard speech comparison library, a daily speech anomaly evaluation model is constructed by using a convolutional neural network learning model, and then a daily speech evaluation coefficient is obtained; by collecting multi-dimensional data, professional speech evaluation is carried out, and then a professional speech evaluation coefficient is obtained; and according to the hearing level evaluation coefficient, the daily speech evaluation coefficient and the professional speech evaluation coefficient, analysis is performed to obtain an analysis result, and a comprehensive analysis scheme is formulated, so that the accuracy is greatly improved.

Description

Technical Field

[0001] The present invention relates to an AI speech anomaly assessment algorithm, and in particular to a speech training hearing aid system equipped with the AI ​​speech anomaly assessment algorithm. Background Art

[0002] Congenital hearing loss is a disease that seriously affects the development of children's hearing and speech. If it is not discovered and intervened in time, it will have a long-term or even irreversible impact on the physical and mental health of the child. Without any intervention, hearing-impaired children may have problems such as delayed language development and slurred speech. In severe cases, they may even become deaf-mute, leading to social disorders and psychological problems. However, through scientific intervention methods, such as hearing aids or cochlear implants and other hearing compensation devices, the hearing condition of children can be significantly improved, laying the foundation for improving their language learning ability.

[0003] In the early stages of life (0-3 years old), this is a critical period for children's language development and a critical stage for the development of the auditory center. Studies have shown that if effective intervention is carried out within 6 months after birth, including wearing hearing aids or cochlear implants, combined with systematic speech rehabilitation training, the language ability of children with hearing loss can reach the development level of normal children. However, in actual applications, the problems of real-time hearing and speech level monitoring of hearing-impaired children and the problem of quantifying speech training results by speech trainers and guardians make it difficult to fully utilize the effects of hearing compensation devices, resulting in the level of language development still lagging behind.

[0004] In this context, it is particularly important to develop a speech training hearing aid system equipped with an AI speech abnormality assessment algorithm. Combined with hearing aids or cochlear implants, through intelligent speech recognition technology, personalized training program design and real-time feedback mechanism, it helps guardians and speech trainers to carry out rehabilitation work more efficiently. For example, in infancy (0-3 years old), the device can assess children's sensitivity to sound and auditory discrimination ability through simple audio stimulation and gamified interactive tests; in the preschool period (3-8 years old), more targeted language training tasks can be designed, such as pronunciation exercises, vocabulary learning, etc.

[0005] In addition, intervention strategies that combine timelines can better help children gradually improve their language skills. In infancy, the focus is on developing basic auditory perception skills; in the preschool period, it is necessary to strengthen the training of language comprehension and expression skills; and after entering school age (3-8 years old), social communication skills and grammatical rules can be further strengthened. Through such phased and systematic intervention, the language development level of children will be significantly improved.

[0006] In short, early detection and scientific intervention of congenital hearing loss are crucial. Hearing compensation through hearing aids or cochlear implants, combined with professional speech training guidance, can effectively improve the language ability of children with hearing loss. The development of intelligent and personalized evaluation and training devices provides guardians and speech trainers with powerful tool support, which helps to maximize the development level of hearing-impaired children to reach the level of normal children and lay a solid foundation for their future integration into society.

[0007] Therefore, a speech training hearing aid system equipped with an AI speech abnormality assessment algorithm is now provided. Summary of the invention

[0008] In order to solve the above technical problems, the object of the present invention is to provide a speech training hearing aid system equipped with an AI speech abnormality assessment algorithm.

[0009] According to one preferred embodiment of the present invention, a speech training hearing aid system equipped with an AI speech abnormality assessment algorithm comprises: The hearing level assessment module is used to collect composite signals and formulate an age-adaptive rule library to conduct a comprehensive hearing assessment on hearing-impaired children, obtain the corresponding hearing threshold discreteness and signal-to-noise ratio attenuation, and then obtain the hearing level assessment coefficient of the corresponding hearing-impaired children; The daily speech assessment module is used to collect speech feature data and develop a standard speech comparison library, and then conduct daily speech assessment on hearing-impaired children. It also uses a convolutional neural network learning model to build a daily speech abnormality assessment model, and then obtain the daily speech assessment coefficient of the corresponding hearing-impaired children. Professional speech assessment module, used to collect multi-dimensional data, conduct professional speech assessment on hearing-impaired children, and then obtain the professional speech assessment coefficient of the corresponding hearing-impaired children; The speech level analysis module is used to obtain the hearing level assessment coefficient, the daily speech assessment coefficient and the professional speech assessment coefficient according to the above modules, and analyze them to obtain the analysis results, and formulate a comprehensive analysis plan for the corresponding hearing-impaired children based on the analysis results.

[0010] According to one preferred embodiment of the present invention, the process of collecting composite signals and formulating an age-adaptive rule base to conduct a comprehensive hearing assessment on hearing-impaired children includes: An embedded sound field generating device is provided to generate a composite signal, wherein the composite signal includes a pure tone signal and a speech noise signal; the pure tone signal is set with different frequency points; the speech noise signal generates a standard composite noise; Setting a speech signal acquisition device to collect the test sound signals of the hearing-impaired children in real time; Formulate an age adaptation rule base, wherein the age adaptation rule base includes age intervals and age correction compensation factors; According to the pure tone signal, speech noise signal and test sound signal, the corresponding hearing threshold dispersion and signal-to-noise ratio attenuation are obtained.

[0011] According to one preferred embodiment of the present invention, the process of obtaining the hearing level assessment coefficient of the corresponding hearing-impaired child according to the corresponding hearing threshold dispersion and signal-to-noise ratio attenuation includes: The test sound signal of the hearing-impaired child is collected in real time according to the different frequency points of the pure tone signal and the speech signal collection device to obtain the hearing threshold dispersion of adjacent frequency points; According to the standard composite noise and the dynamic range of -20dB to +10dB, the environment is divided into a quiet environment, a noisy environment, and a reverberant environment, thereby obtaining the corresponding signal-to-noise ratio attenuation; According to the threshold dispersion, the signal-to-noise ratio attenuation and the age correction compensation factor, a hearing level assessment coefficient corresponding to the hearing-impaired child is obtained.

[0012] According to one preferred embodiment of the present invention, the process of collecting speech feature data and formulating a standard speech comparison library to conduct daily speech assessment on hearing-impaired children includes: Formulate a standard speech comparison library to obtain standard speech data, wherein the standard speech data includes standard acoustic feature data and standard linguistic feature data; A high-precision microphone device is provided to extract speech features of the hearing-impaired child to obtain speech feature data; the speech feature extraction includes acoustic feature extraction and linguistic feature extraction, and the speech feature data includes acoustic feature data and linguistic feature data; Acquiring acoustic feature data according to the acoustic feature extraction; According to the linguistic feature extraction, linguistic feature data is obtained.

[0013] According to one of the preferred embodiments of the present invention, the process of constructing a daily speech anomaly assessment model using a convolutional neural network learning model includes: Obtaining several groups of acoustic feature data and linguistic feature data of different age groups in the historical collection period; Obtaining standard acoustic feature data and standard linguistic feature data of normal children of corresponding age groups in the standard speech comparison library; According to the several groups of acoustic feature data, linguistic feature data of different age groups in the historical collection period, standard acoustic feature data and standard linguistic feature data of the corresponding age groups, a training sample set is formed; Build a standard evaluation model based on the convolutional neural network learning model; The training sample set is input into the standard evaluation model, the standard evaluation model is trained, and the trained standard evaluation model is obtained, and then a daily speech abnormality evaluation model is constructed.

[0014] According to one of the preferred embodiments of the present invention, a daily speech assessment coefficient is obtained based on the daily speech abnormality assessment model.

[0015] According to one preferred embodiment of the present invention, the process of collecting multi-dimensional data and conducting professional speech assessment on hearing-impaired children includes: Conduct multi-dimensional quantitative assessments of hearing-impaired children based on standardized language assessment tools, including: quantitative assessments of language comprehension, language expression, vocabulary, grammar use, and speech clarity; A multi-dimensional data synchronous acquisition device is set up to obtain multi-dimensional data corresponding to quantitative evaluation, including: instruction execution accuracy, reaction time standard deviation, average sentence length, dependency distance, proportion of age-appropriate words, retelling syntactic tree editing distance and phoneme comparison accuracy.

[0016] According to one preferred embodiment of the present invention, the process of obtaining the professional speech assessment coefficient corresponding to the hearing-impaired child includes: According to the instruction execution accuracy, reaction time standard deviation, average sentence length, dependency distance, age-appropriate word ratio, repetition syntactic tree edit distance and phoneme comparison accuracy, the professional speech assessment coefficient of the corresponding hearing-impaired children was obtained, which is recorded as , professional speech assessment coefficient for: ; in, Quantitative evaluation dimension weight coefficients for language understanding, The execution accuracy of compound instructions, is the standard deviation of the reaction time, is the mean of the norm for the same age, is the standard deviation of the norm for the same age, Quantitative evaluation dimension weight coefficient for language expression, is the average sentence length, is the maximum compound instruction execution accuracy value corresponding to age, To prevent zero value, Dependence distance, The percentage of age-appropriate words, The weight coefficient of the vocabulary quantification evaluation dimension, is the edit distance of the paraphrase syntax tree, The weight coefficients of the quantitative evaluation dimensions for grammatical application, is the accuracy of phoneme comparison and Weight coefficients for speech intelligibility quantification evaluation dimensions.

[0017] According to one preferred embodiment of the present invention, the process of analyzing the hearing level assessment coefficient, the daily speech assessment coefficient and the professional speech assessment coefficient, obtaining the analysis results, and formulating a comprehensive analysis plan for hearing-impaired children based on the analysis results includes: According to the hearing level assessment coefficient, daily speech assessment coefficient and professional speech assessment coefficient, the analysis results are recorded as , the analysis results for: ; in, Hearing level assessment coefficient The weight coefficient of Evaluation coefficient for daily speech The weight coefficient of Professional speech assessment coefficient The weight coefficient of Setting standard analysis results ; like At this time, hearing-impaired children can go to normal schools; like When developing a comprehensive analysis plan for hearing-impaired children, the comprehensive analysis plan includes an adaptive speech training plan and speech training outcome assessment; The adaptive speech training program is suitable for the subject to be a guardian, to help hearing-impaired children with speech training; The speech training results evaluation is applicable when the subject is a speech training teacher. The speech training hearing aid plays an evaluation role, providing the speech training teacher with an effect evaluation standard, thereby assisting in completing the speech training course.

[0018] The present invention further provides a computer-readable storage medium, which stores a computer program. The computer program can be executed by a processor to implement the speech training hearing aid system equipped with an AI speech abnormality assessment algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0020] Figure 1 Shown is a schematic diagram of the steps of a speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to the present invention.

[0021] Figure 2Shown is a schematic diagram of the judgment flow of a speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to the present invention.

[0022] Figure 3 Shown is a schematic diagram of a speech training hearing aid system module equipped with an AI speech abnormality assessment algorithm according to the present invention. DETAILED DESCRIPTION

[0023] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0024] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0025] Please combine Figure 1 , Figure 2 as well as Figure 3 , a speech training hearing aid system equipped with an AI speech abnormality assessment algorithm, comprising a control center, wherein the control center is communicatively connected with a hearing level assessment module, a daily speech assessment module, a professional speech assessment module and a speech level analysis module; The hearing level assessment module is used to perform a comprehensive hearing assessment on the hearing-impaired child and obtain a hearing level assessment coefficient of the corresponding hearing-impaired child; It should be further explained that, in the specific implementation process, the specific process of the hearing level assessment module performing a comprehensive hearing assessment on the hearing-impaired child includes: Setting up an embedded sound field generating device; The embedded sound field generating device is used to generate a composite signal using DDS (direct digital synthesis) technology, and the composite signal includes a pure tone signal and a speech noise signal; The pure tone signal is set to seven frequency points of 125Hz, 250Hz, 500Hz, 1kHz, 2kHz, 4kHz, and 8kHz. It should be further explained that each frequency point has a duration of 300ms, a rise and fall time of 50ms, a sound pressure level range of 20-90dB SPL, and a step length of 5dB. The speech noise signal generates a 65dB composite noise background based on the International Speech Equalization Corpus (IEEEC standard), which is recorded as standard composite noise, and the signal-to-noise ratio dynamic range is adjustable from -20dB to +10dB; Setting up speech signal collection device; The speech signal acquisition device is composed of a bone conduction sensor array (including 6 piezoelectric ceramic units) arranged in the mastoid area, with a frequency response range of 80-10kHz, and is used to collect test sound signals of hearing-impaired children in real time; Develop an age-appropriate rule base; The age adaptation rule library includes age ranges and age correction compensation factors. It should be further explained that when the age range is 0-3 years old, the age correction compensation factor is +3dB; when the age range is 4-7 years old, the age correction compensation factor is 0dB; when the age range is greater than 7 years old, the age correction compensation factor is -2dB.

[0026] It should be further explained that, in the specific implementation process, the specific process of the hearing level assessment module obtaining the hearing level assessment coefficient of the corresponding hearing-impaired child includes: According to the different frequency points of the pure tone signal and the speech signal acquisition device, the test sound signal of the hearing-impaired child is collected in real time to obtain the hearing threshold dispersion of adjacent frequency points. It should be further explained that the specific process of obtaining the hearing threshold dispersion of adjacent frequency points includes: The seven frequency points are recorded as i=1, 2, 3, 4, 5, 6, with a total of 6 frequency bands; the hearing threshold corresponding to the test sound signal of the hearing-impaired child collected in real time in the corresponding i frequency band is recorded as ; The threshold discreteness of the i-th frequency band is denoted as , threshold dispersion for: ;in is the hearing threshold collected for the kth time, is the mean value of the hearing threshold collected for k times; According to the standard composite noise and the dynamic range of -20dB to +10dB, the environment is divided into quiet environment, noisy environment, and reverberant environment, which are denoted as j=1, 2, and 3 respectively; The signal-to-noise ratio attenuation at different stages is recorded as , signal-to-noise ratio attenuation for: ;in is the standard reference value, is the actual distance to the ear canal, is the standard ear canal distance; The age correction compensation factor in the age adaptation rule base is recorded as It should be further explained that the age correction compensation factor Automatically load according to the actual age of the hearing-impaired child; According to the threshold dispersion , signal-to-noise ratio attenuation and age correction compensation factor , obtain the hearing level assessment coefficient of the corresponding hearing-impaired child, recorded as ; Listening level assessment coefficient for: ;in is the corresponding threshold discreteness The weight coefficient of is the corresponding signal-to-noise ratio attenuation Speech recognition correction factor in the environment, is the actual age compensation weight coefficient; It should be further explained that according to the threshold dispersion corresponding to the seven frequency points Sum and average, signal-to-noise ratio attenuation under different environments And the actual age of the hearing-impaired child, to obtain the hearing level assessment coefficient .

[0027] The daily speech assessment module is used to perform daily speech assessment on hearing-impaired children, and to construct a daily speech abnormality assessment model to obtain daily speech assessment coefficients corresponding to hearing-impaired children; It should be further explained that, in the specific implementation process, the specific process of the daily speech assessment module performing daily speech assessment on the hearing-impaired child includes: Develop a standard language comparison library; By collecting standard speech data of normal children of different age groups, the standard speech data includes standard acoustic feature extraction data and standard linguistic feature extraction data; Divide the data into different age groups, perform noise reduction, frame division, and endpoint detection on the standard speech data of normal children of different age groups, extract the standard acoustic feature extraction data and standard linguistic feature extraction data of normal children of corresponding age groups, and form a standard speech comparison library; Set up a high-precision microphone device to receive the language performance of hearing-impaired children in natural communication scenarios; According to the high-precision microphone device, speech feature extraction is performed on the hearing-impaired child to obtain speech feature data; it should be further explained that the speech feature extraction includes acoustic feature extraction and linguistic feature extraction, and the speech feature data includes acoustic feature data and linguistic feature data; the acoustic feature extraction is used to extract the pronunciation clarity and accuracy of the hearing-impaired child; the linguistic feature extraction is used to evaluate the vocabulary and grammar of the hearing-impaired child; It should be further explained that, in the specific implementation process, the specific process of extracting acoustic features of hearing-impaired children and obtaining acoustic feature data includes: The acoustic feature data includes 20-dimensional MFCC coefficients, vowel pronunciation accuracy, syllable rate and pause frequency; The high-precision microphone device is used to collect the voice signal of the hearing-impaired child; The speech signal is framed and windowed, the speech signal is divided into 25ms frames, and the frames are shifted by 10ms and weighted by a Hamming window, and the weighted speech signal is passed through a first-order high-pass filter to enhance the high-frequency energy, and FFT calculation is performed to obtain the corresponding amplitude spectrum, and output according to the Mel filter group; it should be further explained that the Mel filter group consists of 40 triangular filters; The output 40-dimensional logarithmic energy is transformed by DCT, and the first 20-dimensional discrete logarithmic energy is retained to obtain the 20-dimensional MFCC coefficients; Through the YIN algorithm and LPC analysis technology, the fundamental frequency of vocal cord vibration and the resonance characteristics of the vocal tract are detected to obtain the corresponding vowel pronunciation accuracy; The speech signal is segmented into syllables, and the energy minimum point is found in the speech signal, and the interval > 80ms is regarded as an independent syllable; the effective speech duration and pause frequency of the speech signal are recorded, and the effective speech duration is recorded as , the pause frequency is recorded as ; According to the number of independent syllables and the effective speech duration, the syllable rate is obtained. for: ;in is the number of independent syllables; Will And pause frequency , recorded as speaking too fast; And pause frequency , recorded as normal speaking speed; And pause frequency , recorded as speaking too slowly; It should be further explained that, in the specific implementation process, the specific process of extracting linguistic features of hearing-impaired children and obtaining linguistic feature data includes: The linguistic feature data include vocabulary richness, syntactic complexity and pronunciation error rate; Counting the number of diverse words according to the speech signal, and performing word frequency distribution statistics on the number of diverse words through a children's language development norm database to obtain the vocabulary richness of the corresponding hearing-impaired child; According to a dependency syntax analyzer (such as a CTB8 model), extract the average dependency distance and the number of clauses of the speech signal; and obtain the corresponding syntactic complexity according to the dependency distance and the number of clauses; The speech signal is phoneme aligned through an ASR model, and pronunciation errors are detected on the phoneme-aligned speech signal to generate a pronunciation error rate; the pronunciation error detection includes substitution errors, omission errors, and distortion errors.

[0028] It should be further explained that, in the specific implementation process, the specific process of building a daily speech abnormality assessment model includes: Obtaining several groups of acoustic feature data and linguistic feature data of different age groups in the historical collection period; It should be further explained that the acoustic feature data and linguistic feature data of several groups of historical collection periods were classified according to the age groups of hearing-impaired children to facilitate the uniformity of the data; Obtaining standard acoustic feature data and standard linguistic feature data of normal children of corresponding age groups in the standard speech comparison library; According to several groups of acoustic feature data and linguistic feature data of different age groups in historical collection periods and standard acoustic feature data and standard linguistic feature data of corresponding age groups; The specific process of constructing a training sample set includes: The acoustic feature data and linguistic feature data of different age groups in several historical collection periods and the standard acoustic feature data and standard linguistic feature data of the corresponding age groups are grouped and labeled, and recorded as is a natural number; Will Several groups of acoustic feature data and linguistic feature data of different age groups in the historical collection period and standard acoustic feature data and standard linguistic feature data of the corresponding age groups are used as sample data, and is less than A natural number, and using the sample data, obtain the mean of the sample data, recorded as a sample set; The acoustic feature data and linguistic feature data of different age groups in the remaining groups during the historical collection period and the standard acoustic feature data and standard linguistic feature data of the corresponding age groups are used as test sets; According to the sample set and the test set, a training sample set is formed; Based on convolutional network neural network, a standard evaluation model is constructed; The training sample set is input into the standard evaluation model, the standard evaluation model is trained, and the trained standard evaluation model is obtained. The trained standard evaluation model is recorded as the daily speech abnormality evaluation model.

[0029] It should be further explained that, in the specific implementation process, the specific process of obtaining the daily speech assessment coefficient of the corresponding hearing-impaired children includes: According to the daily speech abnormality assessment model, the daily speech assessment coefficient is obtained for: ; in, is the weight coefficient of the 20-dimensional MFCC coefficient, and its value range is , is the 20-dimensional MFCC coefficient; is the weight coefficient of vowel pronunciation accuracy, and its value range is , is the vowel pronunciation accuracy; is the weight coefficient of the syllable rate, and its value range is , is the syllable rate; is the weight coefficient of the pause frequency, and its value range is , is the pause frequency; is the weight coefficient of vocabulary richness, and its value range is , for vocabulary richness; is the weight coefficient of syntactic complexity, and its value range is , is the syntactic complexity; is the weight coefficient of the pronunciation error rate, and its value range is , is the pronunciation error rate.

[0030] The professional speech assessment module is used to perform professional speech assessment on the hearing-impaired child and obtain the professional speech assessment coefficient of the corresponding hearing-impaired child; It should be further explained that, in the specific implementation process, the specific process of the professional speech assessment module performing professional speech assessment on the hearing-impaired child includes: Conduct multi-dimensional quantitative assessment on hearing-impaired children based on standardized language assessment tools, including quantitative assessment of language comprehension, language expression, vocabulary, grammar application, and speech clarity; A multi-dimensional data synchronous acquisition device is provided to obtain multi-dimensional data, wherein the multi-dimensional data includes compound instruction execution accuracy, reaction time standard deviation, average sentence length, dependency distance, age-appropriate word ratio, paraphrase syntax tree edit distance, and phoneme comparison accuracy; For the quantitative evaluation of language comprehension, the multidimensional data synchronous acquisition device collects the execution accuracy of compound instructions and the standard deviation of reaction time according to the Z-score standardization technology; for the quantitative evaluation of language expression, the multidimensional data synchronous acquisition device collects the average sentence length and dependency distance according to the percentile ranking conversion technology; for the quantitative evaluation of vocabulary, the multidimensional data synchronous acquisition device collects the proportion of age-adaptive words according to the logarithmic conversion technology and the linear scaling technology; for the quantitative evaluation of grammatical application, the multidimensional data synchronous acquisition device collects the editing distance of the paraphrase syntax tree according to the Levenshtein distance normalization technology; for the quantitative evaluation of speech clarity, the multidimensional data synchronous acquisition device collects the phoneme comparison accuracy according to the Gaussian mixture model probability density model.

[0031] It should be further explained that, in the specific implementation process, the specific process of the professional speech assessment module obtaining the professional speech assessment coefficient of the corresponding hearing-impaired child includes: The execution accuracy of the compound instruction, the standard deviation of the reaction time, the average sentence length, the dependency distance, the proportion of age-appropriate words, the editing distance of the repetition syntactic tree, and the accuracy of the phoneme comparison are marked and recorded as , , , , , as well as ; According to the execution accuracy of the compound instruction , standard deviation of reaction time , average sentence length , Dependence distance , the proportion of age-appropriate words , Paraphrase syntax tree edit distance and phoneme comparison accuracy , and obtain the professional speech assessment coefficient of the corresponding hearing-impaired child, recorded as ; Professional speech assessment coefficient for: ; in, Quantitative evaluation dimension weight coefficient for language understanding, is the mean of the norm for the same age, is the standard deviation of the norm for the same age, Quantitative evaluation dimension weight coefficient for language expression, is the maximum compound instruction execution accuracy value corresponding to age, To prevent zero value, The weight coefficient of the vocabulary quantification evaluation dimension, Quantitative evaluation of grammatical application dimensions weight coefficients and Weight coefficients for speech intelligibility quantification evaluation dimensions.

[0032] The speech level analysis module is used to analyze the hearing level assessment coefficient, the daily speech assessment coefficient and the professional speech assessment coefficient to obtain analysis results, and formulate a comprehensive analysis plan for the hearing-impaired child based on the analysis results; It should be further explained that, in the specific implementation process, the specific process of formulating a comprehensive analysis plan for hearing-impaired children includes: Evaluation coefficient according to listening level , Daily Speech Assessment Coefficient and professional speech assessment coefficient , the analysis results are recorded as , the analysis results for: ;in, Hearing level assessment coefficient The weight coefficient of Evaluation coefficient for daily speech The weight coefficient of Professional speech assessment coefficient The weight coefficient of Setting standard analysis results ; like At this time, hearing-impaired children can go to normal schools; like When developing a comprehensive analysis plan for hearing-impaired children, the comprehensive analysis plan includes an adaptive speech training plan and speech training outcome assessment; The adaptive speech training program is applicable to when the subject is a guardian, and the speech training hearing aid cooperates with the mobile phone APP to guide the guardian and help the hearing-impaired child to conduct speech training, which mainly includes the following parts: 1. Auditory training: gradually improve children's hearing ability from sound perception, discrimination, recognition to understanding; play different types of scenes (animals, vehicles) through mobile phone APP, and play the sound in the ear simultaneously, so that children can identify the source and characteristics of the sound; conduct auditory memory training, such as asking children to repeat a group of words after listening to them; 2. Pronunciation training: Start with oral motor function training (such as pursing lips, grinning, and sticking out tongue) to help children enhance their oral muscle control ability, and then practice pronunciation of single sounds, syllables, words, and sentences. Targeted correction training is carried out for pronunciation errors, such as substitution errors, omission errors, and distortion errors; 3. Language comprehension training: From simple vocabulary comprehension (recognizing common objects and names of people), to sentence comprehension (following instructions and answering simple questions), to paragraph comprehension (listening to stories and answering related questions), gradually improve children's language comprehension ability; 4. Language expression training: from imitating speech, actively expressing needs (such as "I want to drink water"), to describing things, narrating events and having conversations, to improve the accuracy, fluency and richness of children's language expression; 5. Cognitive and social skills training: In line with the laws of cognitive development, children's cognitive scope is expanded through games and activities, such as recognizing colors, shapes, and numbers. At the same time, social scenes are created to allow children to interact with their peers or family members to improve their social skills and communication abilities. The speech training achievement evaluation is applicable to when the subject is a speech training teacher, and the speech training hearing aid plays an evaluation role. , Daily Speech Assessment Coefficient and professional speech assessment coefficient Provide speech training teachers with effectiveness evaluation criteria to assist them in completing speech training courses.

[0033] The embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. The embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit (CPU), the above-mentioned functions defined in the method of the present application are executed. It should be noted that the above-mentioned computer-readable medium of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of an electrical, magnetic, optical, electromagnetic, infrared segment, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more wire segments, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, electrical wire, optical cable, RF, etc., or any suitable combination of the foregoing.

[0034] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0035] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments. Without departing from the principles, the implementation methods of the present invention may be deformed or modified in any way.

Claims

1. A speech training hearing aid system equipped with an AI speech abnormality assessment algorithm, characterized in that: include: The hearing level assessment module is used to collect composite signals and formulate an age-adaptive rule library to conduct a comprehensive hearing assessment on hearing-impaired children, obtain the corresponding hearing threshold discreteness and signal-to-noise ratio attenuation, and then obtain the hearing level assessment coefficient of the corresponding hearing-impaired children; The daily speech assessment module is used to collect speech feature data and develop a standard speech comparison library, and then conduct daily speech assessment on hearing-impaired children. It also uses a convolutional neural network learning model to build a daily speech abnormality assessment model, and then obtain the daily speech assessment coefficient of the corresponding hearing-impaired children. Professional speech assessment module, used to collect multi-dimensional data, conduct professional speech assessment on hearing-impaired children, and then obtain the professional speech assessment coefficient of the corresponding hearing-impaired children; The speech level analysis module is used to obtain the hearing level assessment coefficient, the daily speech assessment coefficient and the professional speech assessment coefficient according to the above modules, and analyze them to obtain the analysis results, and formulate a comprehensive analysis plan for the corresponding hearing-impaired children based on the analysis results.

2. A speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to claim 1, characterized in that: The process of collecting composite signals and developing an age-appropriate rule base to conduct a comprehensive hearing assessment for hearing-impaired children includes: An embedded sound field generating device is provided to generate a composite signal, wherein the composite signal includes a pure tone signal and a speech noise signal; the pure tone signal is set with different frequency points; the speech noise signal generates a standard composite noise; Setting a speech signal acquisition device to collect the test sound signals of the hearing-impaired children in real time; Formulate an age adaptation rule base, wherein the age adaptation rule base includes age intervals and age correction compensation factors; According to the pure tone signal, speech noise signal and test sound signal, the corresponding hearing threshold dispersion and signal-to-noise ratio attenuation are obtained.

3. A speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to claim 2, characterized in that: According to the corresponding hearing threshold dispersion and signal-to-noise ratio attenuation, the process of obtaining the hearing level assessment coefficient of the corresponding hearing-impaired child includes: The test sound signal of the hearing-impaired child is collected in real time according to the different frequency points of the pure tone signal and the speech signal collection device to obtain the hearing threshold dispersion of adjacent frequency points; According to the standard composite noise and the dynamic range of -20dB to +10dB, the environment is divided into a quiet environment, a noisy environment, and a reverberant environment, thereby obtaining the corresponding signal-to-noise ratio attenuation; According to the threshold dispersion, the signal-to-noise ratio attenuation and the age correction compensation factor, a hearing level assessment coefficient corresponding to the hearing-impaired child is obtained.

4. The speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to claim 3, characterized in that: The process of collecting speech feature data and developing a standard speech comparison library for daily speech assessment of hearing-impaired children includes: Formulate a standard speech comparison library to obtain standard speech data, wherein the standard speech data includes standard acoustic feature data and standard linguistic feature data; A high-precision microphone device is provided to extract speech features of the hearing-impaired child to obtain speech feature data; the speech feature extraction includes acoustic feature extraction and linguistic feature extraction, and the speech feature data includes acoustic feature data and linguistic feature data; Acquiring acoustic feature data according to the acoustic feature extraction; According to the linguistic feature extraction, linguistic feature data is obtained.

5. The speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to claim 4, characterized in that: The process of building a daily speech anomaly assessment model using a convolutional neural network learning model includes: Obtaining several groups of acoustic feature data and linguistic feature data of different age groups in the historical collection period; Obtaining standard acoustic feature data and standard linguistic feature data of normal children of corresponding age groups in the standard speech comparison library; According to the several groups of acoustic feature data, linguistic feature data of different age groups in the historical collection period, standard acoustic feature data and standard linguistic feature data of the corresponding age groups, a training sample set is formed; Build a standard evaluation model based on the convolutional neural network learning model; The training sample set is input into the standard evaluation model, the standard evaluation model is trained, and the trained standard evaluation model is obtained, and then a daily speech abnormality evaluation model is constructed.

6. The speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to claim 5, characterized in that: According to the daily speech abnormality assessment model, a daily speech assessment coefficient is obtained.

7. The speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to claim 6, characterized in that: The process of collecting multi-dimensional data and conducting professional speech assessment for hearing-impaired children includes: Conduct multi-dimensional quantitative assessments of hearing-impaired children based on standardized language assessment tools, including: quantitative assessments of language comprehension, language expression, vocabulary, grammar use, and speech clarity; A multi-dimensional data synchronous acquisition device is set up to obtain multi-dimensional data corresponding to quantitative evaluation, including: instruction execution accuracy, reaction time standard deviation, average sentence length, dependency distance, proportion of age-appropriate words, retelling syntactic tree editing distance and phoneme comparison accuracy.

8. The speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to claim 7, characterized in that: The process of obtaining professional speech assessment coefficients for hearing-impaired children includes: According to the instruction execution accuracy, reaction time standard deviation, average sentence length, dependency distance, age-appropriate word ratio, repetition syntactic tree edit distance and phoneme comparison accuracy, the professional speech assessment coefficient of the corresponding hearing-impaired children was obtained, which is recorded as , professional speech assessment coefficient for: ； in, Quantitative evaluation dimension weight coefficient for language understanding, The execution accuracy of compound instructions, is the standard deviation of the reaction time, is the mean of the norm for the same age, is the standard deviation of the norm for the same age, Quantitative evaluation dimension weight coefficient for language expression, is the average sentence length, is the maximum compound instruction execution accuracy value corresponding to age, To prevent zero value, Dependence distance, The percentage of age-appropriate words, The weight coefficient of the vocabulary quantification evaluation dimension, is the edit distance of the paraphrase syntax tree, The weight coefficients of the quantitative evaluation dimensions for grammatical application, is the accuracy of phoneme comparison and Weight coefficients for speech intelligibility quantification evaluation dimensions.

9. The speech training hearing aid system equipped with an AI speech abnormality assessment algorithm according to claim 8, characterized in that: The process of analyzing the hearing level assessment coefficient, the daily speech assessment coefficient, and the professional speech assessment coefficient, obtaining the analysis results, and formulating a comprehensive analysis plan for the hearing-impaired child based on the analysis results includes: According to the hearing level assessment coefficient, daily speech assessment coefficient and professional speech assessment coefficient, the analysis results are recorded as , the analysis results for: ； in, Hearing level assessment coefficient The weight coefficient of Evaluation coefficient for daily speech The weight coefficient of Professional speech assessment coefficient The weight coefficient of Setting standard analysis results ; like At this time, hearing-impaired children can go to normal schools; like When developing a comprehensive analysis plan for hearing-impaired children, the comprehensive analysis plan includes an adaptive speech training plan and speech training outcome assessment; The adaptive speech training program is suitable for the subject to be a guardian, to help hearing-impaired children with speech training; The speech training results evaluation is applicable when the subject is a speech training teacher. The speech training hearing aid plays an evaluation role, providing the speech training teacher with an effect evaluation standard, thereby assisting in completing the speech training course.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to implement a speech training hearing aid system equipped with an AI speech abnormality assessment algorithm as described in any one of claims 1 to 9.

Images