34 results about "Speech disturbances" patented technology

Pronunciation issues for synthesized speech are automatically detected using human recordings as a reference within a Speech Recognition Assisted Evaluation (SRAE) framework including a Text-To-Speech flow and a Speech Recognition (SR) flow. A pronunciation issue detector evaluates results obtained at multiple levels of the TTS flow and the SR flow (e.g. phone, word, and signal level) by using the corresponding human recordings as the reference for the synthesized speech, and outputs possible pronunciation issues. A signal level may be used to determine similarities / differences between the recordings and the TTS output. A model level checker may provide results to the pronunciation issue detector to check the similarities of the TTS and the SR phone set including mapping relations. Results from a comparison of the SR output and the recordings may also be evaluation by the pronunciation issue detector. The pronunciation issue detector outputs a list that lists potential pronunciation issue candidates.

The present invention provides a deaf-mute child speech disorder rehabilitation method and system. A method of software analysis is used, a friendly man-machine interface is used to guide and inspire the learning interest of a deaf child, a problem of complex operation is solved, and the deaf child autonomously learns. The economic cost is greatly reduced, and the popularity of the method and the system are greatly enhanced. Assessment levels are divided according to different conditions of patients, and the training is more targeted. The work intensity of speech rehabilitation therapists is reduced, and the naturalness of the pronunciation of hearing impaired people is greatly improved.

The invention discloses a nasal voice measuring instrument which comprises a nasal voice acquisition unit and a filter amplifier unit. The nasal voice acquisition unit is used for acquiring nasal voice signals. The filter amplifier unit connected with the nasal voice acquisition unit is used for amplifying and filtering the nasal voice signals. The nasal voice measuring instrument is capable of filtering noise in the signals to improve precision in signal acquisition and amplifying the signals at the same time, and is capable of detecting weak information in the signals to solve the problem of signal inaccuracy. The nasal voice measuring instrument can be used with other devices such as computers and software to achieve systematic detection and estimation on speech disorders, and has the advantages of high precision and wide application in clinical applications and scientific research. The invention further provides a measuring method of the nasal voice measuring instrument.

The embodiment of the invention provides a pronunciation correction system. The system comprises: a test module configured to output test content determined based on user learning data and collect first test data input by a user according to the test content, the first test data comprising facial image data and voice data; and a pronunciation correction module which is configured to extract user pronunciation characteristics from the first test data, match corresponding pronunciation correction strategies and feed back pronunciation correction content to the user based on the pronunciation correction strategies, and the pronunciation correction content is used for indicating pronunciation difference types and the corresponding pronunciation correction strategies. The targeted pronunciationcorrection strategy is obtained based on the user test data, and the corresponding pronunciation correction content is matched, so that targeted pronunciation correction feedback is realized, the pronunciation problem of the user can be corrected, the pronunciation correction effect is improved, and the learning experience of the user is improved. In addition, the invention further provides a language pronunciation correction method, a medium and computing equipment.

The invention discloses an interactive oral English correction method. The method comprises the following steps: firstly, preprocessing a received voice signal, screening out noise in the voice signal, then carrying out time domain segmentation on a preprocessed effective signal, carrying out language recognition on each time domain segmentation result, and obtaining an input statement according to corresponding consistent recognition results of each group. Therefore, the syntax structure corresponding to the input statement can be obtained by analyzing the abstract statement combination of the input statement, and prompt is carried out by integrating the pronunciation of the user and the deviation of the corresponding word in the training database. According to the method, the grammatical structure and the pronunciation problem of the user can be synchronously pointed out in the process of interacting with the user, the language environment of oral communication can be simulated, and pronunciation and grammatical training are carried out in the oral communication process. Due to the fact that the training correction method has real-time performance, the training learning effect is better, operation is easy, the teaching cost is lower, and popularization and acceptance are easy.

Disclosed is a computer-based aid for teaching language for children with speech disorders. The teaching aid comprises a plurality of categorized, discrete multimedia files, each of which comprises discrete synchronized audio and video content pertaining to a subject matter, and a means for aggregating two or more discrete multimedia files into a single aggregated multimedia file comprising aggregated synchronized audio and video content comprising an aggregation of the individual discrete content of the of the two or more discrete multimedia files. The aggregated content is to be learnt by a learner.

A system for improving dysarthria speech intelligibility and method thereof, are provided. In the system, user only needs to provides a set of paired corpus including a reference corpus and a patient corpus, and a speech disordering module can automatically generate a new corpus completely synchronous with the reference corpus, and the new corpus can be used as a training corpus for training a dysarthria voice conversion model. The present invention does not need to use a conventional corpus alignment technology or a manual manner to perform pre-processing on the training corpus, so that manpower cost and time cost can be reduced, and synchronization of the training corpus can be ensured, thereby improving both training and conversion qualities of the voice conversion model.

The invention relates to the technical field of voice recognition, and discloses a voice recognition system, and the system is provided with an end point detection: analysis of inputted voice, sign language and lip language is carried out, so as to determine that the voice or action comprises the starting and ending processes of the speaking of a user; noise elimination: external noise except voice is removed, noise generated due to factors of a microphone is also included, and the system needs to be subjected to tests of various kinds of noise; and feature extraction: the voice with noise being removed is extracted, or the characteristics of the body movement of the sign language or the mouth-shaped movement of the lip language are extracted. According to the invention, the special sign language action and lip language action endpoint detection equipment is arranged, so that people with speech disorder can perform character communication with people by using lip language or sign language actions, the function of the system is more powerful, good news is brought to people with speech disorder, and people can freely communicate with people.

The invention relates to a cerebrovascular disease treating drug, which is prepared from 5-30 parts by weight of salvia miltiorrhiza, 10-15 parts by weight of ligusticum chuanxiong hort, 9-15 parts by weight of pangolin, 10-15 parts by weight of polygonatum sibiricum, 9-15 parts by weight of atractylodes macrocephala, 10-15 parts by weight of angelica sinensis, 15-30 parts by weight of milkvetch root, 3-9 parts by weight of panax notoginseng, 6-9 parts by weight of rhizoma cimicifugae, 2.5-5 parts by weight of leech, 6-10 parts by weight of earthworm, 10-15 parts by weight of ox horn, 3-6 parts by weight of scorpion and 4.5-10 parts by weight of bombyx batryticatus. According to the present invention, the cerebrovascular disease treating drug has effects of blood circulation activating, blood stasis dissipating, meridian passage dredging, cerebral vessel expanding, blood circulation improving and blood rheology improving, is used for stroke caused by phlegm stagnation and collateral obstruction, and symptoms such as limb numbness, dizziness, speech disorder, twisted mouth and crooked tongue, dull appearance, hemiparalysis and sudden fainting, can further be used for ischemic cerebral embolism, cerebral vascular atherosclerosis and cerebral hemorrhage sequelae, and further has effects of cerebral blood flow recovering, paralysis rehabilitation promoting, and living quality improving.

The invention discloses a system for converting voice into pictures. The system comprises a voice recognition module, a recognition analysis main module, a big data cloud storage library, an independent processing storage library, a picture preprocessing module, an autonomous learning module, a background management module, a display touch terminal and a client, the display touch terminal comprises a single voice picture, a continuous voice picture, a standard voice picture and a non-professional voice picture which are independently divided into four levels of operable areas, and corresponding areas are selected for actual operation according to requirements of a user. The system can help the user to better understand the voice content in the form of animation and pictures, can present the meaning of helping the user to understand the voice through voice recognition, achieves the determination and thinking association of the voice meaning, and can be better suitable for early education and learning of infants, The system helps nursing personnel to understand the expression and demand of patients with speech disorder, improves and promotes the understanding and mastering of variousteaching and teaching meanings, and is applied to the understanding of professional contents in various professional fields.

The invention relates to the related technical field of medical rehabilitation, in particular to a tongue root trainer, which comprises a spoon handle and a tongue root acting body connected with the spoon handle. The tongue root trainer provided by the invention is suitable for children and adults with speech disorders caused by cerebral palsy, down syndrome and other congenital or acquired problems and articulation problem caused by language development delays and the like in addition to deaf children with strong rote memory ability and touch receptive ability. The tongue root trainer has a wider action range.

The invention discloses a language function rehabilitation training teaching device suitable for speech disorder children. The device comprises a U-shaped frame, the lower surface of the U-shaped frame is fixedly connected with a bracket I, the bottom of the bracket I is fixedly connected with a bottom protection plate, the lower surface of the U-shaped frame is provided with a through hole and is connected with two transmission columns through the through hole in an up-and-down limiting and sliding manner, tongue depressors are fixedly connected to the opposite sides of the two transmission columns, a lifting frame is fixedly connected to the top of each transmission column, and a liquid guiding device is fixedly connected to the lower surface of the lifting frame and the opposite face of the inner wall of the U-shaped frame. Through cooperative use of the structures, the invention solves the problems in the actual use process that: traditional oral muscle training devices for children are single in use effect, systematic training of tongue muscles, lip muscles and palaoropharyngeal muscles of the oral cavity of the children is difficult to achieve, the language function rehabilitation training effect of the children is poor, and inconvenience is brought to use.

The invention discloses a universal directional voice confrontation sample generation method and system, a medium and equipment, and the method comprises the steps: designing a target optimization loss function to achieve the versatility of voice disturbance, minimizing the confidence coefficient of classification as an original correct category, maximizing the confidence coefficient of classification as a target category, and improving the classification accuracy. Meanwhile, the decibel difference between voice disturbance and original voice is introduced into a loss function, the disturbance size is limited, and the lp norm of the voice disturbance is restrained within a specified spherical range; the disturbance is covered by using the daily noise and psychoacoustic principle, and the influence of sound propagation in the air is introduced when the voice disturbance is generated, so that the generated general voice disturbance is still applicable to the physical world; the universal directional speech adversarial sample is generated after the universal speech disturbance is added to any original speech command data, the speech data is wrongly recognized as a specified target category by a speech command classifier based on the convolutional neural network, and the method has great significance for the research on the robustness of the deep neural network.

A method and device for enabling remote speech disorder therapy are presented. The method includes setting a first device with at least one exercise to be performed during a current therapy session, wherein each exercise includes at least a difficulty parameter; receiving a voice production of a user of the first device; processing the received voice production to evaluate a correct execution of the voice production respective of the at least one difficulty parameter; generating a feedback based on the analysis; and outputting the generated feedback to the first device.

The invention discloses a Xiaoshuan Zaizao pill. The pill is prepared from the following raw materials: dragon blood, red peony root, myrrh, Chinese angelica, achyranthes root, red sage root, ligusticum wallichii, cassia twig, pseudo-ginseng, fructus amomi rotundus, radix curcumae, bitter orange, bighead atractylodes rhizome, ginseng, aquilaria wood, little multibanded krait, muscardine silkworm, rhizoma typhon, gastrodia elata, radix stephaniae tetrandrae, pawpaw, scorpion, radix clematidis, radix astragali, oriental waterplantain rhizome, poria cocos, eucommia, sophora bud, tuber of dwarf lilyturf, Chinese magnoliavine, rhizoma drynariae, rosin, haw, cassia bark, borneol, styrax, benzoin, cinnabar, a flavoring agent, and a fresh-keeping agent. The pill can be used to treat sequelae of stroke caused by qi deficiency, blood stagnation, and collateral obstruction due to wind phlegm, such as hemiplegia, facial paralysis, speech disorder, chest distress, lipid increase, and the like.

The invention provides a language function rehabilitation training teaching device suitable for speech impaired children, and relates to the technical field of language function training for speech impaired children, the language function rehabilitation training teaching device comprises a base, the cross section of the base is of a triangular structure, the top end of the base is fixedly connected with a guide rail, a sliding groove is formed in the guide rail, and a motor A is installed on one side of the guide rail; a transmission shaft is arranged on one side of the motor A, penetrates through the guide rail and protrudes out of the guide rail, one side of the transmission shaft is further connected with a screw A, and the motor A and the screw A jointly form a transmission structure. The problem that when a patient is massaged, the training effect becomes poor due to fatigue easily is solved, when a moving block moves, a massage block can be driven to rotate through meshing between a gear and a rack which are installed on the inner side of the moving block, and then the purpose of massaging the cheek of the patient in real time is achieved.

An embodiment of the present invention provides an adaptive correction method for a user's personal pronunciation dictionary model. The method includes: establishing an initial personal pronunciation dictionary model, receiving a user's correction of an erroneous word in the model recognition result, and comparing the confidence in the acoustic score of the erroneous word with the preset confidence when the correction is a matching replacement word. Determine whether there is a pronunciation problem; determine whether the pronunciation relationship between the wrong word and the replacement word belongs to the regular fuzzy sound rule, compare the number of historical corrections, and if it belongs, adjust the pronunciation weight of the same pronunciation as the wrong pronunciation in batches. If it does not belong, adjust it separately Mispronunciation weights, based on the adjusted weights, adaptively correct the individual pronunciation dictionary model. The embodiment of the present invention also provides a system for self-adaptive modification of the user's personal pronunciation dictionary model. The embodiment of the present invention effectively handles recognition errors caused by inaccurate pronunciations of users, and batch corrections can be performed on a class of regular wrong pronunciations.

The invention discloses a language function rehabilitation training teaching device suitable for children with speech disorder, and the device comprises a sealed wall body, and one side of the inner wall of the sealed wall body is provided with anti-falling cotton. According to the device, children can walk on the walking mechanism, the braking structure is arranged, the situation that the children fall down after running can be effectively avoided, walking movement is aerobic movement, the respiratory rate of the children can be adjusted, the children can actively blow air in the game through the respiration promoting mechanism, and the children can play the game conveniently. A wind power detector arranged on the back can measure the speed of air blown out by the child, so that the effect of detecting the air volume is achieved, help is provided for later treatment, meanwhile, a reward mechanism is adopted for rewarding the child on food, and a first motor drives a rotary disc to rotate; the children can naturally take the elastic connecting sleeve on the mounting frame, and the second motor can be started to drive the elastic connecting sleeve to rotate.

The invention discloses a system for lalopathy stress intervention on the basis of real-time rhythm multi-dimensional modeling. The system comprises a collecting device, a calculating device, a judging device and an intervention device, wherein the collecting device collects rhythm data; the calculating device obtains parameter values by calculating the rhythm data; the judging device enables the parameter values to be compared with a standard numerical range of the rhythm data, and if the parameter values are within the standard numerical range, lalopathy does not occur; if the parameter values are lower than or higher than the standard numerical range, the lalopathy occurs and a lalopathy type is judged; and the intervention device performs stress intervention according to the lalopathy type. The system has multi-dimensional characteristic parameters reflecting rhythm lalopathy, can reflect the characteristics of the rhythm lalopathy more comprehensively, enables assessment on lalopathy patients to be more accurate and enables rehabilitation training of the lalopathy patients to be more comprehensive and targeted. The invention further discloses a method for lalopathy stress intervention on the basis of real-time rhythm multi-dimensional modeling.

The invention discloses a virtual reality-based cancer-related cognitive disorder assessment and rehabilitation training system which comprises an assessment system and a rehabilitation system, the assessment system is used for assessing three common damaged dimensions of cancer-related cognitive disorders and comprises a visual and auditory assessment module, a speech fluency assessment module and a digital connection assessment module; the rehabilitation system comprises a single-dimensional rehabilitation training module and a multi-dimensional rehabilitation training module. The assessment system further comprises an assessment system for performing mild dementia screening on an old-age cancer patient and a speech disorder risk evaluation system for a brain tumor patient. The rehabilitation system further comprises a VR immersive reminiscence experience system for carrying out corresponding rehabilitation training on mild dementia patients and a VR speech rehabilitation training system for carrying out corresponding rehabilitation training on speech disorder patients. The multifunctional integrated VR cognitive impairment assessment and rehabilitation training device can highly simulate the daily living environment, and enables patients with cancer-related cognitive impairment to obtain comprehensive cognitive rehabilitation.

The invention discloses an aeration sound interesting trainer for children, belonging to the field of medical instruments, including a base disposed at the bottom, a strut vertically arranged on an upper end face of the base; a rotating device including a fixed housing, a rotating housing, and a rotating electric machine; an air flow detecting device comprising a housing, an air collector, an airflow sensor, a control processing module and a communication module; and a multimedia display which is directly above the air flow detection device through two support rods. The media display is connected with a control processing module through a communication module. A display surface of the multimedia display is oriented in the same direction as the projection direction of the air collector. The invention solves the technical problems of occupying more doctors' resources, difficult detection of training effect and dryness when training the aspirated pronunciation of speech disorder patients.

The invention discloses an anti-Shroom2 autoantibody for detecting autoimmune system diseases as well as a product and application of the anti-Shroom2 autoantibody. The kit prepared by the invention can realize detection of the Shroom2 autoantibody, and can be used for diagnosing nervous system diseases, especially nervous system diseases related to one or more symptoms; the symptoms include cerebellar degeneration, mental behavior disorder, cognitive impairment, visual impairment, memory deterioration, speech disorder, dyskinesia, level of consciousness decline, autonomic nerve dysfunction or hearing disorder. The kit prepared by the application can also be used for distinguishing autoimmune diseases, especially autoimmune diseases and non-autoimmune diseases of a nervous system.

The invention discloses a multi-dimensional voice disorder measuring system based on the multi-dimensional real-time voice modeling technology. The system comprises a collecting device, a calculating device and a judging device, wherein the collecting device collets phoneme data, the calculating device calculates parameter values of the phoneme data, the standard numerical range of the phoneme data is stored in the judging device, the parameter values are compared according to the standard numerical range, voice disorder does not exist if the parameter values are within the standard numerical range, the voice disorder exists if the parameter values are lower than or higher than the standard numerical range, and the type of the existed voice disorder is judged. Real-time detection processing and real-time voice multi-dimensional modeling are carried out on voice and rhinolalia, so segmental phoneme data, supersegmental phoneme data, mixed segmental phoneme data and separted segmental phoneme are evaluated, numerous pointed multi-dimensional characteristic parameters are increased, and therefore evaluation of voice disorder is more comprehensive and precise. The invention further discloses a multi-dimensional voice disorder measuring method based on the multi-dimensional real-time voice modeling technology.

A drug for treatment of cerebrovascular disease is characterized by comprising 5-30 parts by weight of Salvia miltiorrhiza, 10-15 parts by weight of Ligusticum wallichii, 9-15 parts by weight of Manis pentadactyla, 10-15 parts by weight of rhizoma polygonati, 9-15 parts by weight of white atractylodes rhizome, 10-15 parts by weight of Chinese angelica, 15-30 parts by weight of Astragalus membranaceus, 3-9 parts by weight of pseudo-ginseng, 6-9 parts by weight of rhizoma cimicifugae, 2.5-5 parts by weight of leech, 6-10 parts by weight of earthworm, 10-15 parts by weight of ox horn, 3-6 parts by weight of scorpion and 4.5-10 parts by weight of stiff silkworm. The drug has the effects of activating blood circulation to dissipate stasis, clearing and activating the channels and collaterals, expanding blood vessels of brain, and improving blood circulation and blood rheology, and can be used for strokes caused by sputum stasis and collateral obstruction, symptoms are as follows: numbness of limbs, dizziness, speech disorder, skewed mouth and tongue, sluggish look, hemiplegia and sudden faint, also can treat ischemic cerebral embolism, cerebral vascular atherosclerosis and sequela of cerebral hemorrhage, can restore cerebral blood flows, promotes the rehabilitation of paralysis, and improves quality of living and life.

The angel syndrome is a hereditary neurological disorder, and is characterized by development retardation, intellectual disability, serious speech disorder and exercise and balance problems. Provided herein are polynucleotides, vectors, polypeptides, cells, compositions, kits and methods for treating angel syndrome.

A method that includes novel algorithms to biomedical analysis, specifically including the treatment of stuttering, rhotacism, and stutter pauses. Additionally, it provides systematic methods to collect data and treat the patients given the initial data. All the analysis and data collection are done through the convenience of a voice assistant, like Amazon Alexa or Google Home, or through a phone or web application. The claims presented in this disclosure secures the entire combined system and its individual components.

The invention discloses an intelligent diagnosis method, system and readable storage medium for idler faults. The method for intelligent diagnosis of idler faults includes the following steps: S1, collecting audio data of idlers; S2, extracting features of the audio data; the features of the audio data specifically include one or any of sharpness, noise annoyance, and speech interference level Combination; S3, the characteristics of the audio data are input into the trained intelligent classification recognition model, and the intelligent classification recognition model recognizes the running status of the idler; S4, if the idler is running abnormally, perform alarm, monitoring or control operations; The roller is running normally, complete the fault diagnosis of the idler at the current moment, and execute step S5; S5, update the time, repeat steps S1 to S4, and carry out the fault diagnosis of the idler at the next moment. The invention can realize the real-time diagnosis of idler faults, and is easy to implement with low cost and low algorithm complexity.