An active five-tone speech therapy system

Abstract

This invention discloses an active five-tone speech therapy system, primarily targeting individuals with speech dysfunction after stroke. Its core active five-tone speech therapy has been proven in clinical trials to effectively improve spontaneous speech, auditory comprehension, repetition, and naming functions in patients with subacute and chronic aphasia following stroke. Building upon this therapy, this invention adds a module for collecting and differentiating the patient's five internal organ syndrome elements, creating a novel, online, therapist-free remote diagnosis and treatment software that integrates "patient symptom and sign information collection – five internal organ syndrome element assessment – ​​five-tone repertoire recommendation – active five-tone speech therapy – speech collection during training, accuracy and pronunciation standard assessment, training difficulty adjustment, scale evaluation, and training efficacy feedback." This invention can improve the applicability of Western classical melodic intonation therapy to Chinese aphasia patients, mobilize patients' subjective initiative, provide precise five-tone music recommendations and individualized treatment for patients with different syndrome types, enhance the rehabilitation effect and efficiency of existing speech training, reduce the workload of therapists, alleviate the medical and economic burden on patients, and lay the foundation for the development of subsequent active five-tone speech therapy remote diagnosis and treatment equipment.

Description

Technical Field

[0001] This invention relates to the field of speech rehabilitation diagnosis and treatment, and more specifically, to an active five-tone speech therapy system that applies the five-tone theory to post-stroke aphasia. Background Technology

[0002] Aphasia is an acquired language dysfunction syndrome caused by organic brain damage to the language center and related language networks in the cerebral hemispheres. Patients exhibit impaired or lost ability to receive (understand) and use (express) language symbols while remaining conscious. Approximately 80% of aphasia cases result from cerebrovascular accidents causing damage to the neural networks in the left cerebral hemisphere cortex and subcortical regions. Post-stroke aphasia (PSA) is reported to account for about one-third of stroke patients. Because some stroke patients suffer from long-term speech impairment and significantly reduced social skills, their quality of life declines significantly, and this can even lead to an increased mortality rate among stroke patients with PSA.

[0003] Existing speech therapy systems, such as "An Adult Hearing and Speech Rehabilitation System" (patent application number CN201310553325.8), include software that collects patient single-word pronunciation assessments and training data, randomly arranges and shuffles the order, re-evaluates and compares the results with individual words, and focuses on training the patient's repetition function. Another example is "A Speech Recognition Method Specifically for Rehabilitation Training of Aphasia Patients" (patent application number CN202010242125.0). This software uses speech features and a method of training a model by combining data from healthy individuals and patients, which can effectively improve the speech recognition rate of aphasia patients. The constructed model can be applied to rehabilitation training, speech assessment, and assisted communication for aphasia patients. The patent application number is CN201811052361.5. The software and system acquire standard information for aphasia evaluation, including system oral and / or system written standards; acquire aphasia representation information obtained by a target user imitating the system oral and / or system written standards, including oral and / or written information; and compare the standard information and the aphasia representation information to obtain the target user's aphasia evaluation form. The patent application number is CN202211078843.4. This device combines resisted breathing with vocal training. In this method, medical staff select a song and adjust the lyrics. Simultaneously, the patient can see in real-time how many words of the lyrics correspond to their exhalation volume. Only when the patient can exhale continuously for 6 seconds after maximum inhalation can the lyrics be automatically and completely played by the monitoring computer. In this method, the patient can control the song playback through breathing training.

[0004] Existing speech therapy systems have the following drawbacks:

[0005] (1) Most current aphasia rehabilitation software and systems are boring and tedious for patients, making it difficult to get their cooperation, motivate them, and have no significant therapeutic effect.

[0006] (2) Most software or systems fail to integrate diagnostic and treatment functions, with diagnosis and treatment belonging to different software or systems, making the patient's diagnosis and treatment process cumbersome;

[0007] (3) Existing patents do not have TCM syndrome differentiation, do not combine TCM state, and lack personalized and targeted treatment, which greatly affects the effectiveness of their efficacy.

[0008] (4) Some software and systems only serve as a demonstration of vocal organs and muscle movement, and cannot guide caregivers (such as patients' family members) to implement the correct intervention methods. It is difficult to provide long-term assistance and guidance to caregivers to complete the pronunciation correction of patients with behavioral impairment and aphasia.

[0009] (5) Some software and systems have limited application scenarios and rely solely on speech therapy therapists, resulting in slow recovery of speech pronunciation and communication skills for aphasic patients;

[0010] (6) Even if a therapeutic device that combines breathing and vocalization is developed, it still relies heavily on medical staff. Patients cannot intuitively feel whether their exhalation meets the requirements when they make a sound. As a result, they cannot correct their pronunciation errors well. Furthermore, the device cannot adjust the difficulty in real time according to the patient's training progress, resulting in poor therapeutic effects. Summary of the Invention

[0011] This invention proposes an active five-tone speech therapy system for the language function rehabilitation of patients with aphasia after stroke. The system includes remote diagnosis and treatment software with training modes to overcome the problems existing in existing patents.

[0012] The diagnostic and treatment software proposed in this invention includes: "collection of patient symptoms and signs information, judgment of the five internal organs syndrome elements, recommendation of five-tone repertoire, active five-tone speech therapy, collection of speech during training, judgment of accuracy and pronunciation standard, adjustment of training difficulty stages, scale evaluation and feedback on training efficacy."

[0013] The remote diagnosis and treatment software proposed in this invention includes a module for collecting and differentiating the syndrome elements of the five internal organs. This module collects information on the patient's symptoms from multiple sources, including hospitals and the patient's family. Based on the diagnostic criteria of Professor Zhu Wenfeng's "Syndrome Element Differentiation," and according to the weight of the four diagnostic methods in diagnosis, a weighted threshold method is used to determine the syndrome elements of the five internal organs. Then, based on the direct correspondence between the five internal organs and the five musical notes, a suitable melody of the five musical notes is selected for the patient.

[0014] In the remote diagnosis and treatment software proposed in this invention, the recommended pentatonic repertoire includes: each pentatonic melody comprising ① a short audio recording improved and adjusted by this invention; and ② a pentatonic repertoire categorized by this invention. All training stages are conducted according to the corresponding audio, the pentatonic melodies in the repertoire, rhythms, etc.

[0015] The remote diagnosis and treatment software proposed in this invention includes an active five-tone speech therapy method: the melody uses the aforementioned five-tone repertoire, while the rhythm and tone adopt the inherent rhythm and original tone of Mandarin Chinese. The training mode is divided into three stages. The first two stages use phrases with fewer words and higher frequency of daily use; the last stage introduces longer and more complex phrases. In each stage, a candidate must achieve a training score of ≥90% of the total score for five consecutive target phrases to proceed to the next stage.

[0016] An active five-tone speech therapy system includes a mobile terminal. The mobile terminal comprises a patient symptom and sign information collection module, a five-organ syndrome element judgment module, a five-tone repertoire recommendation module, and an active five-tone speech therapy module. The patient symptom and sign information collection module is used to input the patient's symptom and sign information on the mobile terminal. The five-organ syndrome element judgment module is used to determine the syndrome elements of the five organs and, based on the determined direct correspondence between the five organs and the five tones, determine suitable five-tone melodies for the patient. The five-tone repertoire recommendation module recommends suitable melodies based on the judgment results of the five-organ syndrome element judgment module. The patient's "five-tone" melody; all training stages are conducted according to the "five-tone" melody and rhythm; the active five-tone speech therapy module includes entering the training mode using the "five-tone" melody, the melody using the five-tone repertoire, and the rhythm and tone adopting the inherent rhythm and original tone of Chinese; the training mode is divided into three stages. In the first two stages, phrases with fewer words and high frequency of daily use are used; in the last stage, longer and more complex phrases are introduced; in each stage, the training score of 5 consecutive target phrases is ≥90% of the total score of that stage before proceeding to the next stage of training.

[0017] Preferably, the training mode is as follows:

[0018] The first stage of the training mode is as follows: ① Chanting demonstration: The therapist hums twice, which is not scored; ② Singing together: The therapist and patient sing together according to the selected "five-tone" melody audio, accompanied by tapping with the left hand. Success within 4 attempts scores 1 point; otherwise, it scores 0 points, and the next phrase is introduced, starting from ① again; ③ Gradual withdrawal: When the singing is halfway through, the therapist withdraws from chanting, and the patient chants the remaining part alone, accompanied by tapping with the left hand. Success scores 1 point; otherwise, it scores 0 points, and the next phrase is introduced, starting from ① again; ④ Immediate chanting repetition: When the therapist raises their left hand, the subject listens to the demonstration, and then the therapist immediately lowers their left hand, and the patient chants and repeats the phrase alone, accompanied by tapping with the left hand. Success scores 1 point; otherwise, it scores 0 points, and the next phrase is introduced, starting from ① again; ⑤ Immediate chanting response: The therapist immediately chants the question, and the patient chants the target phrase alone without tapping with the left hand. However, when the patient attempts to answer, the therapist may tap the target phrase with their left hand. Success scores 1 point, and a new phrase is introduced, continuing from ① again.

[0019] The second stage of the training mode: ① Chanting demonstration: The therapist performs two humming demonstrations, which are not scored; ② Gradual withdrawal: The therapist and patient sing together according to the selected "pentatonic" melody audio, accompanied by tapping with the left hand. However, halfway through the singing, the therapist withdraws from chanting, and the patient chants the remaining part alone, accompanied by tapping with the left hand. Even if successful on the first try, it must be repeated twice. Success within four attempts earns 1 point; otherwise, 0 points are awarded, and the next phrase is used to start from ① again; ③ Delayed chanting repetition: When the therapist raises their left hand, the patient listens to the demonstration. After a delay of approximately 6 seconds, the therapist lowers their left hand, and the subject chants the phrase independently, accompanied by tapping with their left hand. Success earns 2 points; failure results in returning to step ② and continuing. If successful this time, only 1 point is awarded, and failure again earns 0 points. The subject then starts again from step ① with the next phrase. ④ Delayed chanting response: After a successful response in the previous step, the subject chants the question again after a 6-second delay. The patient chants the target phrase independently without tapping with their left hand. Success earns 2 points; failure results in returning to step ③ and continuing. If successful this time, only 1 point is awarded, and failure again earns 0 points. The subject then starts again from step ① with the next phrase.

[0020] The third stage of the training mode: ① Delayed chanting repetition: When the therapist raises their left hand, the patient listens to the demonstration. After a delay of about 6 seconds, the therapist lowers their left hand, and the patient chants and repeats the phrase independently, accompanied by tapping with their left hand. Success earns 2 points; failure requires practicing chanting together with the therapist withdrawing midway, and the remaining part is completed independently by the patient. Then, the patient tries again to chant and repeat the phrase independently with the delay. If successful this time, only 1 point is awarded; if unsuccessful, 0 points are awarded, and the next phrase is used to start from ① again; ② Normal spoken demonstration: The therapist slowly pronounces the target phrase twice with exaggerated rhythm and emphasis, accompanied by tapping with their left hand. No points are awarded; ③ Gradual withdrawal: The therapist chants with the patient, and the therapist withdraws midway, and the remaining part is completed independently by the patient. Success earns 2 points; failure requires the therapist to accompany the patient in chanting the complete phrase without withdrawing midway. If successful this time, only 1 point is awarded; if it fails again, 0 points are awarded, and the patient must start from ① again with the next phrase. ④ Delayed Normal Spoken Repetition: When the therapist raises their left hand, the patient listens to the demonstration, i.e., the therapist presents the target phrase in normal language without tapping with their left hand. After a delay of about 6 seconds, the therapist lowers their left hand, and the patient repeats the phrase in normal language on their own. If successful, 2 points are awarded; if it fails, the patient returns to ③ and continues. If successful this time, only 1 point is awarded; if it fails again, 0 points are awarded, and the patient must start from ① again with the next phrase. ⑤ Delayed Normal Spoken Response: After a successful response in the previous step, the patient is asked a question in normal language after a 6-second delay. The patient must respond using the target phrase with normal pronunciation. If successful, 2 points are awarded; if it fails, the patient returns to ④ and continues. If successful this time, only 1 point is awarded; if it fails again, 0 points are awarded, and the patient must start from ① again with the next phrase.

[0021] Preferably, the patient symptom and sign information collection module, the five-organ syndrome element judgment module, the five-tone music recommendation module, and the active five-tone speech therapy module are integrated into a remote diagnosis and treatment software. The therapist in the active five-tone speech therapy module is a speech therapy rehabilitation therapist who conducts remote video calls, or a virtual therapist formed by the diagnosis and treatment software and can be displayed on a mobile terminal in real time. The actions and speech of the virtual therapist are operated by the software in advance.

[0022] Preferably, the mobile terminal includes a microphone, a speaker, a screen, and a camera. The microphone is used to collect voice and audio information generated during training, the speaker is used to play voice and audio, the screen is used to display the therapist, and the camera is used to acquire image information of the patient. The mobile terminal (1) also includes a storage device for recording voice, audio, and video information generated by the user during training and testing.

[0023] Preferably, when the therapist is a virtually formed therapist, the active five-tone speech therapy module includes a module for judging accuracy and pronunciation standard, a module for adjusting training difficulty stages, a scale assessment module, and a training efficacy feedback module.

[0024] The accuracy and pronunciation standard judgment module is used to: establish a pre-stored standard corpus, i.e., the first speech, which includes the five-tone music library used in the "five-tone" speech therapy, as well as standard answers for chanting repetition, chanting response, normal spoken repetition, and normal spoken response; collect user speech audio during training and testing, i.e., the second speech, to construct a user corpus; compare and analyze the speech audio collected during training and testing based on the standard corpus to determine the accuracy of the user's second speech, and use the accuracy as the first standard information; on this basis, extract speech features such as pitch, intensity, amplitude, and fundamental frequency from the first and second speech respectively, encode the above speech features to obtain feature vectors, and construct a pronunciation model step to further compare and analyze the pronunciation standard of the second speech in terms of pitch, speech rate, and clarity, and use the pronunciation standard as the second standard information; generate a first analysis report based on the first and second standard information for the next stage of training difficulty adjustment;

[0025] The training difficulty stage adjustment module is used to: calculate scores according to the scoring principles of the three stages of the "Active Five-Tone Speech Therapy Module" training mode based on the accuracy and pronunciation standard in the first analysis report, summarize the user's training completion rate in the current difficulty stage, so as to adjust the training content in a timely manner and determine whether the user can enter the next difficulty stage. The specific judgment criteria are: the training score of 5 consecutive target phrases in each stage is ≥ 90% of the total score of that stage, in order to enter the next stage of training.

[0026] The scale assessment module is used to: provide a separate scale assessment module containing: various clinically commonly used aphasia assessment scales and daily communication ability assessment scales. The clinically commonly used aphasia assessment scales include the Western Aphasia Test Suite, the Chinese Aphasia Examination, the Chinese Standard Aphasia Examination Scale from the China Rehabilitation Research Center, and the Boston Diagnostic Aphasia Examination. The daily communication ability assessment scales include the Daily Living Communication Ability Assessment and the Chinese Functional Language Communication Ability Test. Users conduct functional assessments using the scale assessment module before initial training and after phased training, respectively, to establish a pre-stored standard corpus, i.e., third speech, which contains standard answers to the aforementioned scales. Data is collected during the scale assessment process. The user's voice audio, i.e., the fourth language, is used to construct a user corpus. Based on the standard corpus, the voice audio collected during the scale assessment process is compared and analyzed to determine the accuracy rate of the user's fourth language, and the accuracy rate is used as the third standard information. On this basis, speech features such as pitch, intensity, amplitude, and fundamental frequency are extracted from the third and fourth languages ​​respectively. These speech features are encoded to obtain feature vectors, and a pronunciation model is constructed. The pronunciation standard of the fourth language, such as pitch, speech rate, and clarity, is further compared and analyzed, and the pronunciation standard is used as the fourth standard information. Based on the scale scoring rules, the scale score is determined according to the third and fourth standard information, and a second analysis report is generated.

[0027] The training efficacy feedback module is used to send the aforementioned second analysis report to a mobile terminal for display after the scale assessment is completed.

[0028] Preferably, when no speech audio is detected during the scale assessment process or the collected speech audio does not match the pre-stored standard corpus at all, the software will automatically play the pronunciation of the first Chinese character in the standard answer to the question to guide the user to imitate the correct fourth pronunciation; if no speech audio is detected or the collected speech audio still does not match the pre-stored standard corpus, the complete standard answer to the question will be played automatically.

[0029] Preferably, it also includes a breathing and vocalization training terminal, which is wirelessly or wiredly connected to the mobile terminal, and the breathing and vocalization training terminal can perform in-depth breathing training in addition to speech training.

[0030] Further, the breathing and vocalization training terminal includes: a fixed outer frame, an inner rotating cylinder assembly, a curve first processing module, and a straw. The inner rotating cylinder assembly is rotatably disposed within the fixed outer frame, and the curve first processing module is disposed on one side of the fixed outer frame. The inner rotating cylinder assembly includes a rotating cylinder and a lower sealing cap assembly. The rotating cylinder includes an inner cylinder and an outer cylinder. The upper ends of the inner and outer cylinders are sealed with upper sealing plates, and the lower ends are provided with openings. Several sealing plates are provided between the outer cylinder and the inner cylinder. The space enclosed by two adjacent sealing plates, the inner cylinder, the upper sealing plate, and the outer cylinder forms several training air chambers. A piston is slidably disposed in each training air chamber. A vent valve is provided at the upper end of the upper sealing plate corresponding to each training air chamber. The cap assembly includes an inner rotating cap and an outer sealing ring. The inner rotating cap can be embedded in the outer sealing ring and rotate relative to the outer sealing ring. The outer sealing ring is fixedly connected to the outer cylinder. The inner rotating cap is fixedly disposed at the lower end of the fixed outer frame. A rotation drive mechanism capable of driving the inner rotating cylinder assembly to rotate is disposed within the fixed outer frame. The first straw is inserted into the inner rotating cap from the lower end of the fixed outer frame, and its end contacts the training air chamber. The first curve processing module includes a writing component and an erasing component. The writing component can draw the required breathing volume corresponding to the patient's singing during training as lines on the outside of the rotating cylinder when the inner rotating cylinder assembly rotates, forming a standard breathing training curve. The outer cylinder is made of transparent material.

[0031] Further, the fixed outer frame includes an upper limit ring, a lower limit ring seat, and a left support plate and a right support plate connected between the upper limit ring and the lower limit ring seat. The first curve processing module is installed on the left support plate. The left support plate is provided with a slide groove 1 and a slide groove 2. The writing component 1 includes a slider 2, a telescopic rod 2, and a writing pen 1. A lead screw 2 is rotatably installed in the slide groove 2. The lead screw 2 is driven to rotate by a drive motor 2. The slider 2 is provided with an internal thread. The slider 2 is inserted into the lead screw 2 and slides up and down driven by the lead screw 2. One end of the telescopic rod 2 is fixed to the right side of the slider 2. The other end of the telescopic rod is fixedly equipped with a writing pen. The telescopic rod can move the tip of the writing pen away from or towards the rotating cylinder. The erasing assembly includes a slider, a telescopic rod, and an erasing block. A lead screw is rotatably installed in the slide groove. The lead screw is driven to rotate by a drive motor. The slider has an internal thread and is inserted into the lead screw, which drives it to slide up and down. One end of the telescopic rod is fixed to the right side of the slider, and the other end of the telescopic rod is fixedly equipped with an erasing block. The telescopic rod can move the erasing block away from or towards the rotating cylinder.

[0032] Furthermore, a piston adjustment power assembly is also provided inside the inner cylinder. The piston adjustment power assembly includes a power mounting plate fixedly installed inside the inner cylinder, a lead screw five rotatably installed on the power mounting plate, and a guide rod fixedly installed on the power mounting plate. The upper end of the lead screw five is rotatably installed inside the upper sealing plate. The guide rod is fixedly installed inside the upper sealing plate. A piston adsorption slider is passed through the lead screw five and the guide rod. The rotation of the lead screw five can drive the piston adsorption slider to slide up and down. A first magnet is provided on the side of the piston that contacts the inner cylinder, and a second magnet is provided on the side of the piston adsorption slider that contacts the inner cylinder. The first magnet and the second magnet attract each other.

[0033] The beneficial effects of the present invention are:

[0034] (1) This invention develops a diagnostic and treatment software that performs online diagnosis of the five internal organs and selects the corresponding five-tone mode to remotely conduct active five-tone speech training for patients with aphasia after stroke. Through online diagnosis of the five internal organs, individualized treatment is achieved, which improves the treatment effect and initiative of patients. Through personalized diagnosis and music selection, melodies and rhythms are incorporated into speech therapy. The five-tone music library contains various types of music that patients are familiar with, which can easily mobilize the enthusiasm of patients and improve the therapeutic effect.

[0035] (2) It has achieved the scientific effectiveness of computer assessment and treatment, reduced human resource costs, improved scientific reliability, and can be promoted and applied on a large scale, saving social resources. Through personalized TCM syndrome differentiation, it can accurately identify and evaluate the patient's syndrome type and implement speech rehabilitation in a targeted and personalized manner. Through the assessment of language function scale, it can further clarify the patient's aphasia subtype in order to match the appropriate speech training difficulty.

[0036] (3) This invention is a new technology applied to post-stroke aphasia. By differentiating the syndromes of the five internal organs in traditional Chinese medicine, the reliability and validity are more accurate and reliable, saving manpower costs and being easy to use. At the same time, it further reduces the economic pressure on patients' families. The active five-tone speech training also improves patients' negative emotions from another perspective, and enhances patients' quality of life and self-confidence.

[0037] (4) This invention helps patients undergo gradual training through voice prompts and image guidance, and simultaneously guides caregivers (such as patients' family members) to implement correct interventions. It can intuitively assist and guide caregivers to complete pronunciation correction for patients with behavioral impairment and aphasia.

[0038] (5) This invention integrates diagnosis and treatment functions into one, with assessment and treatment belonging to the same software, simplifying the diagnosis and treatment process for patients; it expands the application scenarios of the software, allowing patients to use the remote diagnosis and treatment software through platforms such as mobile phones, tablets, and computers, thereby achieving remote rehabilitation, rather than relying solely on speech therapy therapists, and improving the effectiveness of language function recovery for aphasia patients.

[0039] (6) In addition, the present invention also provides a breathing and vocal training terminal for breathing training and vocal training. The training terminal can automatically form a breathing training curve according to the length of breathing training desired by the patient. According to the different exhalation intensity required by the patient when singing different pitches, it can intuitively reflect the patient's breathing intensity when singing and compare it with the standard breathing intensity, so that the patient can automatically adjust the breathing intensity when singing. This not only trains vocalization but also trains the breathing intensity when vocalizing, so that the patient's breath is stable when vocalizing, thereby further improving the effect of vocal training.

[0040] (7) The breathing and vocalization training terminal of the present invention calculates different breathing training curves based on the size of the exhaled breath according to the rhythm of the melody being sung. Then, the patient can adjust according to the training curve based on their own exhaled breath, so that the breath can be stable above the standard in each beat. Breath stability training can also be carried out during vocalization training. If the patient's breath has reached the specified stability requirements after multiple training sessions, the patient can also actively lengthen the duration of the breath in each beat, so as to carry out more adaptive training.

[0041] (8) The breathing and vocalization training terminal of the present invention can be used by two patients. Each patient can perform breathing training without interfering with the other, which can significantly improve training efficiency. At the same time, the generated breathing training curve can be generated by computer software and then drawn on the outer wall of the exhalation rotating cylinder by the curve generation pen, thereby automatically forming the training curve. The degree of automation is high. Moreover, a pen and an eraser are respectively set at both ends of the exhalation rotating cylinder. The pen and eraser are set adjacent to each other, so that different training curves can be provided to two different patients at the same time.

[0042] (9) This application realizes an active guidance teaching function for patients' exhalation by setting up a piston adjustment power component. The software converts the exhalation intensity and speed required to produce a certain tone into the rotation speed of the drive motor. The rotation speed of the drive motor can guide the speed and degree of piston rise. Since the piston is magnetically connected, the piston can also have a certain rise and fall error when guided, making the exhalation guidance for patients flexible and not causing discomfort to patients during exhalation. Since the patient's exhalation is actively guided by the drive motor, he can better remember the exhalation degree and volume required to produce this tone, which can make the breathing training effect better. It is no longer the teacher or computer passively telling the patient how to operate when exhaling. The patient can feel it immediately, thereby increasing the active guidance teaching function. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the modules included in the mobile terminal of the present invention;

[0044] Figure 2 This is a schematic diagram of the modules included in the active five-tone speech therapy module of the present invention;

[0045] Figure 3 This is a schematic diagram of the treatment system structure of the present invention;

[0046] Figure 4 A schematic diagram of the first embodiment of the breathing and vocalization training terminal;

[0047] Figure 5 This is a schematic diagram of the first embodiment with a fixed outer frame;

[0048] Figure 6 This is a top view with a fixed outer frame;

[0049] Figure 7 This is a schematic diagram of the structure of the first embodiment of the inner rotating cylinder assembly;

[0050] Figure 8 An exploded view of a first embodiment of the inner rotating cylinder assembly;

[0051] Figure 9 A schematic diagram of the structure after half of the rotating cylinder has been cut off;

[0052] Figure 10 The front view and sectional view of the inner rotating cylinder assembly;

[0053] Figure 11 for Figure 10 BB sectional view and enlarged view C;

[0054] Figure 12 A schematic diagram of the piston and piston-adsorption slider structure;

[0055] Figure 13 This is a schematic diagram of the outer sealing ring structure;

[0056] Figure 14 This is a schematic diagram of the internal rotating cover structure;

[0057] Figure 15 This is a schematic diagram of a straw structure;

[0058] Figure 16 A schematic diagram of the second embodiment of the breathing and vocalization training terminal;

[0059] Figure 17 This is a schematic diagram of the structure of the second embodiment of the inner rotating cover.

[0060] Label Explanation

[0061] 1. Mobile terminal; 2. Breathing and vocalization training terminal; 3. Fixed outer frame; 4. Inner rotating cylinder assembly; 5. First curve processing module; 6. Second curve processing module; 7. Straw 1; 8. Upper limit ring; 9. Lower limit ring seat; 10. Left support plate; 11. Right support plate; 12. Writing component 1; 13. Erasing component 1; 14. Erasing component 2; 15. Writing component 2; 16. Rotating cylinder; 17. Lower cover assembly; 8. Inner rotating cover; 19. Outer sealing ring; 20. Inner cylinder; 21. Outer cylinder; 22. Sealing plate; 23. Upper sealing plate; 24. Training air chamber; 25. Air release valve; 26. Piston adjustment power assembly; 27. Piston suction slider; 28. Piston; 29. ​​Slide groove one; 30. Slide groove two; 31. Lead screw one; 32. Lead screw two; 33. Slider one; 34. Slider two; 35. Telescopic rod one; 36. Telescopic rod two; 37. Cleaning 38. Block 1; 39. Writing Pen 1; 40. Slide 3; 41. Slide 4; 42. Lead Screw 3; 43. Lead Screw 4; 44. Slider 3; 45. Slider 4; 46. Telescopic Rod 3; 47. Telescopic Rod 4; 48. Writing Pen 2; 49. Clearing Block 2; 50. Through Hole 1; 51. Power Mounting Plate; 52. Lead Screw 5; 53. Guide Rod; 54. Drive Motor 1; 55. Fan-shaped Piston Plate; 56. Piston Inner Flange; 57. Lead screw hole; 58. Guide slide hole; 59. Lower ring plate; 60. Outer ring wall; 61. First recess; 62. First inner hole; 63. Rotating circular plate; 64. Limiting inner ring; 65. Through hole two; 66. Second recess; 67. Upper edge plate; 68. Lower edge plate; 69. Horizontal pipe section; 70. Vertical pipe section; 71. Suction tube two; 72. Through hole three; 73. Through hole four; 74. Upper rotary bearing; 75. Lower rotary bearing. Detailed Implementation

[0062] The present invention will be further described below with reference to embodiments, but this is not intended to limit the present invention in any way. Any modifications or substitutions made based on the teachings of the present invention shall fall within the protection scope of the present invention.

[0063] An active five-tone speech therapy system, such as Figure 1-3As shown, it includes a mobile terminal 1, which includes a patient symptom and sign information collection module, a five-organ syndrome element judgment module, a five-tone music recommendation module, and an active five-tone speech therapy module. The patient symptom and sign information collection module is used to input the patient's symptom and sign information on the mobile terminal 1. The symptom information can be collected from various sources, such as hospitals or the patient's family. The five-organ syndrome element judgment module determines the syndrome elements of the five organs based on the dialectical standards of Professor Zhu Wenfeng's "Syndrome Element Diagnosis" and the weighted threshold method according to the weight of the four diagnostic methods in diagnosis. Based on the direct correspondence between the five organs and the five tones, it determines the suitable "five-tone" melody for the patient. The five-tone music recommendation module recommends suitable music based on the judgment results of the five-organ syndrome element judgment module. The patient's "five-tone" melody, each "five-tone" melody includes ①: a short audio file modified and adjusted by this invention; ②: a five-tone music library categorized by this invention; all training stages are conducted according to the corresponding audio file, the "five-tone" melody in the music library, rhythm, etc.; the active five-tone speech therapy module includes entering the training mode using the "five-tone" melody, specifically, the melody uses the five-tone music, and the rhythm and tone adopt the inherent rhythm and original tone of Chinese; the training mode is divided into three stages. In the first two stages, phrases with fewer words and high daily usage frequency are used; in the last stage, longer and more complex phrases are introduced; in each stage, the training score of 5 consecutive target phrases is ≥90% of the total score of that stage before proceeding to the next stage of training.

[0064] Specifically, the first stage of the training model is as follows: ① Chanting demonstration: The therapist performs two humming demonstrations (not scored); ② Singing together: The therapist and patient sing together according to the selected "five-tone" melody audio, accompanied by tapping with the left hand (1 point for success within 4 attempts; otherwise, 0 points, and start again from ① with the next phrase); ③ Gradual withdrawal: When the singing is halfway through, the therapist withdraws from chanting, and the patient chants the remaining part alone, accompanied by tapping with the left hand (1 point for success; otherwise, 0 points, and start again from ① with the next phrase); ④ Immediate chanting repetition: When the therapist raises their left hand, the subject listens to the demonstration, and then the therapist immediately lowers their left hand, and the patient chants and repeats the phrase alone, accompanied by tapping with the left hand (1 point for success; otherwise, 0 points, and start again from ① with the next phrase); ⑤ Immediate chanting response: The therapist immediately chants the question, and the patient chants the target phrase alone without tapping with the left hand, but when the patient attempts to answer, the therapist may tap the target phrase with the left hand (1 point for success, and a new phrase is introduced to continue from ①).

[0065] The second stage of the training model: ① Chanting demonstration: The therapist performs two humming demonstrations (not scored); ② Gradual withdrawal: The therapist and patient sing together according to the selected "pentatonic" melody audio, accompanied by tapping with the left hand. However, halfway through the singing, the therapist withdraws from chanting, and the patient chants the remaining part alone, accompanied by tapping with the left hand. Even if successful on the first try, it must be repeated twice (1 point for success within 4 attempts; otherwise, 0 points, and start from ① again with the next phrase); ③ Delayed chanting repetition: The patient listens to the demonstration when the therapist raises their left hand. After a delay of about 6 seconds, the therapist lowers their left hand, and the subject sings the recitation alone, accompanied by tapping with their left hand (success, 2 points; failure, return to step ② and continue, this time if successful, only 1 point, failure again, 0 points, and start from step ① again with the next phrase); ④ Delayed singing response: after the previous step is successful, delay for 6 seconds and sing the question again, the patient sings the target phrase alone, without tapping with their left hand (success, 2 points; failure, return to step ③ and continue, this time if successful, only 1 point, failure again, 0 points, and start from step ① again with the next phrase).

[0066] The third stage of the training model: ① Delayed chanting repetition: When the therapist raises their left hand, the patient listens to the demonstration. After a delay of about 6 seconds, the therapist lowers their left hand, and the patient chants and repeats the phrase independently, accompanied by tapping with their left hand (success, 2 points; failure, then practice chanting together while the therapist withdraws midway, and the patient completes the remaining part independently, then try again for the patient to chant and repeat the phrase independently. If successful this time, only 1 point is awarded; if unsuccessful again, 0 points are awarded, and the next phrase is used to start from ① again); ② Normal spoken demonstration: The therapist slowly pronounces the target phrase twice with exaggerated rhythm and emphasis, accompanied by tapping with their left hand (no points are awarded); ③ Gradual withdrawal: The therapist chants with the patient while the therapist withdraws midway, and the patient completes the remaining part independently (success, 2 points; failure, then the therapist tries to chant the complete phrase with the patient without withdrawing midway). If successful this time, only 1 point is awarded; if it fails again, 0 points are awarded, and the patient must start from ① again with the next phrase. ④ Delayed normal verbal repetition: When the therapist raises their left hand, the patient listens to the demonstration, that is, the therapist presents the target phrase in normal language without tapping with their left hand. After a delay of about 6 seconds, the therapist lowers their left hand, and the patient repeats the phrase in normal language on their own (success is awarded 2 points; failure means returning to ③ and continuing; if successful this time, only 1 point is awarded; if it fails again, 0 points are awarded, and the patient must start from ① again with the next phrase). ⑤ Delayed normal verbal response: After a successful step in the previous step, a question is asked in normal language after a 6-second delay. The patient must answer using the target phrase with normal pronunciation (success is awarded 2 points; failure means returning to ④ and continuing; if successful this time, only 1 point is awarded; if it fails again, 0 points are awarded, and the patient must start from ① again with the next phrase).

[0067] Specifically, the patient symptom and sign information collection module, the five-organ syndrome element judgment module, the five-tone music recommendation module, and the active five-tone speech therapy module are integrated into a remote diagnosis and treatment software. The therapist in the active five-tone speech therapy module can be a speech therapy rehabilitation therapist who conducts remote video calls, or a virtual therapist formed by the diagnosis and treatment software and displayed in real time on the mobile terminal 1. The actions and speech of the virtual therapist can be operated by the software in advance.

[0068] Furthermore, the mobile terminal 1 includes a microphone, a speaker, a screen, and a camera. The microphone is used to collect voice and audio information generated during the training process, the speaker is used to play voice and audio, the screen is used to display the therapist, and the camera is used to acquire image information of the patient. The mobile terminal 1 also includes a storage device for recording voice, audio, and video information generated by the user during training and testing.

[0069] Specifically, when the therapist is a virtual therapist, the active five-tone speech therapy module includes a module for judging accuracy and pronunciation standard, a module for adjusting training difficulty stages, a scale assessment module, and a training efficacy feedback module.

[0070] The accuracy and pronunciation standardization judgment module is used to: establish a pre-stored standard corpus (first speech), which includes the five-tone music library used in the aforementioned "five-tone" speech therapy, as well as standard answers for chanting repetition, chanting response, normal spoken repetition, and normal spoken response. During training and testing, user speech audio (second language) is collected to construct a user corpus. Based on the standard corpus (first speech), the speech audio (second language) collected during training and testing is compared and analyzed to determine the accuracy of the user's second speech, and the accuracy is used as the first standard information. On this basis, by extracting the pitch, intensity, amplitude, fundamental frequency, and other speech features of the first and second speech respectively, encoding these speech features to obtain feature vectors, and constructing a pronunciation model, the module further compares and analyzes the pronunciation standardization of the second speech, including pitch, speech rate, and clarity, and uses the pronunciation standardization as the second standard information. A first analysis report is generated based on the first and second standard information and used for adjusting the training difficulty in the next stage.

[0071] The training difficulty adjustment module is used to: calculate scores according to the scoring principles of the three stages of the "Active Five-Tone Speech Therapy Module" training mode based on the accuracy and pronunciation standard in the first analysis report, summarize the user's training completion rate in the current difficulty stage, so as to adjust the training content in a timely manner and determine whether the user can enter the next difficulty stage (the user can enter the next stage of training only if the training score of 5 consecutive target phrases in each stage is ≥ 90% of the total score of that stage).

[0072] The scale assessment module provides a separate scale assessment module containing various commonly used clinical aphasia assessment scales, including the Western Aphasia Battery (WAB), the Aphasia Battery of Chinese (ABC), the Chinese Rehabilitation Research Center Standard Aphasia Examination (CRRCCAAE), and the Boston Diagnostic Aphasia Examination (BDAE), as well as assessment scales for daily communication skills such as the Communication Activities of Daily Living (CADL) and the Chinese Functional Communication Profile (CFCP). Users conduct functional assessments using this scale assessment module before initial training and after subsequent training phases. A pre-stored standard corpus (third language) containing standard answers to the aforementioned scales is established; during the scale assessment process, user audio recordings (fourth language) are collected to construct a user corpus. Based on a standard corpus (third speech), the audio recordings (fourth language) collected during the scale assessment process are compared and analyzed to determine the accuracy rate of the user's fourth speech, which is then used as the third standard information. Building upon this, speech features such as pitch, intensity, amplitude, and fundamental frequency are extracted from both the third and fourth speech, and these features are encoded to obtain feature vectors. Furthermore, a pronunciation model is constructed to compare and analyze the pronunciation standard of the fourth speech, including pitch, speech rate, and clarity, which is then used as the fourth standard information. Based on the scale's scoring rules and the third and fourth standard information, the scale score is determined, generating a second analysis report.

[0073] The training efficacy feedback module is used to send the aforementioned second analysis report to mobile terminal 1 for display after the scale assessment is completed. It not only displays the current second analysis report but also allows users to retrieve previous analysis reports through the history log for comparison of improvements after each training phase.

[0074] Furthermore, if no speech audio (fourth language) is detected during the scale assessment process, or if the collected speech audio (fourth language) does not match the pre-stored standard corpus (third language), the software will automatically play the pronunciation of the first Chinese character in the standard answer to the question (third language) to guide the user to imitate the correct fourth language pronunciation; if no speech audio (fourth language) is detected or if the collected speech audio (fourth language) still does not match the pre-stored standard corpus (third language), the software will automatically play the complete standard answer to the question (third language).

[0075] Furthermore, it also includes a breathing and vocalization training terminal 2, which is wirelessly or wiredly connected to the mobile terminal 1. The breathing and vocalization training terminal 2 can perform in-depth breathing training in addition to speech training. Compared with existing breathing training, it can ensure that breathing training is smoother and more stable when pronouncing words.

[0076] Furthermore, the mobile terminal 1 is a mobile phone, tablet, or computer.

[0077] Specifically, such as Figure 4-15As shown, the breathing and vocalization training terminal 2 of the first embodiment includes: a fixed outer frame 3, an inner rotating cylinder assembly 4, a curve first processing module 5, and a straw 7. The inner rotating cylinder assembly 4 is rotatably disposed within the fixed outer frame 3, and the curve first processing module 5 is disposed on one side of the fixed outer frame 3. The inner rotating cylinder assembly 4 includes a rotating cylinder 16 and a lower sealing cap assembly 17. The rotating cylinder 16 includes an inner cylinder 20 and an outer cylinder 21. The upper ends of the inner cylinder 20 and the outer cylinder 21 are closed with an upper sealing plate 23, and the lower ends are provided with openings. A plurality of sealing partitions 22 are disposed between the outer cylinder 21 and the inner cylinder 20. Two adjacent sealing partitions 22, the inner cylinder 20, and the inner cylinder 20 are connected. The space enclosed by the upper sealing plate 23 and the outer cylinder 21 forms several training air chambers 24. A piston 28 is slidably arranged in each training air chamber 24. A vent valve 25 is provided on the upper end of the upper sealing plate 23 corresponding to each training air chamber 24. The lower sealing cover assembly 17 includes an inner rotating cover 18 and an outer sealing ring 19. The inner rotating cover 18 can be embedded in the outer sealing ring 19 and rotate relative to the outer sealing ring 19. The outer sealing ring 19 is fixedly connected to the outer cylinder 21. The inner rotating cover 18 is fixedly arranged at the lower end of the fixed outer frame 3. A rotary drive mechanism (not shown in the figure) is provided in the fixed outer frame 3 to drive the inner rotating cylinder assembly 4 to rotate. As illustrated, a rotary drive motor can be installed on the fixed outer frame 3, with a drive gear on the rotary output shaft of the rotary motor, and a gear ring on the outer periphery of the inner rotating cylinder assembly 4. The rotation of the inner rotating cylinder assembly 4 is achieved through the meshing of the drive gear and the gear ring. The straw 7 is inserted into the inner rotating cover 18 from the lower end of the fixed outer frame 3, with its end contacting the training air chamber 24. The curve first processing module 5 includes a writing component 12 and an erasing component 13. The writing component 12 can draw the required breathing volume corresponding to the patient's vocalization during training as lines on the rotating cylinder 18 when the inner rotating cylinder assembly 4 rotates. On the outside of 6, a standard breathing training curve is formed. The outer cylinder 21 is made of transparent material, so the height position of the piston 28 inside the outer cylinder 21 can be seen through the outer cylinder 21. During breathing training, the patient exhales into the straw 7 while pronouncing the vowel "u". At this time, the piston 28 can slide up and down with the size of the exhalation. When the pronunciation is standard and the exhalation is appropriate, the piston 28 is always on the standard breathing training curve. If the exhalation is too low or too high, the piston 28 will deviate from the standard breathing training curve. The patient can intuitively feel whether the exhalation is appropriate through the curve. This is more intuitive than the breathing training method of the prior art. The erasing component 13 can erase the old training curve after the inner rotating cylinder component 4 rotates once, while the writing component 12 can draw the new training curve on the outer cylinder 21.

[0078] like Figure 5As shown, the fixed outer frame 3 includes an upper limit ring 8, a lower limit ring seat 9, and a left support plate 10 and a right support plate 11 connected between the upper limit ring 8 and the lower limit ring seat 9. The curve first processing module 5 is installed on the left support plate 10. The left support plate 10 is provided with a first slide groove 29 and a second slide groove 30. The writing component 12 includes a second slider 34, a second telescopic rod 36, and a first writing pen 38. A second lead screw 32 is rotatably arranged in the second slide groove 30. The second lead screw 32 is driven to rotate by a second drive motor (not shown in the figure). The second slider 34 is provided with an internal thread. The second slider 34 is inserted into the second lead screw 32 and is driven by the second lead screw 32 to slide up and down. One end of the second telescopic rod 36 is fixed to the right side of the second slider 34. The other end of the telescopic rod 36 is fixedly provided with a writing pen 38. The telescopic rod 36 can drive the tip of the writing pen 38 away from or towards the rotating cylinder 16. The erasing assembly 13 includes a slider 33, a telescopic rod 35, and an erasing block 37. A lead screw 31 is rotatably provided in the slide groove 29. The lead screw 31 is driven to rotate by a drive motor 3 (not shown in the figure). The slider 33 is provided with an internal thread. The slider 33 is inserted into the lead screw 31 and is driven by the lead screw 31 to slide up and down. One end of the telescopic rod 35 is fixed to the right side of the slider 33. The other end of the telescopic rod 35 is fixedly provided with an erasing block 37. The telescopic rod 35 can drive the erasing block 37 away from or towards the rotating cylinder 16.

[0079] like Figure 9 , 12 As shown, a piston adjustment power assembly 26 is also provided inside the inner cylinder 20. The piston adjustment power assembly 26 includes a power mounting plate 50 fixedly installed inside the inner cylinder 20, a lead screw 51 rotatably installed on the power mounting plate 50, and a guide rod 52 fixedly installed on the power mounting plate 50. The upper end of the lead screw 51 is rotatably installed inside the upper sealing plate 23, and the guide rod 52 is fixedly installed inside the upper sealing plate 23. A piston suction slider 27 passes through the lead screw 51 and the guide rod 52. The piston adjustment power assembly 26 can be driven by the rotation of the lead screw 51. The piston suction slider 27 slides up and down. A first magnet is provided on the side of the piston 28 that contacts the inner cylinder 20, and a second magnet is provided on the side of the piston suction slider 27 that contacts the inner cylinder. The first magnet and the second magnet attract each other. Thus, the position of the piston 28 can be adjusted by sliding the piston suction slider 27. The ease of raising and lowering the piston 28 can also be adjusted by adjusting the screw 51. This allows for the adjustment of the amount of air that the patient needs to blow into the straw according to the actual situation, and intelligent adjustment of the breathing intensity based on the patient's condition.

[0080] Additionally, a drive motor 53 is fixedly installed below the power mounting plate 50. The power output end of the drive motor 53 is fixedly connected to the lower end of the lead screw 51. Rotation of the drive motor 53 drives the lead screw 51 to rotate. A guide hole 57 and a lead screw hole 56 are provided inside the piston-adsorption slider 27. The lead screw 51 passes through the lead screw hole 56 and engages with the thread inside the lead screw hole 56. The guide rod 52 passes through the guide hole 57. The first magnet can be a permanent magnet or an electromagnet, and the corresponding second magnet can be an electromagnet or a permanent magnet. When the piston position needs to be adjusted using the piston-adsorption slider 27, the electromagnet is energized and becomes magnetic. At this time, it can attract the permanent magnet and drive the piston to adjust its position. After use, the piston needs to be adjusted to the lowest position. The piston-adsorbent slider 27 can also actively control the patient's exhalation, providing active guidance for the patient's exhalation. Specifically, the exhalation intensity and speed required to produce a certain tone are converted into the rotational speed of the drive motor 53. The rotational speed of the drive motor 53 guides the speed and extent of the piston's rise. Since the piston is magnetically connected, it can also have a certain degree of rising and falling error when guided, making the exhalation guidance for the patient flexible and preventing discomfort. Because the patient's exhalation is actively guided by the drive motor 53, it can better remember the required exhalation intensity and volume for producing that tone, thus improving the breathing training effect. It is no longer a matter of the teacher or computer passively telling the patient how to exhale; the patient can feel it immediately, thereby enhancing the active guidance teaching function.

[0081] In addition, the vent valve 25 can be a valve structure whose opening is controlled by a solenoid valve. By adjusting its opening size, the amount of exhaled air required to enter the training air chamber 24 when the piston rises and falls to different heights can be controlled.

[0082] Furthermore, in order to improve the attraction force of the first magnet and the second magnet, the piston 28 includes a fan-shaped piston plate 54 and an inner piston flange 55. The thickness of the inner piston flange 55 is greater than that of the fan-shaped piston plate 54 and the same as that of the piston attraction slider 27. The first magnet is disposed on the inner piston flange 55.

[0083] The specific function of the piston adjustment power assembly 26 is as follows: In

[0084] Furthermore, the number of the partitions 22 can be selected as 2, 4, 6 or even more, preferably 4, such as... Figure 8 As shown.

[0085] like Figure 10-11As shown in Figures 13-14, the outer sealing ring 19 includes a lower ring plate 58 and an outer ring wall 59. A first inner hole 61 is formed on the inner side of the lower ring plate 58, and the outer ring wall 59 is provided on the outer side of the lower ring plate 58. A first recess 60 is provided at the upper end of the lower ring plate 58 near the edge of the first inner hole 61.

[0086] The inner rotating cover 18 includes a rotating circular plate 62 and a limiting inner ring 63. The limiting inner ring 63 is disposed at the upper end of the rotating circular plate 62, and the center of the limiting inner ring 63 is concentric with the center of the rotating circular plate 62. A second recess 65 is provided on the outer periphery of the rotating circular plate 62, thereby forming an upper edge plate 66 and a lower edge plate 67 on the outer edge of the rotating circular plate 62. The rotating circular plate 62 is rotatably disposed in the first inner hole 61. The upper edge plate 66 is embedded in the first recess 60. The upper end of the upper edge plate 66 is flush with the uppermost end of the lower ring plate 58, and the lower end of the lower edge plate 67 is lower than the lowermost end of the lower ring plate 58. In this way, the airtightness can be well guaranteed when the inner rotating cover 18 rotates relative to the outer sealing plate 19.

[0087] Furthermore, an internal thread is provided on the inner side of the outer ring wall 59, and an external thread is provided at the corresponding position at the lower end of the outer cylinder 21. The tightening or loosening of the outer sealing ring 19 and the outer cylinder 21 is achieved through the internal thread and the external thread.

[0088] When the outer sealing ring 19 is tightened to the lower end of the outer cylinder 21, the limiting inner ring 63 is inserted into the inner side of the inner cylinder 20, and the lower end of the sealing plate 22 is in close contact with the upper surface of the rotating circular plate 62.

[0089] A through hole 49 is provided at the lower end of the limiting ring seat 9, and a through hole 64 is provided on the rotating circular plate 62. When the rotating circular plate 62 is fixed to the limiting ring seat 9, the through hole 49 and the through hole 64 overlap each other. After the straw 7 passes through the through hole 49 and the through hole 64, the upper side of the end of the straw 7 is flush with the upper surface of the rotating circular plate 62.

[0090] When the rotating drive mechanism drives the inner rotating cylinder assembly 4 to rotate, the length of each syllable during the patient's breathing training can be controlled by changing the driving speed of the rotating drive mechanism. For example, the time it takes for the adjacent septa 22 of each training air chamber 24 to pass through the straw can be considered as one syllable. In order to give the patient an intuitive feeling of the syllable, the thickness of the septa 22 is greater than or equal to the diameter of the straw 7. In this way, when the septa 22 rotates to the position of the straw 7, the end of the straw 7 can be completely blocked. Thus, the patient can clearly feel that the blowing is obstructed. At this time, it means that the length of one syllable has ended, and the patient can inhale to carry out the breathing training of the next syllable length, thereby achieving the purpose of autonomous breathing training.

[0091] Furthermore, such as Figure 15 As shown, the straw 7 includes a horizontal tube section 68 and a vertical tube section 69, with the vertical tube section 69 disposed at one end of the horizontal tube section 68.

[0092] Furthermore, such as Figure 5-6 As shown, a second curve processing module 6 is also provided on the other side of the fixed outer frame 3. The second curve processing module 6 is opposite to the first curve processing module 5. The second curve processing module 6 includes a second writing component 15 and a second erasing component 14.

[0093] The second curve processing module 6 is mounted on the right support plate 11. The right support plate 11 is provided with a slide groove 39 and a slide groove 40. The writing component 2 15 includes a slider 3 43, a telescopic rod 3 45, and a writing pen 2 47. A lead screw 3 41 is rotatably disposed in the slide groove 3 39. The lead screw 3 41 is driven to rotate by a drive motor 4 (not shown in the figure). The slider 3 43 is provided with an internal thread. The slider 3 43 is inserted into the lead screw 3 41 and is driven by the lead screw 3 41 to slide up and down. One end of the telescopic rod 3 45 is fixed to the left side of the slider 3 43, and the writing pen 2 47 is fixedly disposed at the other end of the telescopic rod 3 45. The telescopic rod 3 45 can... The pen tip of the writing pen 47 moves away from or closer to the rotating cylinder 16. The erasing assembly 14 includes a slider 44, a telescopic rod 46, and a cleaning block 48. A lead screw 42 is rotatably installed in the groove 40. The lead screw 42 is driven to rotate by a drive motor 5 (not shown in the figure). The slider 44 has an internal thread and is inserted into the lead screw 42, which drives it to slide up and down. One end of the telescopic rod 46 is fixed to the left side of the slider 44, and the other end of the telescopic rod 46 is fixed to the cleaning block 48. The telescopic rod 46 can move the cleaning block 48 away from or closer to the rotating cylinder 16.

[0094] like Figure 6 As shown, the line connecting the writing component 12 and the writing component 2 15 intersects the line connecting the erasing component 13 and the erasing component 2 14 at point O. Point O passes through the line connecting the center of the upper limit ring 8 and the center of the lower limit ring seat 9. The angle between the writing component 12 and the horizontal line is α, and the angle between the writing component 2 15 and the horizontal line is δ, where α = δ. The angle between the erasing component 13 and the horizontal line is β, and the angle between the erasing component 2 14 and the horizontal line is γ, where β = γ. Further, α = β. Further, the value range of α is (0°, 7.5°).

[0095] Along Figure 6The eraser assembly 13, the writing assembly 12, the eraser assembly 2 14, and the writing assembly 2 15 are arranged in a clockwise direction as shown by the arrow. The inner rotating cylinder assembly 4 rotates in the same direction as the arrow, which is clockwise.

[0096] Furthermore, both the writing pen 38 and the writing pen 47 are erasable water-based pens, and both the erasing block 37 and the erasing block 48 are capable of erasing the ink from the water-based pens.

[0097] It also includes an upper rotary bearing 73 and a lower rotary bearing 74. The upper rotary bearing 73 is disposed at the upper end of the upper limit ring 8 and the outer cylinder 21, and the lower rotary bearing 74 is disposed at the lower end of the lower limit ring seat 9 and the outer cylinder 21.

[0098] like Figure 16-17 As shown, the breathing and vocalization training terminal 2 of the second embodiment also includes a second straw 70, a fourth through hole 72 is provided on the lower limiting ring seat 9, and a third through hole 71 is provided on the rotating circular plate 62. The third through hole 71 corresponds to the fourth through hole 72. The second straw 70 passes through the third through hole 71 and the fourth through hole 72, and its end is flush with the upper surface of the rotating circular plate 62.

[0099] Furthermore, the second straw 70 and the first straw 7 have the same structure, and the horizontal tube segments 68 of the first straw 7 and the second straw 70 are located on the same straight line.

[0100] The second embodiment of the breathing and vocalization training terminal 2 can simultaneously provide breathing training for two patients. The cooperation between the first curve processing module 5 and the second curve processing module 6 allows different training curves to be displayed when the outer cylinder faces different patients. Specifically, the first curve processing module 5 reads the first piece of music provided to patient one for breathing training, and the second curve processing module 6 reads the second piece of music provided to patient two for breathing training. Based on the speed of the first and second pieces, the power of the rotation drive mechanism driving the inner rotating cylinder assembly 4 is selected, thereby enabling the selection of the straw. The duration T of the time it takes for the first and / or second straws to pass through a training air chamber 24 is recorded as the duration of one beat of breathing training. Based on the selected beat duration, the writing components 12 and 25 are controlled to write simultaneously. At the same time, the erasing components 24 and 13 erase the curves written by the writing components 12 and 25, respectively, so that the breathing training curves in front of patients 1 and 2 are updated in real time. In addition, when both patients 1 and 2 want to adjust their breathing training, they only need to adjust the rotation speed of the inner rotating cylinder component 4.

[0101] Furthermore, the working method of the breathing and vocalization sequence terminal 2 in the first embodiment is described as follows: When a patient needs to perform breathing training, they can select a suitable rhythmic piece according to their own situation, such as a simple 1 (do), 2 (re), 3 (mi), 4 (fa), 5 (sol), 6 (la), 7 (si). The patient can select the tempo of the above-mentioned piece according to their own situation, such as one note lasting 1-6 seconds. The rotation speed of the inner rotating cylinder assembly 4 can be set accordingly, and the time it takes for the straw to pass through one training air chamber 24 can be set accordingly. The slider 34 indicates the time interval at which the straw passes through one training air chamber 24. After 1-6 seconds, the piston moves to the appropriate height, allowing the curve drawn on the outer tube to be drawn at different heights according to different notes. Simultaneously, the telescopic rod 36 extends, causing the writing pen 38 to contact the outer tube. At the same time, the patient presses their lips to the end of the straw and hums the tune by emitting the vowel "u". Since different pitches require different amounts of exhaled air, the amount of air required for the piston to rise and fall to the standard breathing line position can be pre-adjusted to the exhalation volume required for producing the standard sound. For example, when corresponding to "1 (do)," the exhalation volume required to produce "u" can be pre-adjusted using the vent valve. The opening size of 25 and / or the rotation speed of the drive motor 53 within the piston adjustment power assembly 26 are used to achieve this. The rotation of the drive motor 53 can adjust the up and down movement of the piston. When the patient's breathing is relatively weak in the initial stage, the drive motor 53 can assist in reducing the expiratory volume required when the piston rises or falls. At this time, the main focus is on training the patient's breathing stability. After the inner rotating cylinder assembly 4 rotates one revolution, the erasing assembly 13 is activated, and the telescopic rod 35 extends to contact the outside of the inner rotating cylinder assembly 4, erasing the writing pen 38. Then, the writing pen 38 can continue to draw a new breathing curve. This allows the terminal 2 to continuously perform breathing training. During training, if the patient needs to rest, the rotation of the inner rotating cylinder assembly 4 can be stopped, and the first curve processing module 5 will also stop operating. When it is necessary to adjust the pace of breathing training, the rotation speed of the inner rotating cylinder assembly 4 can be adjusted. Since the patient can see the intensity of their own breathing and has a corresponding breathing curve, the patient's enthusiasm for treatment can be greatly improved, and the training effect on stable breathing is also greatly enhanced. Compared with existing passive breathing training, the patient can better experience the fun of training and enhance the patient's cooperation.

[0102] While the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Any person skilled in the art can make many possible variations and modifications to the technical solutions of the present invention, or modify them into equivalent embodiments, without departing from the scope of the present invention. Therefore, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention, without departing from the scope of the present invention, should fall within the protection scope of the present invention.

Claims

1. An active five-tone speech therapy system, comprising a mobile terminal (1), characterized in that: The mobile terminal (1) includes a patient symptom and sign information collection module, a five-organ syndrome element judgment module, a five-tone music recommendation module, and an active five-tone speech therapy module. The patient symptom and sign information collection module is used to input the patient's symptom and sign information on the mobile terminal (1). The five-organ syndrome element judgment module is used to determine the syndrome elements of the five organs and, according to the determined direct correspondence between the five organs and the five tones, determine the suitable "five-tone" melody for the patient. The five-tone music recommendation module recommends the suitable "five-tone" melody for the patient based on the judgment result of the five-organ syndrome element judgment module. All training stages are carried out according to the "five-tone" melody and rhythm. The active five-tone speech therapy module includes using the "five-tone" melody to enter the training mode, using the five-tone music for the melody, and adopting the inherent rhythm and original tone of Chinese for the rhythm and tone. Only when the training score of 5 consecutive target phrases in each stage is ≥90% of the total score of that stage can the next stage of training be entered. The patient symptom and sign information collection module, the five visceral syndrome element judgment module, the five-tone music recommendation module and the active five-tone speech therapy module are integrated into a remote diagnosis and treatment software. The therapist in the active five-tone speech therapy module is a speech therapy rehabilitation therapist who conducts remote video calls, or a virtual therapist formed by the diagnosis and treatment software and can be displayed on the mobile terminal (1) in real time. The actions and speech of the virtual therapist are operated by the software in advance. When the therapist is a virtual therapist, the active five-tone speech therapy module includes a module for judging accuracy and pronunciation standard, a module for adjusting training difficulty stages, a scale assessment module, and a training efficacy feedback module. The accuracy and pronunciation standard judgment module is used to: establish a pre-stored standard corpus, namely the first speech, which includes the five-tone music library used in the above-mentioned "five-tone" speech therapy, as well as the standard answers for chanting repetition, chanting response, normal spoken repetition, and normal spoken response; During training and testing, users' voice audio is collected, i.e., second voice recordings are used to build a user corpus. Based on a standard corpus, the audio recordings collected during training and testing are compared and analyzed to determine the accuracy rate of the user's second speech, which is used as the first standard information. Building upon this, the pitch, intensity, amplitude, and fundamental frequency of the first and second speech are extracted separately, and these features are encoded to obtain feature vectors. A pronunciation model is then constructed, and the pronunciation standard of the second speech (pitch, speed, and clarity) is further compared and analyzed, with the standard pronunciation level used as the second standard information. A first analysis report is generated based on the first and second standard information for adjusting the training difficulty in the next stage. The training difficulty stage adjustment module is used to: calculate scores according to the accuracy and pronunciation standard in the first analysis report, based on the scoring principles of the three stages of the "Active Five-Tone Speech Therapy Module" training mode, summarize the user's training completion rate in the current difficulty stage, so as to adjust the training content in a timely manner and determine whether the user can enter the next difficulty stage. The specific judgment criteria are: the training score of 5 consecutive target phrases in each stage is ≥ 90% of the total score of that stage, in order to enter the next stage of training. The scale assessment module is used to: provide a separate scale assessment module containing: various clinically commonly used aphasia assessment scales and daily communication ability assessment scales. The clinically commonly used aphasia assessment scales include the Western Aphasia Test Suite, the Chinese Aphasia Examination, the Chinese Standard Aphasia Examination Scale from the China Rehabilitation Research Center, and the Boston Diagnostic Aphasia Examination. The daily communication ability assessment scales include the Daily Living Communication Ability Assessment and the Chinese Functional Language Communication Ability Test. Users conduct functional assessments using the scale assessment module before initial training and after phased training, respectively, to establish a pre-stored standard corpus, i.e., third speech, which contains the standard answers for the aforementioned scales. During the scale assessment process... The process involves collecting users' voice audio (i.e., fourth language) to construct a user corpus; comparing and analyzing the voice audio collected during the scale assessment process against a standard corpus to determine the accuracy rate of the user's fourth language, and using the accuracy rate as the third standard information; further, by extracting the pitch, intensity, amplitude, and fundamental frequency of the third and fourth languages ​​respectively, encoding these features to obtain feature vectors, and constructing a pronunciation model, the process further compares and analyzes the pitch, speech rate, and clarity of the fourth language to determine the standard of pronunciation, and using the standard of pronunciation as the fourth standard information; based on the scale scoring rules, and according to the third and fourth standard information, determining the scale score, and generating a second analysis report; The training efficacy feedback module is used to send the above second analysis report to the mobile terminal (1) for display after the scale assessment is completed.

2. The active five-tone speech therapy system as described in claim 1, characterized in that: The specific training modes are as follows: The first stage of the training mode is as follows: ① Chanting demonstration: The therapist hums twice, which is not scored; ② Singing together: The therapist and patient sing together according to the selected "five-tone" melody audio, accompanied by tapping with the left hand. Success within 4 attempts scores 1 point; otherwise, it scores 0 points, and the next phrase is introduced, starting from ① again; ③ Gradual withdrawal: When the singing is halfway through, the therapist withdraws from chanting, and the patient chants the remaining part alone, accompanied by tapping with the left hand. Success scores 1 point; otherwise, it scores 0 points, and the next phrase is introduced, starting from ① again; ④ Immediate chanting repetition: When the therapist raises their left hand, the subject listens to the demonstration, and then the therapist immediately lowers their left hand. The patient chants and repeats the phrase alone, accompanied by tapping with the left hand. Success scores 1 point; otherwise, it scores 0 points, and the next phrase is introduced, starting from ① again; ⑤ Immediate chanting response: The therapist immediately chants the question, and the patient chants the target phrase alone without tapping with the left hand. However, when the patient attempts to answer, the therapist may tap the target phrase with their left hand. Success scores 1 point, and a new phrase is introduced, continuing from ①. The second stage of the training mode: ① Chanting demonstration: The therapist performs two humming demonstrations, which are not scored; ② Gradual withdrawal: The therapist and patient sing together according to the selected "pentatonic" melody audio, accompanied by tapping with the left hand. However, halfway through the singing, the therapist withdraws from chanting, and the patient chants the remaining part alone, accompanied by tapping with the left hand. Even if successful on the first try, it must be repeated twice. Success within four attempts earns 1 point; otherwise, 0 points are awarded, and the next phrase is used to start from ① again; ③ Delayed chanting repetition: When the therapist raises their left hand, the patient listens to the demonstration. After a delay of approximately 6 seconds, the therapist lowers their left hand, and the subject chants the phrase independently, accompanied by tapping with their left hand. Success earns 2 points; failure results in returning to step ② and continuing. If successful this time, only 1 point is awarded, and failure again earns 0 points. The subject then starts again from step ① with the next phrase. ④ Delayed chanting response: After a successful response in the previous step, the subject chants the question again after a 6-second delay. The patient chants the target phrase independently without tapping with their left hand. Success earns 2 points; failure results in returning to step ③ and continuing. If successful this time, only 1 point is awarded, and failure again earns 0 points. The subject then starts again from step ① with the next phrase. The third stage of the training mode: ① Delayed chanting repetition: When the therapist raises their left hand, the patient listens to the demonstration. After a delay of about 6 seconds, the therapist lowers their left hand, and the patient chants and repeats the phrase independently, accompanied by tapping with their left hand. Success earns 2 points; failure requires practicing chanting together with the therapist withdrawing midway, and the remaining part is completed independently by the patient. Then, the patient tries again to chant and repeat the phrase independently with the delay. If successful this time, only 1 point is awarded; if unsuccessful, 0 points are awarded, and the next phrase is used to start from ① again; ② Normal spoken demonstration: The therapist slowly pronounces the target phrase twice with exaggerated rhythm and emphasis, accompanied by tapping with their left hand. No points are awarded; ③ Gradual withdrawal: The therapist chants with the patient, and the therapist withdraws midway, and the remaining part is completed independently by the patient. Success earns 2 points; failure requires the therapist to accompany the patient in chanting the complete phrase without withdrawing midway. If successful this time, only 1 point is awarded; if it fails again, 0 points are awarded, and the patient must start from ① again with the next phrase. ④ Delayed Normal Spoken Repetition: When the therapist raises their left hand, the patient listens to the demonstration, i.e., the therapist presents the target phrase in normal language without tapping with their left hand. After a delay of about 6 seconds, the therapist lowers their left hand, and the patient repeats the phrase in normal language on their own. If successful, 2 points are awarded; if it fails, the patient returns to ③ and continues. If successful this time, only 1 point is awarded; if it fails again, 0 points are awarded, and the patient must start from ① again with the next phrase. ⑤ Delayed Normal Spoken Response: After a successful response in the previous step, the patient is asked a question in normal language after a 6-second delay. The patient must respond using the target phrase with normal pronunciation. If successful, 2 points are awarded; if it fails, the patient returns to ④ and continues. If successful this time, only 1 point is awarded; if it fails again, 0 points are awarded, and the patient must start from ① again with the next phrase.

3. The active five-tone speech therapy system as described in claim 1, characterized in that: The mobile terminal (1) includes a microphone, a speaker, a screen, and a camera. The microphone is used to collect voice and audio information generated during training. The speaker is used to play voice and audio. The screen is used to display the therapist. The camera is used to acquire image information of the patient. The mobile terminal (1) also includes a storage device for recording voice, audio, and video information generated by the user during training and testing.

4. The active five-tone speech therapy system as described in claim 1, characterized in that: If no audio is detected during the scale assessment or the collected audio does not match the pre-stored standard corpus, the software will automatically play the pronunciation of the first Chinese character in the standard answer to guide the user to imitate the correct fourth pronunciation; if no audio is detected or the collected audio still does not match the pre-stored standard corpus, the software will automatically play the complete standard answer to the question.

5. The active five-tone speech therapy system as described in claim 4, characterized in that: It also includes a breathing and vocalization training terminal (2), which is wirelessly or wiredly connected to the mobile terminal (1). The breathing and vocalization training terminal (2) can perform in-depth breathing training in addition to speech training.

6. The active five-tone speech therapy system as described in claim 5, characterized in that: The breathing and vocalization training terminal (2) includes: a fixed outer frame (3), an inner rotating cylinder assembly (4), a curve first processing module (5), and a straw (7). The inner rotating cylinder assembly (4) is rotatably disposed within the fixed outer frame (3), and the curve first processing module (5) is disposed on one side of the fixed outer frame (3). The inner rotating cylinder assembly (4) includes a rotating cylinder (16) and a lower cover assembly (17). The rotating cylinder (16) includes an inner cylinder (20) and an outer cylinder (21). The upper ends of the inner cylinder (20) and the outer cylinder (21) are sealed with upper sealing plates (23), and the lower ends are provided with openings. Several sealing plates (22) are provided between the outer cylinder (21) and the inner cylinder (20). The space enclosed by two adjacent sealing plates (22), the inner cylinder (20), the upper sealing plate (23) and the outer cylinder (21) forms several training air chambers (24). A piston (28) is slidably provided in each training air chamber (24). A piston (28) is provided on the upper end of the upper sealing plate (23) corresponding to each training air chamber (24). The lower sealing cap assembly (17) includes an inner rotating cap (18) and an outer sealing ring (19). The inner rotating cap (18) can be embedded in the outer sealing ring (19) and rotate relative to the outer sealing ring (19). The outer sealing ring (19) is fixedly connected to the outer cylinder (21). The inner rotating cap (18) is fixedly disposed at the lower end of the fixed outer frame (3). A rotation drive mechanism capable of driving the inner rotating cylinder assembly (4) to rotate is provided in the fixed outer frame (3). The straw One (7) is inserted into the inner rotating cover (18) from the lower end of the fixed outer frame (3), and the end contacts the training air chamber (24). The first curve processing module (5) includes a writing component (12) and an erasing component (13). The writing component (12) can draw the required breathing volume corresponding to the patient's singing during training in the form of lines on the outside of the rotating cylinder (16) when the inner rotating cylinder component (4) rotates, forming a standard breathing training curve. The outer cylinder (21) is made of transparent material.

7. The active five-tone speech therapy system as described in claim 6, characterized in that: The fixed outer frame (3) includes an upper limit ring (8), a lower limit ring seat (9), and a left support plate (10) and a right support plate (11) connecting the upper limit ring (8) and the lower limit ring seat (9). The first curve processing module (5) is installed on the left support plate (10). The left support plate (10) is provided with a first slide groove (29) and a second slide groove (30). The first writing component (12) includes a second slider (34), a second telescopic rod (36), and a first writing pen (38). A second lead screw (32) is rotatably provided in the second slide groove (30). The second lead screw (32) is driven to rotate by a second drive motor. The second slider (34) is provided with an internal thread. The second slider (34) is inserted into the second lead screw (32) and is driven to slide up and down by the second lead screw (32). One end of the second telescopic rod (36) is fixed to the right side of the second slider (34). The other end of the second (36) is fixedly provided with a writing pen (38). The telescopic rod second (36) can drive the pen tip of the writing pen (38) away from or close to the rotating cylinder (16). The erasing component first (13) includes a slider first (33), a telescopic rod first (35) and a cleaning block first (37). A lead screw first (31) is rotatably provided in the slide groove first (29). The lead screw first (31) is driven to rotate by a drive motor third. The slider first (33) is provided with an internal thread. The slider first (33) is inserted into the lead screw first (31) and is driven by the lead screw first (31) to slide up and down. One end of the telescopic rod first (35) is fixed to the right side of the slider first (33). The other end of the telescopic rod first (35) is fixedly provided with a cleaning block first (37). The telescopic rod first (35) can drive the cleaning block first (37) away from or close to the rotating cylinder (16).

8. The active five-tone speech therapy system as described in claim 7, characterized in that: The inner cylinder (20) is also provided with a piston adjustment power assembly (26). The piston adjustment power assembly (26) includes a power mounting plate (50) fixedly installed on the inner side of the inner cylinder (20), a lead screw (51) rotatably installed on the power mounting plate (50), and a guide rod (52) fixedly installed on the power mounting plate (50). The upper end of the lead screw (51) is rotatably installed in the upper sealing plate (23), and the guide rod (52) is fixedly installed in the upper sealing plate (23). A piston adsorption slider (27) is passed through the lead screw (51) and the guide rod (52). The rotation of the lead screw (51) can drive the piston adsorption slider (27) to slide up and down. A first magnet is provided on the side of the piston (28) that contacts the inner cylinder (20), and a second magnet is provided on the side of the piston adsorption slider (27) that contacts the inner cylinder. The first magnet and the second magnet attract each other.

Images