Intelligent Rehabilitation System for Voice Quality Disorders Based on the ICF-RFT Framework

Abstract

This invention relates to the field of voice and speech rehabilitation, and particularly to an intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework. The system comprises eight rehabilitation modules: fundamental frequency perturbation, amplitude perturbation (exaggerated), amplitude perturbation (lowed), glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , nasal flow (exaggerated), and nasal flow (lowed). The system includes a real-time voice quality signal detection and processing unit, a main control unit, a real-time audiovisual and tactile feedback unit, a rehabilitation cloud platform, a mobile rehabilitation learning machine, and a mobile intelligent screening device. This invention can set rehabilitation modes according to the user's voice quality disorder type, thereby intelligently selecting rehabilitation treatment content and steps. It is simple and convenient to operate, and can provide differentiated treatment plans based on different genders, age groups, and degrees of voice quality damage.

Description

Technical Field

[0001] This invention relates to the field of voice and speech rehabilitation technology, and in particular to an intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework. Background Technology

[0002] Normal voice production requires a sudden inhalation of a large amount of air followed by a steady exhalation to maintain sufficient subglottic pressure, enabling the vocal cords to vibrate. The resulting sound energy pulse signal passes through the vocal tract, where it is modulated by the resonance cavities formed by the various organs within the tract, and finally propagates into the air to form sound waves. Voice therapy must consider the process of voice production; dysfunction at any stage of this process can lead to abnormal voices, such as breathy sounds, hoarseness, or abnormal resonance quality.

[0003] b3101 sound quality, also known as voice quality function, refers to the function that produces sound characteristics, including pitch, resonance and other features, including high and low pitch function, excessive nasality (hypernasal function), insufficient nasality (hyponasal function), difficulty in vocalization, hoarseness or harshness; it can be measured by 6 parameters, including fundamental frequency perturbation, glottal noise, amplitude perturbation, second formant frequencies F2 / i / and F2 / u / , and nasal flow.

[0004] For voice quality issues related to b3101, the most common and well-known method in voice therapy is facilitation therapy, which can be divided into three main categories based on breathing, vocalization, and resonance functions. Each category contains more than ten targeted treatment methods. Therapists can select the corresponding method based on the patient's vocal function impairment and combine it with modern voice therapy techniques. Real-time monitoring tracks the effectiveness of each treatment, helping therapists and patients to clearly understand the progress after each treatment and obtain feedback on efficacy from objective data. However, for some patients who are limited by geographical location, economic conditions, transportation, or other factors and cannot easily access in-person rehabilitation, current technology has not yet developed intelligent online rehabilitation technologies.

[0005] In view of the above problems, the present invention is proposed to provide an intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework, which can perform real-time measurement and assessment and audiovisual and tactile feedback rehabilitation therapy. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by proposing an intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] The intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework includes the following system structure: a real-time voice quality signal detection and processing unit, a main control unit, a real-time audiovisual and tactile feedback unit, a rehabilitation cloud platform, a mobile rehabilitation learning machine, and a mobile intelligent screening instrument. The real-time voice quality signal detection and processing unit is used to detect and process the user's voice quality signal data in real time.

[0009] The main control unit is connected to the real-time voice quality signal detection and processing unit, and is used for the precise assessment and rehabilitation training of eight related modules, including fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , formant frequency perturbation, nasal flow, and nasooral resonance ratio, throughout the entire process of functional assessment, plan formulation, treatment implementation, and efficacy evaluation.

[0010] The real-time audiovisual tactile feedback unit is connected to the main control unit and uses real-time audiovisual tactile feedback technology to display and provide feedback on voice quality rehabilitation results and related information in real time. The rehabilitation cloud platform is connected to the main control unit via the Internet for data sharing. The mobile rehabilitation learning machine is connected to the rehabilitation cloud platform for assigning rehabilitation tasks and popularizing rehabilitation training. The mobile intelligent screening instrument is connected to the real-time voice quality signal detection and processing unit for comprehensive examination. It obtains the average value of ICF functional impairment through the Internet connection to the rehabilitation cloud platform, providing a diversion for the accurate voice quality assessment and real-time monitoring unit.

[0011] Preferably, the main control unit includes a voice quality accurate assessment and real-time monitoring unit, a rehabilitation mode and rehabilitation step execution unit, an ICF conversion unit, and a voice quality four-stage rehabilitation content push unit. The voice quality accurate assessment and real-time monitoring unit performs accurate assessment and real-time monitoring of data to determine the existence and type of voice quality disorder. The rehabilitation mode and rehabilitation step execution unit sets the rehabilitation mode and executes the rehabilitation steps. The ICF conversion unit performs ICF conversion on the accurate assessment and real-time monitoring data. The voice quality four-stage rehabilitation content push unit selects and pushes rehabilitation content for each of the four stages.

[0012] Preferably, the real-time voice quality signal detection and processing unit is used to acquire user voice quality signal data in real time. The voice quality data is obtained by connecting a microphone and / or an electroglot diagram to a single-channel low-pass filter or a dual-channel low-pass filter.

[0013] The main technical parameters for analyzing and controlling data error and response rate error of the input voice quality signal are as follows:

[0014] a) Frequency error: ±2%;

[0015] b) Voltage error: ±3%;

[0016] c) Real-time response rate of the base frequency: ≤6ms.

[0017] d) LPC real-time response rate: ≤45ms (LPC: Linear Prediction Spectrum);

[0018] e) Real-time spectrogram resolution:

[0019] Narrowband (60Hz): 12.7ms ± 4%;

[0020] Midband (120Hz): 12.7ms ± 4%;

[0021] Broadband (240Hz): 12.7ms ± 4%.

[0022] The real-time voice quality signal detection and processing unit includes the b3101 comprehensive inspection module: it obtains three measurement values—fundamental frequency perturbation, glottal noise energy, and amplitude perturbation—from the voice quality signal; it then performs ICF damage degree conversion on each objective measurement value to obtain the average voice quality ICF value; (as shown in Table 1).

[0023] Table 1: ICF Comprehensive Examination of Voice Quality and Function

[0024]

[0025] The b3101 intelligent rehabilitation module in the main control unit includes eight modules: fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , formant frequency perturbation F2f, nasal flow, and nasooral resonance ratio (NOR). Each module includes: precise assessment, damage level conversion (as shown in Table 2), and its rehabilitation content push and real-time monitoring (e.g., ...). Figure 2 (As shown).

[0026] Table 2: ICF Voice Quality Function Precision Assessment

[0027]

[0028]

[0029] like Figure 3 A voice quality disorder intelligent rehabilitation system based on the ICF-RFT framework applies a non-therapeutic fundamental frequency perturbation intelligent rehabilitation method, including the following steps:

[0030] S21: ICF fundamental frequency perturbation = 1, corresponding real-time monitoring data. The first stage is: breathing relaxation training, real-time slow and steady exhalation method, and real-time feedback training for voiced and unvoiced sounds (combined with real-time slow and steady exhalation method); the second stage is: real-time counting method, real-time sound duration feedback training (combined with real-time counting method), real-time rapid forceful exhalation method, and real-time loudness feedback training (combined with real-time rapid forceful exhalation method); the third stage is: real-time singing method, real-time sound duration feedback training (combined with real-time singing method), real-time word-by-word sentence length increase method, and real-time sound duration feedback training (combined with real-time word-by-word sentence length increase method); the fourth stage is: real-time singing method (combined with syllable duration practice) and real-time slow and steady exhalation method (combined with pause and onset practice).

[0031] S22: ICF fundamental frequency perturbation = 2, corresponding real-time monitoring data. The first stage is: real-time flat-tone forward beeping, real-time pitch feedback training (combined with real-time flat-tone forward beeping), real-time chanting method, and real-time pitch feedback training (combined with real-time chanting method); the second stage is: real-time flat-tone rotating beeping, real-time pitch feedback training (combined with real-time flat-tone rotating beeping), real-time humming method, and real-time pitch feedback training (perception); the third stage is: real-time rising beeping, real-time pitch feedback training (perception), real-time pitch gradient training method, and real-time pitch feedback training (combined with real-time pitch gradient training method); the fourth stage is: real-time humming method (combined with pitch change practice) and real-time chanting method (combined with andante stress);

[0032] S23: ICF fundamental frequency perturbation = 3, corresponding real-time monitoring data. The first stage is: real-time flat forward toeing and real-time pitch feedback training (combined with real-time flat forward toeing); the second stage is: real-time flat rotating toeing and real-time pitch feedback training (combined with real-time flat rotating toeing), real-time humming method, and real-time pitch feedback training (combined with real-time humming method); the third stage is: throat massage method, real-time chanting method, and real-time pitch feedback training (combined with real-time chanting method); the fourth stage is: real-time flat forward toeing (combined with syllable duration practice) and real-time chanting method (combined with slow tempo emphasis).

[0033] S24: ICF fundamental frequency perturbation = 4, corresponding real-time monitoring data. The first stage is: throat massage method, jitter real-time feedback training (perception); the second stage is: neck relaxation training, jitter real-time feedback training; the third stage is: real-time flat tone forward humming, tone real-time feedback training (combined with real-time flat tone forward humming), tone real-time feedback training (perception); the fourth stage is: real-time vowel slow tempo stress reading, real-time vowel ascending tempo stress reading.

[0034] like Figure 4 The intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework applies a non-therapeutic approach to amplitude perturbation-based intelligent rehabilitation caused by hyperactive vocal function, including the following steps:

[0035] S31: ICF amplitude perturbation (exaggeration) = 1, corresponding real-time monitoring data. The first stage is: neck relaxation training, real-time chanting method, and loudness real-time feedback training (combined with real-time chanting method); the second stage is: real-time flat-tone forward tapping, pitch real-time feedback training (combined with real-time flat-tone forward tapping), real-time humming method, loudness real-time feedback training (combined with real-time humming method), real-time loudness gradient training method (changing loudness), and loudness real-time feedback training (combined with real-time loudness gradient training); the third stage is: real-time flat-tone rotating tapping, pitch real-time feedback training (combined with real-time flat-tone rotating tapping), real-time singing method, loudness real-time feedback training (combined with real-time singing method), and Shimmer real-time feedback training (perception); the fourth stage is: real-time singing method (combined with syllable duration training) and real-time humming method (combined with fast-paced emphasis).

[0036] S32: ICF amplitude perturbation (exaggeration) = 2, corresponding real-time monitoring data. The first stage is: real-time breath-based articulation method, real-time feedback training of voiced and unvoiced sounds (combined with real-time breath-based articulation method), real-time chanting method, and real-time loudness feedback training (combined with real-time chanting method); the second stage is: real-time level tone forward tapping, real-time pitch feedback training (combined with real-time level tone forward tapping), real-time humming method, and real-time loudness feedback training (combined with real-time humming method); the third stage is: real-time level tone rotating tapping, real-time pitch feedback training (combined with real-time level tone rotating tapping), real-time loudness gradient training method (changing loudness), and real-time loudness feedback training (combined with real-time loudness gradient training); the fourth stage is: real-time chanting method combined with slow tempo emphasis, and real-time chanting method combined with andante emphasis;

[0037] S33: ICF amplitude perturbation (exaggeration) = 3, corresponding real-time monitoring data. The first stage is: neck relaxation training and shimmer real-time feedback training; the second stage is: real-time breath-based articulation, voiced / unvoiced sound real-time feedback training (combined with real-time breath-based articulation), real-time chanting, and loudness real-time feedback training (combined with real-time chanting); the third stage is: real-time level tone forward humming, pitch real-time feedback training (combined with real-time level tone forward humming), real-time humming, and loudness real-time feedback training (combined with real-time humming); the fourth stage is: real-time humming (combined with slow tempo emphasis) and real-time humming (combined with andante emphasis).

[0038] S34: ICF amplitude perturbation (exaggeration) = 4, corresponding real-time monitoring data. The first stage is: throat massage method, Shimmer real-time feedback training (perception); the second stage is: neck relaxation training, Shimmer real-time feedback training (perception); the third stage is: real-time breath-based articulation method, voiceless and voiced sound real-time feedback training (combined with real-time breath-based articulation method), Shimmer real-time feedback training (perception); the fourth stage is: real-time vowel slow stress reading, real-time vowel ascending stress reading.

[0039] like Figure 5 The intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework applies a non-therapeutic approach to amplitude perturbation-based intelligent rehabilitation for vocal dysfunction, including the following steps:

[0040] S41: ICF amplitude perturbation (low) = 1, corresponding real-time monitoring data. The first stage is: neck relaxation training, real-time forceful chair-moving method, and loudness real-time feedback training (combined with real-time forceful chair-moving method); the second stage is: real-time flat forward toeing, pitch real-time feedback training (combined with real-time flat forward toeing), real-time humming method, loudness real-time feedback training (combined with real-time humming method), real-time loudness gradient training method, and loudness real-time feedback training (combined with real-time loudness gradient training); the third stage is: real-time flat rotating toeing, pitch real-time feedback training (combined with real-time flat rotating toeing), real-time singing method, loudness real-time feedback training (combined with real-time singing method), and Shimmer real-time feedback training (perception); the fourth stage is: real-time singing method (combined with syllable duration practice) and real-time humming method (combined with fast-paced emphasis).

[0041] S42: ICF amplitude perturbation (low) = 2, corresponding real-time monitoring data. The first stage is: real-time flat pitch forward toeing, real-time pitch feedback training (combined with real-time flat pitch forward toeing), real-time forceful chair-moving method, and real-time loudness feedback training (combined with real-time forceful chair-moving method); the second stage is: real-time pitch-changing forward toeing, real-time pitch feedback training (combined with real-time pitch-changing forward toeing), real-time collision method, real-time loudness feedback training (combined with real-time collision method), real-time humming method, and real-time loudness feedback training (combined with real-time humming method); the third stage is: real-time flat pitch rotating toeing, real-time pitch feedback training (combined with real-time flat pitch rotating toeing), real-time loudness gradient training method, and real-time loudness feedback training (combined with real-time loudness gradient training); the fourth stage is: real-time humming method (combined with slow tempo emphasis) and real-time humming method (combined with andante emphasis).

[0042] S43: ICF amplitude perturbation (low) = 3, corresponding real-time monitoring data. The first stage is: real-time flat pitch forward tapping, real-time pitch feedback training (combined with real-time flat pitch forward tapping); the second stage is: real-time pitch-changing forward tapping, real-time pitch feedback training (combined with real-time pitch-changing forward tapping), real-time forceful chair-moving method, real-time loudness feedback training (combined with real-time forceful chair-moving method); the third stage is: real-time flat pitch rotating tapping, real-time pitch feedback training (combined with real-time flat pitch rotating tapping), real-time humming method, real-time pitch feedback training (combined with real-time humming method); the fourth stage is: real-time humming method (combined with slow tempo emphasis), real-time humming method (combined with andante emphasis).

[0043] S44: ICF amplitude perturbation (low) = 4, corresponding real-time monitoring data. The first stage is: throat massage method, Shimmer real-time feedback training (perception); the second stage is: neck relaxation training, Shimmer real-time feedback training (perception); the third stage is: neck relaxation training, real-time forceful chair moving method, loudness real-time feedback training (combined with real-time forceful chair moving method); the fourth stage is: real-time vowel slow tempo stress reading, real-time vowel ascending tempo stress reading.

[0044] like Figure 6 A smart rehabilitation system for voice quality disorders based on the ICF-RFT framework applies a non-therapeutic glottic noise energy smart rehabilitation method, including the following steps:

[0045] S51: ICF glottal noise energy = 1, corresponding real-time monitoring data. The first stage is: real-time level tone rotation tapping, real-time pitch feedback training (combined with real-time level tone rotation tapping), real-time masking method, and real-time feedback training of voiced and unvoiced sounds (combined with real-time masking method); the second stage is: real-time rising tone tapping, real-time pitch feedback training (combined with real-time rising tone tapping), real-time chair-moving method, and real-time loudness feedback training (combined with real-time chair-moving method); the third stage is: real-time collision method, real-time loudness feedback training (combined with real-time collision method), real-time loudness gradient training method, real-time loudness feedback training (combined with real-time loudness gradient training method), and real-time noise NNE feedback training (combined with real-time loudness gradient training method); the fourth stage is: real-time collision method (combined with loudness change practice) and real-time chair-moving method (combined with loudness change practice);

[0046] S52: ICF glottal noise energy = 2, corresponding real-time monitoring data. The first stage is: real-time flat pitch forward tapping, real-time pitch feedback training (combined with real-time flat pitch forward tapping), real-time chewing method, and real-time feedback training of voiced and unvoiced sounds (combined with real-time chewing method); the second stage is: real-time flat pitch rotating tapping, real-time pitch feedback training (combined with real-time flat pitch rotating tapping), real-time collision method, and real-time loudness feedback training (combined with real-time collision method); the third stage is: real-time rising pitch tapping, real-time pitch feedback training (combined with real-time rising pitch tapping), real-time loudness gradient training method, real-time loudness feedback training (combined with real-time loudness gradient training method), real-time forceful chair-moving method, and real-time loudness feedback training (combined with real-time forceful chair-moving method); the fourth stage is: real-time chewing method (combined with pause onset practice) and real-time collision method (combined with loudness change practice);

[0047] S53: ICF glottal noise energy = 3, corresponding real-time monitoring data. The first stage is: real-time flat pitch forward tapping, real-time pitch feedback training (combined with real-time flat pitch forward tapping); the second stage is: real-time rising pitch tapping, real-time pitch feedback training (combined with real-time rising pitch tapping), real-time chewing method, real-time feedback training of voiced and unvoiced sounds (combined with real-time chewing method); the third stage is: real-time flat pitch rotating tapping, real-time pitch feedback training (combined with real-time flat pitch rotating tapping), real-time arm swinging and pushing method, real-time loudness feedback training (combined with real-time arm swinging and pushing method), real-time forceful chair moving method, real-time loudness feedback training (combined with real-time forceful chair moving method); the fourth stage is: real-time chewing method (combined with syllable duration practice), real-time chewing method (combined with pitch change practice);

[0048] S54: ICF glottal noise energy = 4, corresponding real-time monitoring data. The first stage is: throat massage method, noise NNE real-time feedback training (perception); the second stage is: neck relaxation training, voiced and unvoiced sound real-time feedback training (perception); the third stage is: real-time level tone forward pluck, pitch real-time feedback training (combined with real-time level tone forward pluck), real-time arm swing and push method, loudness real-time feedback training (combined with real-time arm swing and push method); the fourth stage is: real-time vowel slow tempo stress reading, real-time vowel ascending tempo stress reading.

[0049] like Figure 7 A voice quality disorder intelligent rehabilitation system based on the ICF-RFT framework applies a non-therapeutic formant frequency F2 / i / intelligent rehabilitation execution method, including the following steps:

[0050] S61: ICF formant frequency F2 / i / =1, corresponding real-time monitoring data. The first stage is: oral relaxation training and formant real-time feedback training (perception); the second stage is: real-time rising pitch tapping, real-time pitch feedback training (combined with real-time rising pitch tapping), real-time front position method, and formant real-time feedback training (combined with real-time front position method); the third stage is: real-time front position method, formant real-time feedback training (combined with real-time front position method), real-time pitch gradient training method, and real-time pitch feedback training (combined with real-time pitch gradient training method); the fourth stage is: real-time front position method (combined with slow tempo emphasis) and real-time front position method (combined with andante emphasis).

[0051] S62: ICF formant frequency F2 / i / =2, corresponding real-time monitoring data. The first stage is: oral relaxation training and formant real-time feedback training (perception); the second stage is: real-time rising pitch tapping, real-time pitch feedback training (combined with real-time rising pitch tapping), real-time front position method, and real-time pitch feedback training (combined with real-time front position method); the third stage is: real-time rising pitch tapping, real-time pitch feedback training (combined with rising pitch tapping), real-time tongue extension method, and formant real-time feedback training (combined with real-time tongue extension method); the fourth stage is: real-time tongue extension method (combined with slow tempo emphasis) and real-time tongue extension method (combined with andante emphasis).

[0052] S63: ICF formant frequency F2 / i / =3, corresponding real-time monitoring data. The first stage is: oral relaxation training and formant real-time feedback training (perception); the second stage is real-time level pitch forward pluck and pitch real-time feedback training (combined with real-time level pitch forward pluck); the third stage is: real-time rising pitch pluck and pitch real-time feedback training (combined with real-time rising pitch pluck); the fourth stage is: oral relaxation training (combined with slow tempo emphasis) and oral relaxation training (combined with andante emphasis).

[0053] S64: ICF formant frequency F2 / i / =4. The first stage is: oral relaxation training and formant real-time feedback training (perception); the second stage is: oral relaxation training and tongue area map real-time feedback training (perception); the third stage is: oral relaxation training and pitch real-time feedback training (perception); the fourth stage is: oral relaxation training (combined with slow tempo emphasis) and oral relaxation training (combined with andante emphasis).

[0054] like Figure 8 A voice quality disorder intelligent rehabilitation system based on the ICF-RFT framework applies a non-therapeutic formant frequency F2 / u / intelligent rehabilitation execution method, including the following steps:

[0055] S71: ICF formant frequency F2 / u / =1, corresponding real-time monitoring data. The first stage is: oral relaxation training and formant real-time feedback training (perception); the second stage is: real-time falling tone tapping, real-time pitch feedback training (combined with real-time falling tone tapping), real-time back position tone method, and formant real-time feedback training (combined with real-time back position tone method); the third stage is: real-time back position tone method, real-time pitch feedback training (combined with real-time back position tone method), real-time pitch gradient training method, and real-time pitch feedback training (combined with real-time pitch gradient training method); the fourth stage is: real-time back position tone method (combined with slow tempo emphasis) and real-time back position tone method (combined with andante emphasis).

[0056] S72: ICF formant frequency F2 / u / =2, corresponding real-time monitoring data. The first stage is: oral relaxation training and formant real-time feedback training (perception); the second stage is: real-time falling tone tapping, real-time tone feedback training (combined with real-time falling tone tapping), real-time back position tone method, and real-time tone feedback training (combined with real-time back position tone method); the third stage is: real-time falling tone tapping, real-time tone feedback training (combined with real-time falling tone tapping), real-time tone gradient training method, and real-time tone feedback training (combined with real-time tone gradient training method); the fourth stage is: real-time back position tone method (combined with slow tempo emphasis) and real-time back position tone method (combined with andante emphasis).

[0057] S73: ICF resonant frequency F2 / u / =3, corresponding real-time monitoring data. The first stage is: oral relaxation training and real-time feedback training of resonant frequencies (perception); the second stage is: real-time level pitch forward pluck and real-time pitch feedback training (combined with real-time level pitch forward pluck); the third stage is: real-time falling pitch pluck and real-time pitch feedback training (combined with real-time falling pitch pluck); the fourth stage is: oral relaxation training (combined with slow tempo emphasis) and oral relaxation training (combined with andante emphasis).

[0058] S74: ICF formant frequency F2 / u / =4, corresponding real-time monitoring data. The first stage is: oral relaxation training and formant real-time feedback training (perception); the second stage is: oral relaxation training and tongue area map real-time feedback training (perception); the third stage is: oral relaxation training and pitch real-time feedback training (perception); the fourth stage is: oral relaxation training (combined with slow tempo emphasis) and oral relaxation training (combined with andante emphasis).

[0059] like Figure 9 The intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework applies a non-therapeutic method for intelligent rehabilitation of nasal flow (hypertension), including the following steps:

[0060] S81: ICF nasal flow (hyperactive) = 1, corresponding real-time monitoring data. The first stage is: real-time nasal relaxation training, real-time nasal / lateral sound stimulation, and nasal flow NL real-time feedback training (combined with real-time nasal / lateral sound stimulation); the second stage is: real-time nasal relaxation training, real-time oral resonance training, and voiceless / voiced sound real-time feedback training (combined with real-time oral resonance); the third stage is: real-time pitch gradient training, real-time pitch feedback training (combined with real-time pitch gradient training), real-time oral resonance, and nasal flow NL real-time feedback training (combined with real-time oral resonance); the fourth stage is: real-time nasal / lateral sound stimulation (combined with andante rhythm), and real-time oral resonance (combined with andante rhythm).

[0061] S82: ICF nasal flow (hyperactive) = 2, corresponding real-time monitoring data. The first stage is: real-time nasal relaxation training, real-time nasal / lateral sound stimulation, and nasal flow NL real-time feedback training (combined with real-time nasal / lateral sound stimulation); the second stage is: real-time nasal relaxation training, real-time oral resonance training, and voiceless / voiced sound real-time feedback training (combined with real-time oral resonance); the third stage is: real-time oral resonance, pitch real-time feedback training (combined with real-time oral resonance), real-time pitch gradient training, and pitch real-time feedback training (combined with real-time pitch gradient training); the fourth stage is: real-time nasal / lateral sound stimulation (combined with slow tempo) and real-time nasal / lateral sound stimulation (combined with andante tempo).

[0062] S83: ICF nasal flow (hyperactive) = 3, corresponding real-time monitoring data. The first stage is: real-time nasal relaxation training and nasal flow NL real-time feedback training (perception); the second stage is: real-time nasal relaxation training, real-time oral resonance training, and voiceless / voiced sound real-time feedback training (combined with real-time oral resonance training); the third stage is: real-time nasal relaxation training, real-time oral resonance training, and voiceless / voiced sound real-time feedback training (combined with real-time oral resonance training); the fourth stage is: real-time oral resonance training (combined with slow tempo) and real-time oral resonance training (combined with andante tempo).

[0063] S84: ICF resonant frequency F2 / u / =4, corresponding real-time monitoring data. The first stage is: real-time nasal relaxation training and nasal flow NL real-time feedback training (perception); the second stage is real-time oral resonance method and voiced / unvoiced sound real-time feedback training (combined with real-time oral resonance method); the third stage is: real-time nasal relaxation training and pitch real-time feedback training (perception); the fourth stage is: real-time nasal relaxation training (combined with slow tempo emphasis) and real-time nasal relaxation training (combined with andante emphasis).

[0064] like Figure 10 A voice quality disorder intelligent rehabilitation system based on the ICF-RFT framework applies a non-therapeutic nasal flow (low) intelligent rehabilitation implementation method, including the following steps:

[0065] S91: ICF nasal flow (low) = 1, corresponding real-time monitoring data. The first stage is: real-time nasal relaxation training, real-time nasal / lateral sound stimulation, and nasal flow NL real-time feedback training (combined with real-time nasal / lateral sound stimulation); the second stage is: real-time nasal relaxation training, real-time nasal resonance training, and voiceless / voiced sound real-time feedback training (combined with real-time nasal resonance); the third stage is: real-time pitch gradient training, real-time pitch feedback training (combined with real-time pitch gradient training), real-time nasal resonance, and nasal flow NL real-time feedback training (combined with real-time nasal resonance); the fourth stage is: real-time nasal / lateral sound stimulation (combined with andante rhythm), and real-time nasal resonance (combined with andante rhythm).

[0066] S82: ICF nasal flow (hyperactive) = 2, corresponding real-time monitoring data. The first stage is: real-time nasal relaxation training, real-time nasal / lateral sound stimulation, and nasal flow NL real-time feedback training (combined with real-time nasal / lateral sound stimulation); the second stage is: real-time nasal relaxation training, real-time nasal resonance training, and voiceless / voiced sound real-time feedback training (combined with nasal resonance); the third stage is: real-time nasal resonance, real-time pitch feedback training (combined with real-time nasal resonance), real-time pitch gradient training, and real-time pitch feedback training (combined with real-time pitch gradient training); the fourth stage is: real-time nasal / lateral sound stimulation (combined with slow tempo) and real-time nasal / lateral sound stimulation (combined with andante tempo).

[0067] S83: ICF nasal flow (hyperactive) = 3, corresponding real-time monitoring data. The first stage is: real-time nasal relaxation training and nasal flow NL real-time feedback training (perception); the second stage is: real-time nasal relaxation training, real-time nasal resonance training, and voiceless / voiced sound real-time feedback training (combined with real-time nasal resonance training); the third stage is: real-time nasal relaxation training, real-time nasal resonance training, and voiceless / voiced sound real-time feedback training (combined with real-time nasal resonance training); the fourth stage is: real-time nasal resonance training (combined with slow tempo) and real-time nasal resonance training (combined with andante tempo).

[0068] S84: ICF resonant frequency F2 / u / =4, corresponding real-time monitoring data, the first stage is: real-time nasal relaxation training, nasal flow NL real-time feedback training (perception); the second stage is: real-time nasal relaxation training, nasal flow NL real-time feedback training (perception); the third stage is: real-time nasal relaxation training, pitch real-time feedback training (perception); the fourth stage is: real-time nasal relaxation training (combined with slow tempo emphasis) and real-time nasal relaxation training (combined with andante emphasis).

[0069] The intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework classifies the types of voice quality disorders, including:

[0070] (1) Neurogenic voice disorder: Neurogenic voice disorder usually refers to voice problems caused by neurological diseases, impaired muscle control and nerve innervation of muscles related to the breathing, vocalization, resonance and articulation systems, or voice disorders caused by any damage or disease of the peripheral or central nervous system. Medical diagnoses related to neurogenic voice disorder include: vocal cord paralysis, laryngeal dystonia, essential tremor, Parkinson's disease, Huntington's disease, myasthenia gravis, multiple sclerosis, amyotrophic lateral sclerosis, progressive supranuclear palsy, multiple system atrophy and acquired brain injury (traumatic brain injury, TBI, CVA), etc.

[0071] (2) Organic voice disorders: voice problems caused by organic lesions of the vocal organs, including congenital abnormalities of the vocal organs, proliferative lesions of the vocal cords, laryngeal tumors, inflammatory lesions of the larynx and other lesions of the vocal cords, such as laryngomalacia, vocal cord nodules, vocal cord polyps, vocal cord cysts, vocal cord scars, laryngeal leukoplakia, vocal cord Reynolds edema, laryngeal papilloma, vocal cord atrophy, etc.

[0072] (3) Functional voice disorder: There is no organic or neurological lesion in the vocal system (respiratory organs, vocal organs and resonating organs), but the voice is abnormal in pitch, loudness and tone quality mainly due to abuse, misuse or disuse of the voice.

[0073] This invention provides a voice therapy instrument with real-time voice quality measurement and audiovisual feedback technology, comprising: a real-time processing unit, a main control unit, and a real-time voice quality audiovisual feedback unit; the real-time processing unit is used to acquire data on fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequencies F2 / i / , F2 / u / , formant frequency disturbance, nasal flow, and nasooral resonance ratio when a user produces speech; the main control unit is connected to the real-time processing unit and is used to measure relevant data of voice production parameters in real time and determine the user's existing voice quality disorder, and select rehabilitation content and steps for speech voice rehabilitation according to the type of voice quality disorder; the real-time voice quality audiovisual feedback unit is connected to the main control unit and is used to display relevant data of voice quality parameters and rehabilitation results of voice quality disorder in real time.

[0074] The voice therapy instrument proposed in this invention, which utilizes real-time voice quality measurement and audiovisual feedback technology, includes a main control unit comprising: a rehabilitation mode and content storage module, a neurogenic voice disorder rehabilitation step and execution module, and a real-time fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , and nasal flow measurement and discrimination module. The rehabilitation mode and content storage module stores the rehabilitation content used during the rehabilitation process, including fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , nasal flow focused rehabilitation mode, fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , and nasal flow rehabilitation content, as well as real-time feedback of fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , and nasal flow data, and adds fundamental frequency perturbation, amplitude perturbation, glottal noise energy, and formant frequency F2 / i / . The training content includes formant frequencies F2 / u / and nasal flow. The rehabilitation steps and execution module for neurogenic voice disorders are connected to the rehabilitation mode and content storage module and the real-time audiovisual feedback unit. This module is used to set the rehabilitation mode for fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequencies F2 / i / , formant frequencies F2 / u / , and nasal flow, and to run the rehabilitation content for these parameters in real time. The real-time measurement and discrimination module for fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequencies F2 / i / , formant frequencies F2 / u / , and nasal flow is connected to the real-time processing unit and the rehabilitation execution module. This module is used to judge the state of fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequencies F2 / i / , formant frequencies F2 / u / , and nasal flow when the user produces speech, and to determine the existing voice quality disorder in the user's voice.

[0075] This invention also proposes a voice therapy method using real-time voice quality measurement and audiovisual feedback technology. Relevant parameters include fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , and nasal flow. Using the aforementioned voice therapy instrument, the voice therapy method includes:

[0076] Step 1:

[0077] The real-time fundamental frequency jitter, amplitude jitter, and glottal noise energy measurement unit measures the user's fundamental frequency jitter, amplitude jitter, and glottal noise energy data in real time. Fundamental frequency jitter refers to the rate of change of sound wave frequency when producing a stable voice. It measures the fundamental frequency difference between a given period and several adjacent periods, expressed as a percentage (%). It primarily reflects the degree of roughness, and secondarily the degree of hoarseness, and is one of the best indicators for measuring voice quality related to vibration sources. Generally, the normal value is less than 0.5%. Glottal noise (NNE) refers to the disturbance noise energy caused by air leakage due to incomplete glottal closure during voice production, expressed as decibels (dB). It primarily reflects the degree of breathiness, and secondarily the degree of hoarseness, and is also one of the best indicators for measuring voice quality related to vibration sources. Generally, the normal value is less than -10 dB. Amplitude perturbation (shimmer) refers to the rate of change of sound wave amplitude when phoning with a stable voice. It measures the amplitude difference between a given period and several preceding or subsequent periods, expressed as a percentage (%). It primarily reflects the degree of hoarseness. Generally, the normal value is less than 3%. Sound wave amplitude can be obtained by measuring the peak-to-peak value of the sound wave. The data is sent in real-time to the real-time fundamental frequency perturbation, amplitude perturbation, and glottal noise energy measurement and discrimination module in the main control unit.

[0078] The real-time formant frequency F2 / i / and formant frequency F2 / u / measurement units measure the user's formant frequency F2 / i / and formant frequency F2 / u / data in real time. If the F2 value of / u / is lower than that of normal individuals of the same age and sex or within the normal range, it indicates good posterior tongue movement ability; if it is higher than that of normal individuals of the same age and sex, it indicates anterior focusing. If the F2 value of / i / is higher than that of normal individuals of the same age and sex or within the normal range, it indicates good anterior tongue movement ability; if it is lower than that of normal individuals of the same age and sex, it indicates posterior focusing. The data is sent in real time to the real-time formant frequency F2 / i / and formant frequency F2 / u / measurement and discrimination module in the main control unit.

[0079] The real-time nasal flow measurement unit measures the user's nasal flow data in real time. Nasal flow (NL) is the ratio of nasal sound pressure level N to output sound pressure level (the sum of oral sound pressure level O and nasal sound pressure level N) during speech, expressed as a percentage. It primarily reflects the nasal energy during speech and can help determine whether there is hypernasalization or hyponasalization. If the nasal flow in a hypernasalization test exceeds the upper limit of the normal range for individuals of the same age and sex, it indicates hypernasalization; if the nasal flow in a hyponasalization test does not reach the lower limit of the normal range for individuals of the same age and sex, it indicates hyponasalization. The data is sent in real time to the real-time nasal flow measurement and discrimination module in the main control unit.

[0080] The real-time fundamental frequency perturbation discrimination module makes a discrimination based on the fundamental frequency perturbation data and the fundamental frequency perturbation target value and its corresponding range, and sends the discrimination result to the neurogenic voice disorder rehabilitation steps and execution module in real time;

[0081] The real-time amplitude perturbation discrimination module makes a discrimination based on the amplitude perturbation data and the amplitude perturbation target value and its corresponding range, and sends the discrimination result to the neurogenic voice disorder rehabilitation steps and execution module in real time;

[0082] The real-time glottal noise energy discrimination module makes a discrimination based on the glottal noise energy data and the glottal noise energy target value and its corresponding range, and sends the discrimination result to the neurogenic voice disorder rehabilitation steps and execution module in real time;

[0083] The real-time formant frequency F2 / i / discrimination module makes a discrimination based on the formant frequency F2 / i / data and the target value of formant frequency F2 / i / and its corresponding range, and sends the discrimination result to the neurogenic voice disorder rehabilitation steps and execution module in real time;

[0084] The real-time formant frequency F2 / u / discrimination module makes a discrimination based on the formant frequency F2 / u / data and the target value of formant frequency F2 / u / and its corresponding range, and sends the discrimination result to the neurogenic voice disorder rehabilitation steps and execution module in real time;

[0085] The real-time nasal flow discrimination module makes a discrimination based on the nasal flow data and the target value of nasal flow and its corresponding range, and sends the discrimination result to the neurogenic voice disorder rehabilitation steps and execution module in real time;

[0086] Step 3:

[0087] The neurogenic voice disorder rehabilitation steps and execution module set the fundamental frequency perturbation rehabilitation mode according to the discrimination result, read and execute the fundamental frequency perturbation rehabilitation mode and fundamental frequency perturbation rehabilitation content from the rehabilitation mode and content storage module, and transmit the rehabilitation execution result to the real-time fundamental frequency perturbation audiovisual feedback unit.

[0088] The neurogenic voice disorder rehabilitation steps and execution module set an amplitude perturbation rehabilitation mode according to the discrimination result, read and execute the amplitude perturbation rehabilitation mode and amplitude perturbation rehabilitation content from the rehabilitation mode and content storage module, and transmit the rehabilitation execution result to the real-time amplitude perturbation audiovisual feedback unit.

[0089] The neurogenic voice disorder rehabilitation steps and execution module set the glottic noise energy rehabilitation mode according to the discrimination result, read and execute the glottic noise energy rehabilitation mode and glottic noise energy rehabilitation content from the rehabilitation mode and content storage module, and transmit the rehabilitation execution result to the real-time glottic noise energy audiovisual feedback unit.

[0090] The neurogenic voice disorder rehabilitation steps and execution module set the formant frequency F2 / i / rehabilitation mode according to the discrimination result, read and execute the formant frequency F2 / i / rehabilitation mode and formant frequency F2 / i / rehabilitation content from the rehabilitation mode and content storage module, and transmit the rehabilitation execution result to the real-time formant frequency F2 / i / audiovisual feedback unit.

[0091] The neurogenic voice disorder rehabilitation steps and execution module set the formant frequency F2 / u / rehabilitation mode according to the discrimination result, read and execute the formant frequency F2 / u / rehabilitation mode and formant frequency F2 / u / rehabilitation content from the rehabilitation mode and content storage module, and transmit the rehabilitation execution result to the real-time formant frequency F2 / u / audiovisual feedback unit.

[0092] The neurogenic voice disorder rehabilitation steps and execution module set a nasal flow rehabilitation mode based on the discrimination result, read and execute the nasal flow rehabilitation mode and nasal flow rehabilitation content from the rehabilitation mode and content storage module, and transmit the rehabilitation execution result to the real-time nasal flow audiovisual feedback unit.

[0093] Step Four:

[0094] The real-time fundamental frequency perturbation audiovisual feedback unit displays the results of the fundamental frequency perturbation rehabilitation in real time.

[0095] The real-time amplitude perturbation audiovisual feedback unit displays the amplitude perturbation rehabilitation execution results in real time.

[0096] The real-time glottal noise energy audiovisual feedback unit displays the glottal noise energy rehabilitation execution results in real time.

[0097] The real-time resonant frequency F2 / i / audiovisual feedback unit displays the resonant frequency F2 / i / rehabilitation execution result in real time.

[0098] The real-time resonant frequency F2 / u / audiovisual feedback unit displays the resonant frequency F2 / u / rehabilitation execution result in real time.

[0099] The real-time nasal flow audiovisual feedback unit displays the nasal flow rehabilitation execution results in real time.

[0100] The voice therapy method, which utilizes real-time measurement of fundamental frequency perturbations and audiovisual feedback technology, further includes, in step two:

[0101] Step a: The real-time fundamental frequency perturbation measurement and discrimination module converts the user's fundamental frequency perturbation data into a fundamental frequency perturbation limit value, and qualitatively judges the degree of fundamental frequency perturbation damage;

[0102] Step b: The real-time fundamental frequency perturbation measurement and discrimination module sends the user information and fundamental frequency perturbation disorder type information to the rehabilitation execution module;

[0103] Step c: The real-time baseband perturbation measurement and discrimination module makes a discrimination based on user information and the baseband perturbation data and the baseband perturbation focusing limit value.

[0104] The voice therapy method, which utilizes real-time amplitude perturbation measurement and audiovisual feedback technology, further includes, in step two:

[0105] Step a: The real-time amplitude perturbation measurement and discrimination module converts the user's amplitude perturbation data into amplitude perturbation limit values ​​and qualitatively judges the degree of amplitude perturbation damage;

[0106] Step b: The real-time amplitude perturbation measurement and discrimination module sends the user information and amplitude perturbation obstacle type information to the rehabilitation execution module;

[0107] Step c: The real-time amplitude perturbation measurement and discrimination module makes a discrimination based on user information and the amplitude perturbation data and amplitude perturbation focusing limit value.

[0108] The voice therapy method, which utilizes real-time measurement of glottal noise energy and audiovisual feedback technology, further includes, in step two:

[0109] Step a: The real-time glottal noise energy measurement and discrimination module converts the user's glottal noise energy data into a glottal noise energy limit value and qualitatively judges the degree of glottal noise energy damage.

[0110] Step b: The real-time glottal noise energy measurement and discrimination module sends the user information and glottal noise energy impairment type information to the rehabilitation execution module;

[0111] Step c: The real-time glottal noise energy measurement and discrimination module makes a discrimination based on the user information and the glottal noise energy data and the glottal noise energy focusing limit value.

[0112] Among them, the voice therapy method using real-time measurement of formant frequency F2 / i and audiovisual feedback technology, step two further includes:

[0113] Step a: The real-time resonant frequency F2 / i / measurement and discrimination module converts the user's resonant frequency F2 / i / data into a resonant frequency F2 / i / limit value, and performs a qualitative discrimination on the degree of damage to the resonant frequency F2 / i / .

[0114] Step b: The real-time formant frequency F2 / i / measurement and discrimination module sends the user information and the formant frequency F2 / i / obstacle type information to the rehabilitation execution module;

[0115] Step c: The real-time resonant frequency F2 / i / measurement and discrimination module makes a discrimination based on the user information and the resonant frequency F2 / i / data and the resonant frequency F2 / i / focusing limit value.

[0116] Among them, the voice therapy method using real-time measurement of formant frequency F2 / u / and audiovisual feedback technology, step two further includes:

[0117] Step a: The real-time resonant frequency F2 / u / measurement and discrimination module converts the user's resonant frequency F2 / u / data into a resonant frequency F2 / u / limit value, and performs a qualitative discrimination on the degree of damage to the resonant frequency F2 / u / .

[0118] Step b: The real-time formant frequency F2 / u / measurement and discrimination module sends the user information and the formant frequency F2 / u / obstacle type information to the rehabilitation execution module;

[0119] Step c: The real-time resonant frequency F2 / u / measurement and discrimination module makes a discrimination based on the user information and the resonant frequency F2 / u / data and the resonant frequency F2 / u / focusing limit value.

[0120] The voice therapy method, which utilizes real-time nasal flow measurement and audiovisual feedback technology, further includes, in step two:

[0121] Step a: The real-time nasal flow measurement and discrimination module converts the user's nasal flow data into a nasal flow limit value and qualitatively judges the degree of nasal flow damage;

[0122] Step b: The real-time nasal flow measurement and discrimination module sends the user information and nasal flow disorder type information to the rehabilitation execution module;

[0123] Step c: The real-time nasal flow measurement and discrimination module makes a discrimination based on user information and the nasal flow data and nasal flow focusing limit value.

[0124] In the voice therapy method proposed in this invention, which uses real-time measurement of fundamental frequency perturbation and audiovisual feedback technology, step four further includes: repeating steps one to four until the user's speech voice no longer has fundamental frequency perturbation disorder or the rehabilitation training position is stopped.

[0125] In the voice therapy method proposed in this invention, which uses real-time amplitude perturbation measurement and audiovisual feedback technology, step four further includes: repeating steps one to four until the user's speech voice no longer has amplitude perturbation impairment or the rehabilitation training position is stopped.

[0126] In the voice therapy method proposed in this invention, which uses real-time measurement of glottal noise energy and audiovisual feedback technology, step four further includes: repeating steps one to four until the user's speech voice no longer has glottal noise energy impairment or the rehabilitation training position is stopped.

[0127] In the voice therapy method proposed in this invention, which uses real-time measurement of formant frequency F2 / i / and audiovisual feedback technology, step four further includes: repeating steps one to four until the user's speech voice no longer has formant frequency F2 / i / impairment or the rehabilitation training position is stopped.

[0128] In the voice therapy method proposed in this invention, which uses real-time measurement of formant frequency F2 / u / and audiovisual feedback technology, step four further includes: repeating steps one to four until the user's speech voice no longer has formant frequency F2 / u / impairment or the rehabilitation training position is stopped.

[0129] In the voice therapy method proposed in this invention, which uses real-time nasal flow measurement and audiovisual feedback technology, step four further includes: repeating steps one to four until the user's speech voice no longer has nasal flow obstruction or the rehabilitation training position is stopped.

[0130] The beneficial effects of this invention are as follows:

[0131] This invention can set a rehabilitation mode according to the user's voice quality disorder type, thereby intelligently selecting rehabilitation treatment content and steps. It is simple and convenient to operate, reduces the difficulty of treating voice quality disorders, and can provide differentiated treatment plans based on different genders, age groups, and degrees of voice quality damage. Attached Figure Description

[0132] Figure 1 This is a block diagram of the intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework proposed in this invention;

[0133] Figure 2 This is a block diagram of the b3101 intelligent rehabilitation module's rehabilitation content push (four stages) and real-time monitoring system proposed in this invention.

[0134] Figure 3 This is a block diagram of the intelligent rehabilitation system with fundamental frequency perturbation ICF=1 proposed in this invention;

[0135] Figure 4 This is a block diagram of the amplitude perturbation (hyperactivity) ICF=2 intelligent rehabilitation system proposed in this invention;

[0136] Figure 5 This is a block diagram of the amplitude perturbation (low) ICF=3 intelligent rehabilitation system proposed in this invention;

[0137] Figure 6 This is a block diagram of the intelligent rehabilitation system with glottal noise energy ICF=4 proposed in this invention;

[0138] Figure 7 This is a block diagram of the intelligent rehabilitation system with resonant peak frequency F2 / i / ICF = 1 proposed in this invention;

[0139] Figure 8 This is a block diagram of the intelligent rehabilitation system with resonance peak frequency F2 / u / ICF=2 proposed in this invention;

[0140] Figure 9 This is a block diagram of the intelligent rehabilitation system for nasal flow NL (hyperincreased) ICF=3 proposed in this invention;

[0141] Figure 10 This is a block diagram of the intelligent rehabilitation system for nasal flow (NL) (low) ICF=4 proposed in this invention;

[0142] Figure 11 This is a voice quality report proposed in this invention. Detailed Implementation

[0143] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0144] In the technical solution of this invention embodiment, the voice therapy instrument based on the real-time measurement and audiovisual feedback technology of fundamental frequency perturbation includes a real-time processing unit, a main control unit, and a real-time fundamental frequency perturbation audiovisual feedback unit. The real-time processing unit is used to acquire fundamental frequency perturbation data when the user produces speech sounds in real time. The main control unit is connected to the real-time processing unit and is used to measure the fundamental frequency perturbation data in real time and determine the user's fundamental frequency perturbation disorder, selecting rehabilitation content and steps for speech and voice rehabilitation based on the type of fundamental frequency perturbation disorder. The real-time fundamental frequency perturbation audiovisual feedback unit is connected to the main control unit and is used to display the fundamental frequency perturbation data and the rehabilitation results of the fundamental frequency perturbation disorder in real time.

[0145] In the technical solution of this invention embodiment, the voice therapy instrument based on real-time amplitude perturbation measurement and audiovisual feedback technology includes a real-time processing unit, a main control unit, and a real-time amplitude perturbation audiovisual feedback unit. The real-time processing unit is used to acquire amplitude perturbation data when the user produces speech sounds in real time. The main control unit is connected to the real-time processing unit and is used to measure the amplitude perturbation data in real time and determine the amplitude perturbation disorder present in the user, selecting rehabilitation content and steps for speech and voice rehabilitation based on the type of amplitude perturbation disorder. The real-time amplitude perturbation audiovisual feedback unit is connected to the main control unit and is used to display the amplitude perturbation data and the rehabilitation results of the amplitude perturbation disorder in real time.

[0146] In the technical solution of this invention embodiment, the voice therapy instrument based on the real-time measurement and audiovisual feedback technology of glottal noise energy includes a real-time processing unit, a main control unit, and a real-time glottal noise energy audiovisual feedback unit. The real-time processing unit is used to acquire glottal noise energy data when the user produces speech sounds in real time. The main control unit is connected to the real-time processing unit and is used to measure the glottal noise energy data in real time and determine the user's existing glottal noise energy disorder, selecting rehabilitation content and steps for speech and voice rehabilitation based on the type of glottal noise energy disorder. The real-time glottal noise energy audiovisual feedback unit is connected to the main control unit and is used to display the glottal noise energy data and the rehabilitation results of the glottal noise energy disorder in real time.

[0147] In the technical solution of this invention embodiment, the voice therapy instrument using real-time formant frequency F2 / i / measurement and audiovisual feedback technology includes a real-time processing unit, a main control unit, and a real-time formant frequency F2 / i / audiovisual feedback unit. The real-time processing unit is used to acquire formant frequency F2 / i / data when the user produces speech sounds in real time. The main control unit is connected to the real-time processing unit and is used to measure the formant frequency F2 / i / data in real time and determine the user's existing formant frequency F2 / i / impairment, selecting rehabilitation content and steps for speech and voice rehabilitation based on the type of formant frequency F2 / i / impairment. The real-time formant frequency F2 / i / audiovisual feedback unit is connected to the main control unit and is used to display the formant frequency F2 / i / data and the rehabilitation results of the formant frequency F2 / i / impairment in real time.

[0148] In the technical solution of this invention embodiment, the voice therapy instrument using real-time formant frequency F2 / u / measurement and audiovisual feedback technology includes a real-time processing unit, a main control unit, and a real-time formant frequency F2 / u / audiovisual feedback unit. The real-time processing unit 1 is used to acquire formant frequency F2 / u / data when the user produces speech sounds in real time. The main control unit is connected to the real-time processing unit and is used to measure the formant frequency F2 / u / data in real time and determine the user's existing formant frequency F2 / u / impairment, selecting rehabilitation content and steps for speech and voice rehabilitation based on the type of formant frequency F2 / u / impairment. The real-time formant frequency F2 / u / audiovisual feedback unit is connected to the main control unit and is used to display the formant frequency F2 / u / data and the rehabilitation results of the formant frequency F2 / u / impairment in real time.

[0149] In the technical solution of this invention embodiment, the voice therapy instrument based on real-time nasal flow measurement and audiovisual feedback technology includes a real-time processing unit, a main control unit, and a real-time nasal flow audiovisual feedback unit. The real-time processing unit is used to acquire nasal flow data in real time when the user produces speech. The main control unit is connected to the real-time processing unit and is used to measure nasal flow data in real time and determine the user's nasal flow disorder, selecting rehabilitation content and steps for speech and voice rehabilitation based on the type of nasal flow disorder. The real-time nasal flow audiovisual feedback unit is connected to the main control unit and is used to display nasal flow data and rehabilitation results for nasal flow disorder in real time.

[0150] The data on voice quality measured by users is analyzed and compared with a norm database. This database contains normalized mean voice quality, standard deviation of voice quality, and voice quality limits (0, 1, 2, 3, 4) for different genders and age groups (4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17 years old, and 18-99 years old as the adult group, for a total of 15 groups).

[0151] Specifically, the real-time processing unit 1 includes a single-channel low-pass filter, an electroglottic transducer, and a microphone. The real-time processing unit offers four gain levels for the acquired speech audio: 25dB, 30dB, 35dB, and 40dB, with an error of ±1.0dB per level (100Hz-700Hz fundamental frequency range). The single-channel low-pass filter uses three low-pass frequencies: 5kHz, 10kHz, and 20kHz. Selecting the 5kHz level: when the input signal frequency is 5.5kHz, the maximum attenuation is ≥50dB. Selecting the 10kHz level: when the input signal frequency is 11Hz, the maximum attenuation is ≥50dB. Selecting the 20kHz level: when the input signal frequency is 22kHz, the maximum attenuation is ≥50dB. When there is no signal input, the static noise is ≤1mV.

[0152] See Figure 1 The main control unit 2 includes: a rehabilitation mode and content storage module 21, a neurogenic voice disorder rehabilitation step and execution module 22, and a real-time voice quality parameter measurement and discrimination module 23. The rehabilitation mode and content storage module 21 stores the voice quality parameter focusing rehabilitation mode, voice quality parameter rehabilitation content, and real-time feedback of voice quality parameter data used during the rehabilitation process. It also stores content for reducing voice quality parameter training (awakening, induction, increasing loudness / pitch, self-feedback), and increasing voice quality parameter training (awakening, induction, pitch shifting, self-feedback). The content includes audio materials for playback, static image materials and / or dynamic video materials for display, and text and / or numbers for display. The neurogenic voice disorder rehabilitation step and execution module 22 is connected to the rehabilitation mode and content storage module 21 and the real-time audiovisual feedback unit, and is used to set the voice quality parameter rehabilitation mode and run the voice quality parameter rehabilitation content in real time. The real-time voice quality parameter measurement and discrimination module 23 is connected to the real-time processing unit 1 and the rehabilitation execution module. It is used to determine the voice quality parameters of the user when producing speech in real time, and to determine the voice quality disorder of the user. The real-time voice quality parameter measurement and discrimination module 23 judges the voice quality focus type based on the voice quality parameter data and the corresponding voice quality focus limit value.

[0153] Specifically, for each age group of the same sex (15 groups for males and 15 groups for females), the formula for judging the degree of damage of hyperactive voice quality function is: mean voice quality + mean voice quality * voice quality limit; the formula for judging the degree of damage of degraded voice quality function is: mean voice quality - mean voice quality * voice quality limit; based on the above formulas, the precise voice quality assessment module 21 can identify the type of voice quality disorder, specifically degraded voice quality function and hyperactive voice quality function.

[0154] Example 1: First, a profile is created based on the user's information (see profile form). Then, the user is instructed to pronounce / ae—— / as loudly as possible and maintain it for several seconds. By analyzing the user's voice quality signal, the user's fundamental frequency tremor data can be obtained. The assessed voice quality report can be found here. Figure 11 As shown, specifically, in terms of voice quality, this patient (a 45-year-old female) has neurogenic voice disorder.

[0155] File Form:

[0156]

[0157]

[0158] In terms of language:

[0159] Rear focusing, the resonant frequency F2 / i / is 1719Hz, indicating severe damage;

[0160] The front-focused image shows a resonant peak frequency F2 / u / of 956.8 Hz, indicating moderate damage.

[0161] Regarding the voice:

[0162] 1. Glottal closure:

[0163] Moderate breath noise, with a glottal noise NNE of -7.21 dB.

[0164] 2. Regularity of vocal cord vibration:

[0165] Moderately rough sound, with a fundamental frequency jitter of 1.54%.

[0166] Moderate hoarseness with slight perturbation: Shimmer 7.47%;

[0167] In response to the above-mentioned obstacles, the rehabilitation steps and implementation module 22 for neurogenic voice disorders provide suggestions and implementation methods, as follows: 1. Real-time feedback therapy, selecting treatment methods such as real-time pitch feedback training and real-time loudness feedback training; 2. Traditional treatment, selecting treatment methods such as neck relaxation training, physiological abdominal breathing training, breathing relaxation training, humming method, counting method, slow and steady exhalation method, vocal cord relaxation training, warble method, arm swing and push method, mouth opening method, musical pitch matching method, pitch gradient training, chanting method, etc.

[0168] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A smart rehabilitation system for voice quality disorders based on the ICF-RFT framework, the system structure of which includes: The real-time voice quality signal detection and processing unit (1), main control unit (2), real-time audiovisual tactile feedback unit (3), rehabilitation cloud platform (4), mobile rehabilitation learning machine (5), and mobile intelligent screening instrument (6) are characterized in that: the real-time voice quality signal detection and processing unit (1) is used to detect and process the user's voice quality signal data in real time; The main control unit (2) is connected to the real-time voice quality signal detection and processing unit (1) for the precise assessment and rehabilitation training of eight related modules, including fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , formant frequency disturbance, nasal flow, and nasooral resonance ratio. This process runs through the entire process of functional assessment, plan formulation, treatment implementation, and efficacy evaluation. The real-time audiovisual tactile feedback unit (3) is connected to the main control unit (2) and uses real-time audiovisual tactile feedback technology to display and provide feedback on the results and related information of voice quality rehabilitation in real time; the rehabilitation cloud platform (4) is connected to the main control unit (2) via the Internet for data sharing; the mobile rehabilitation learning machine (5) is connected to the rehabilitation cloud platform (4) for assigning rehabilitation tasks and popularizing rehabilitation training; the mobile intelligent screening instrument (6) is connected to the real-time voice quality signal detection and processing unit (1) for comprehensive examination, and obtains the average value of ICF functional damage through the Internet connection to the rehabilitation cloud platform (4) to provide a diversion for the voice quality accurate assessment and real-time monitoring unit (21); The main control unit (2) includes a voice quality accurate assessment and real-time monitoring unit (21), a rehabilitation mode and rehabilitation step execution unit (22), an ICF conversion unit (23), and a voice quality four-stage rehabilitation content push unit (24). The voice quality accurate assessment and real-time monitoring unit (21) performs accurate assessment and real-time monitoring of data to determine whether there is a voice quality disorder and to determine the type of voice quality disorder; the rehabilitation mode and rehabilitation step execution unit (22) sets the rehabilitation mode and executes the rehabilitation steps; the ICF conversion unit (23) performs ICF conversion on the data of accurate assessment and real-time monitoring; and the voice quality four-stage rehabilitation content push unit (24) selects the four-stage rehabilitation content to push.

2. The intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework according to claim 1, characterized in that, The real-time voice quality signal detection and processing unit (1) is used to acquire the user's voice quality signal data in real time. The voice quality data is obtained by connecting a microphone to a single-channel low-pass filter, a dual-channel low-pass filter, and / or by an electroglottogram.

3. The intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework according to claim 2, characterized in that, The main technical parameters for analyzing and controlling data error and response rate error of the input voice quality signal are as follows: a) Frequency error: ±2%; b) Voltage error: ±3%; c) Real-time response rate of fundamental frequency: ≤6ms; d) LPC real-time response rate: ≤45ms; e) Real-time spectrogram resolution: Narrowband: 12.7ms ± 4%; Midband: 12.7ms ± 4%; Broadband: 12.7ms ± 4%.

4. The intelligent rehabilitation system for voice quality disorders based on the ICF-RFT framework according to claim 1, characterized in that, The real-time voice quality signal detection and processing unit (1) includes a b3101 comprehensive inspection module: obtaining three measurement values—fundamental frequency perturbation, glottal noise energy, and amplitude perturbation—through the voice quality signal; performing ICF damage degree conversion on each objective measurement value to obtain the average voice quality ICF value; The main control unit (2) includes eight modules: fundamental frequency perturbation, amplitude perturbation, glottal noise energy, formant frequency F2 / i / , formant frequency F2 / u / , formant frequency perturbation F2f, nasal flow, and naso-oral resonance ratio NOR. Each module includes: accurate assessment, damage degree conversion and its rehabilitation content push and real-time monitoring.

Images