CN103519785A - Multi-dimensional voice disorder measuring system based on multi-dimensional real-time voice modeling and method thereof - Google Patents
Multi-dimensional voice disorder measuring system based on multi-dimensional real-time voice modeling and method thereof Download PDFInfo
- Publication number
- CN103519785A CN103519785A CN201310459854.1A CN201310459854A CN103519785A CN 103519785 A CN103519785 A CN 103519785A CN 201310459854 A CN201310459854 A CN 201310459854A CN 103519785 A CN103519785 A CN 103519785A
- Authority
- CN
- China
- Prior art keywords
- voice
- phoneme data
- real
- disorder
- segmental
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 208000011293 voice disease Diseases 0.000 title claims abstract description 79
- 238000000034 method Methods 0.000 title claims abstract description 20
- 210000001331 nose Anatomy 0.000 claims description 27
- 210000003928 nasal cavity Anatomy 0.000 claims description 11
- 238000012544 monitoring process Methods 0.000 claims description 9
- 238000001228 spectrum Methods 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 abstract description 5
- 208000027765 speech disease Diseases 0.000 abstract description 3
- 206010013952 Dysphonia Diseases 0.000 abstract 1
- 238000011897 real-time detection Methods 0.000 abstract 1
- 238000005259 measurement Methods 0.000 description 27
- 239000000463 material Substances 0.000 description 18
- 206010068319 Oropharyngeal pain Diseases 0.000 description 14
- 201000007100 Pharyngitis Diseases 0.000 description 14
- 238000012360 testing method Methods 0.000 description 14
- 238000004458 analytical method Methods 0.000 description 13
- 238000001514 detection method Methods 0.000 description 11
- 238000012549 training Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 210000000214 mouth Anatomy 0.000 description 8
- 230000000630 rising effect Effects 0.000 description 8
- 230000002159 abnormal effect Effects 0.000 description 6
- 230000004069 differentiation Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 210000005182 tip of the tongue Anatomy 0.000 description 4
- 238000012546 transfer Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 241001413866 Diaphone Species 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
Images
Landscapes
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention discloses a multi-dimensional voice disorder measuring system based on the multi-dimensional real-time voice modeling technology. The system comprises a collecting device, a calculating device and a judging device, wherein the collecting device collets phoneme data, the calculating device calculates parameter values of the phoneme data, the standard numerical range of the phoneme data is stored in the judging device, the parameter values are compared according to the standard numerical range, voice disorder does not exist if the parameter values are within the standard numerical range, the voice disorder exists if the parameter values are lower than or higher than the standard numerical range, and the type of the existed voice disorder is judged. Real-time detection processing and real-time voice multi-dimensional modeling are carried out on voice and rhinolalia, so segmental phoneme data, supersegmental phoneme data, mixed segmental phoneme data and separted segmental phoneme are evaluated, numerous pointed multi-dimensional characteristic parameters are increased, and therefore evaluation of voice disorder is more comprehensive and precise. The invention further discloses a multi-dimensional voice disorder measuring method based on the multi-dimensional real-time voice modeling technology.
Description
Technical field
The invention belongs to speech measures field, relate in particular to a kind of voice disorder multi-dimensional measuring system and method thereof based on real-time voice Conceptual Modeling technology.
Background technology
Voice disorder is the important component part of speech disorder, to the assessment of voice disorder, is therefore also the important module in speech disorder assessment.At present, the domestic assessment for voice disorder comprises subjective evaluation and two aspects of objective measurement; Subjective evaluation is mainly for specific test phonetic material, by analyzing patient, read aloud the appraisal result of fc-specific test FC phonetic material, obtain patient's the results such as continuous speech definition, and compare with the result of normal population, thereby realize the subjective evaluation to patient's phonetic function; Objective measurement comprises the measurement of supersegmental phoneme voice and the measurement of segmental phoneme voice, by analyzing patient's Speech acoustics signal, obtain the relevant parameters,acoustic of its supersegmental phoneme and segmental phoneme, and compare with the relevant parameter of normal population, thereby realize the objective measurement to patient's phonetic function.
But, be there is to defect in the objective measurement of voice disorder at present: characteristic parameter is less targetedly, can not reflect more all sidedly the problem of voice disorder supersegmental phoneme part and segmental phoneme part.Therefore, the objective measurement of carrying out voice disorder from single dimension can not reflect patient's actual speech problem comprehensively, thereby the assessment result of patient's voice disorder is impacted.
Summary of the invention
The present invention has overcome in existing voice difficulty measurement system that single dimension is measured and the less defect of characteristic parameter targetedly, has proposed a kind of voice disorder multi-dimensional measuring system and method thereof based on real-time voice Conceptual Modeling technology.
The present invention proposes a kind of voice disorder measuring system based on real-time voice Conceptual Modeling technology, comprising: harvester, it gathers phoneme data; Accountant, calculates the parameter value of described phoneme data; And judgment means, its standard figures scope by described parameter value and described phoneme data contrasts, if described parameter value, within the scope of described standard figures, represents not exist voice disorder; If described parameter value, below or above described standard figures scope, represents the voice disorder type that exists voice disorder judgement to exist.
In the voice disorder measuring system based on real-time voice Conceptual Modeling technology that the present invention proposes, described harvester comprises omnidirectional microphone harvester separated with mouth and nose.
In the voice disorder measuring system based on real-time voice Conceptual Modeling technology that the present invention proposes, described phoneme data comprise segmental phoneme data, supersegmental phoneme data, hybrid segmental phoneme data and separate type segmental phoneme data.
In the voice disorder measuring system based on real-time voice Conceptual Modeling technology that the present invention proposes, described segmental phoneme data comprise the related data with speech intelligibility.
In the voice disorder measuring system based on real-time voice Conceptual Modeling technology that the present invention proposes, described supersegmental phoneme data comprise the related data with tonal variations rate.
In the voice disorder measuring system based on real-time voice Conceptual Modeling technology that the present invention proposes, described hybrid segmental phoneme data comprise pure and impure signal to noise ratio rate, articulation type ratio, place of articulation ratio, cue and the time ratio of supplying gas.
In the voice disorder measuring system based on real-time voice Conceptual Modeling technology that the present invention proposes, described separate type segmental phoneme data comprise nose flow, nasal cavity linear prediction spectrum and nasal cavity power spectrum.
In the voice disorder measuring system based on real-time voice Conceptual Modeling technology that the present invention proposes, further comprise supervising device, the rehabilitation efficacy of the front and back difference monitoring voice disorder of the described phoneme data that it obtains according to multi collect.
The invention allows for a kind of voice disorder measuring method based on real-time voice Conceptual Modeling technology, comprise the following steps: step 1: gather phoneme data; Step 2: the parameter value that calculates described phoneme data; Step 3: contrast described parameter value according to described standard figures scope, if described parameter value, within the scope of described standard figures, represents not exist voice disorder; If described parameter value, below or above described standard figures scope, represents the voice disorder type that exists voice disorder judgement to exist.
In the voice disorder measuring method based on real-time voice Conceptual Modeling technology that the present invention proposes, further comprise step 4: according to the rehabilitation efficacy of the front and back difference monitoring voice disorder of the described phoneme data that repeatedly record.
The present invention is based on real-time voice Conceptual Modeling technology, broken through the restriction of carrying out voice disorder objective measurement from single dimension, added the detection analysis to separate type rhinophonia signal, from segmental phoneme, supersegmental phoneme, hybrid segmental phoneme and a plurality of dimensions of separate type segmental phoneme, analyze, extract the characteristic parameter of a plurality of dimensions for the objective measurement of voice disorder, make voice disorder user's assessment more effective with training.
Utilization of the present invention realizes speech rehabilitation effect monitoring based on real-time voice Conceptual Modeling technology, comprise the check analysis of measurement data and carry out Single subject experiment, thereby the periodic measurement result of utilizing multidimensional characteristic parameter in voice disorder during rehabilitation training is monitored the effect of its training, make rehabilitation training strategy can access feedback and adjustment timely.
The present invention has set up a comprehensive pointed voice disorder multidimensional measure method again comprehensively based on real-time voice Conceptual Modeling technology, scattered objective measurement Technology Integration is got up, and added the monitoring of rehabilitation efficacy, form a set of complete voice disorder composite measurement evaluation system, for voice disorder user's rehabilitation lays the foundation.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of appraising of speech intelligibility;
Fig. 2 is the result figure of speech intelligibility test parameter;
Fig. 3 is the result figure of tonal variations rate parameter;
Fig. 4 is the oscillogram of pure and impure signal to noise ratio rate;
Fig. 5 is the result figure of articulation type ratio;
Fig. 6 is the formant trajectory figure of test cue;
Fig. 7 is the oscillogram that detects nose flow;
Fig. 8 is the result figure of monitoring rehabilitation efficacy;
Fig. 9 is the flow chart of the voice disorder multidimensional measure method based on real-time voice Conceptual Modeling technology;
Figure 10 is the schematic diagram of voice disorder multi-dimensional measuring system;
Figure 11 is the flow chart that obtains phoneme data.
The specific embodiment
In conjunction with following specific embodiments and the drawings, the present invention is described in further detail.Implement process of the present invention, condition, experimental technique etc., except the content of mentioning specially below, be universal knowledege and the common practise of this area, the present invention is not particularly limited content.
What Figure 10 showed is voice disorder multi-dimensional measuring system of the present invention, comprising harvester 1, accountant 2 and judgment means 3.
Harvester 1 comprises omnidirectional microphone, the separated harvester of mouth and nose.Harvester 1 adopts voice Conceptual Modeling technology, by omnidirectional microphone, the separated harvester of mouth and nose, gathers phoneme data respectively from voice, two dimensions of rhinophonia.
Judgment means 3 is connected with accountant 2, stores the standard figures scope that phoneme data are relevant in judgment means 3.The regime values of each phoneme parameter of each age level different crowd of standard figures Range Representation is interval, judgment means 3 judges that according to standard figures scope whether the parameter value of phoneme data is abnormal, if parameter value within the scope of standard figures, represents that these phoneme data are normal; If parameter value, higher or lower than standard figures scope, represents to exist voice disorder.
Further, voice disorder multi-dimensional measuring system of the present invention also comprises supervising device 4.Supervising device 4 is by the difference between a plurality of phoneme data that repeatedly measure, thus the rehabilitation efficacy of monitoring before and after voice rehabilitation training.
What Figure 11 showed is the flow process that obtains various phoneme data, and phoneme data comprise segmental phoneme data, supersegmental phoneme data, hybrid segmental phoneme data and separate type segmental phoneme data.
Segmental phoneme packet is containing the data relevant to speech intelligibility.The parameter value of segmental phoneme data comprises: word inteligibility, sentence definition, continuous speech definition etc., its integral body has reflected user's speech articulation.By the parameter value of a plurality of segmental phonemes is compared with corresponding critical field, can carry out many-side assessment to speech intelligibility.
Supersegmental phoneme packet is containing the related data with tonal variations rate.For supersegmental phoneme, tonal variations rate refers to the fundamental frequency standard deviation of voice signal.The measurement of tonal variations rate comprises rising tune, falling tone, rising-falling tone energy force measurement, adopt respectively with after rising tune, falling tone and the pronunciation of rising-falling tone form, the value of the fundamental frequency standard deviation of the voice signal obtaining, has reflected the intonation changing capability of continuous speech, thereby reflects the fluency of voice.
Hybrid segmental phoneme data comprise pure and impure signal to noise ratio rate, articulation type ratio, place of articulation ratio, cue and the time ratio etc. of supplying gas.Pure and impure signal to noise ratio rate refers to the ratio of the duration of sore throat relieving and the duration of voiced sound in input voice segments, has reflected the transfer capability of pure and impure sound in continuous speech.Articulation type ratio refers to the ratio of the duration of certain articulation type and the duration of whole voice segments in input voice segments, has reflected the order of accuarcy of articulation type in continuous speech.Place of articulation ratio refers to the ratio of the duration of certain place of articulation and the duration of whole voice segments in input voice segments, has reflected the order of accuarcy of place of articulation in continuous speech.Cue refers to that in input voice segments, sore throat relieving, to the formant trajectory of voiced sound transition, has reflected that the order of accuarcy that in continuous speech, sore throat relieving pronounces and sore throat relieving are to the order of accuarcy of voiced sound transition.The time ratio of supplying gas refers to the ratio of aspirated sound duration and the duration of whole voice segments, has reflected aspirated order of accuarcy in continuous speech.
Separate type segmental phoneme packet containing to nose flow, data that nasal cavity linear prediction spectrum is relevant with nasal cavity power spectrum.Nose flow refers to the ratio of nasal cavity sound pressure level and output sound pressure level (oral cavity sound pressure level and nasal cavity sound pressure level sum), has reflected the ability of rhinophonia pronunciation in continuous speech.Nasal cavity linear prediction spectrum is the spectrogram of formant frequency, bandwidth and the amplitude that can observe rhinophonia, the order of accuarcy of vocal tract shape while having reflected rhinophonia pronunciation in continuous speech.Nasal cavity power spectrum is the spectrogram that can observe rhinophonia harmonic structure and concentration of energy region, and while having reflected rhinophonia pronunciation in continuous speech, energy is with the characteristic of frequency change.
Above-mentioned segmental phoneme data, supersegmental phoneme data, hybrid segmental phoneme data and separate type segmental phoneme data, wherein each phoneme data is indicated respectively different voice disorder types.What Fig. 9 showed is the flow process that the present invention is based on the voice disorder multidimensional measure method of real-time voice Conceptual Modeling technology, the voice disorder multidimensional measure method that the present invention is based on real-time voice Conceptual Modeling is by the multidimensional measure rounded analysis to phoneme data, judges whether to exist the type of the voice disorder of voice disorder and existence.
Embodiment 1: speech intelligibility is measured
In the present embodiment, adopt mike to record by linearity input the voice that user reads aloud, to gather the segmental phoneme data of these voice.In test phonetic material, be provided with some appraising point, can appraise for speech intelligibility, as shown in Figure 1.After appraising in full, in accountant 2 these voice of calculating, correct number accounts for the percentage ratio of total appraising point number, and this percentage ratio is the parameter value of segmental phoneme data, comprising: the parameters such as word inteligibility, sentence definition, continuous speech definition, as shown in Figure 2.Judgment means 3, according to user's information, is compared user's parameters value with its age, the corresponding standard figures scope of sex, if parameter value within standard figures scope, illustrates user's speech intelligibility ability, be normal; If parameter value is lower than standard figures scope, there is voice disorder in explanation, and this obstacle is speech intelligibility obstacle.
Embodiment 2: voice fluency is measured
Test phonetic material is respectively and adopts with rising tune, falling tone and rising-falling tone form, the voice that harvester 1 collection user sends according to test phonetic material are to obtain supersegmental phoneme data, and accountant 2 calculating that these supersegmental phoneme data are carried out to fundamental frequency standard deviation obtain the parameter value of its tonal variations rate.
The present embodiment is measured as example with tone rising tune, test phonetic material is from minimum tone, to be raised to vowel/i--i/ that maximum tone is sent out as far as possible with rising tune form, user pronounces according to this test phonetic material, and harvester 1 obtains the phoneme data in this voice signal.Preferably, user's voice are carried out the pretreatment such as pre-filtering, amplification, cancellation of DC offset and normalization in the gatherer process of harvester 1, to improve the signal to noise ratio of voice signal, improve certainty of measurement.Then 2 pairs of phoneme data of accountant are carried out pure and impure sound differentiation, and calculate the fundamental frequency value of voiced sound part and the standard deviation of analyzing fundamental frequency value.Judgment means 3 is by user's fundamental frequency standard deviation and corresponding standard figures scope comparison, if its fundamental frequency standard deviation within standard figures scope, illustrates that user's tone rising tune ability is normal; Otherwise its fundamental frequency standard deviation is excessive or too smallly all illustrate that user's tone rising tune ability exists obstacle, has fluency obstacle thereby disclose.
For example, Fig. 3 has shown the tonal variations rate value of a tone rising tune ability patients with abnormal, and the standard deviation of its fundamental frequency value is 9.24Hz.Because normal person's standard figures scope is 30Hz～70Hz, and this patient compares with normal person, and its value is far smaller than term of reference, so judgement exists voice disorder, is specially and exists tonal variations rate abnormal.
Tone falling tone and rising-falling tone energy force measurement are the same, and three's result has reflected different tonal variations abilities, abnormal as long as one of them value exists, and all illustrate that user's tonal variations ability exists obstacle, and the fluency of voice exists abnormal.
With the following Examples 3 to embodiment 7, after hybrid segmental phoneme characteristic parameter is measured, user's the box-like segmental phoneme data analysis of different blended, differentiates the order of severity of the hybrid segmental phoneme obstacle of user.
Embodiment 3: pure and impure sound transfer capability is measured
Pure and impure signal to noise ratio rate mainly refers to the ratio of the duration of sore throat relieving and the duration of voiced sound in voice segments.Harvester 1 gathers hybrid segmental phoneme data from user's voice segments, 2 pairs of hybrid segmental phoneme data of accountant are carried out pure and impure sound differentiation to obtain the duration of its sore throat relieving part and voiced sound part, and by the former duration the duration divided by the latter, thereby obtain a kind of parameter value of hybrid segmental phoneme, i.e. pure and impure sound rate value.
The phonetic material " handcuff " that not only comprises sore throat relieving but also comprise voiced sound that this enforcement take that user sends is (kao4) example.By harvester 1, gather after its phoneme data, accountant 2 carries out pure and impure sound differentiation to it, detects also mark of its sore throat relieving part and voiced sound part, then calculates respectively total duration of its sore throat relieving part and voiced sound part and compares, final its pure and impure sound rate value that obtains, as shown in Figure 4.Judgment means 3 is compared user's pure and impure sound rate value with standard figures scope, if within its pure and impure sound rate value drops on standard figures scope, illustrate that user's pure and impure sound transfer capability is normal; Otherwise there is voice disorder in excessive or too small all explanation of its pure and impure sound rate value, is specially pure and impure sound transfer capability and has obstacle.
Embodiment 4: articulation type is measured
Articulation type ratio mainly refers to the ratio of the duration of certain articulation type and the duration of whole voice segments in input voice segments, is articulation type rate value.By voice being carried out to the analysis of sound-type and the calculating on duration and border obtains the duration of its specific articulation type and the duration of whole voice segments.Wherein the analysis of sound-type comprises the detection of the articulation types such as resonant detection, vowel detection, voiced consonant's detection, rhinophonia detection, turbid fricative detection, plosive and sore throat relieving detection.
The present embodiment with carry out plosive articulation type ratio be measured as example, user is sent to the phonetic material " bag (bao1)/throwing (pao1) " that comprises plosive, harvester 1 gather in user speech material hybrid segmental phoneme data, 2 pairs of hybrid segmental phoneme data of accountant are carried out pure and impure sound differentiation, comprise and carry out resonant detection, vowel detection and tenuis detect to obtain the analysis result of sound-type, analysis result is as Fig. 5, accountant 2 carries out sound-type mark again and border is determined, calculate respectively the duration of its tenuis part and total duration of whole phonetic material and compare, final its plosive rate value that obtains, it is a kind of parameter value of hybrid segmental phoneme data.
Judgment means 3 is compared user's plosive rate value with standard figures scope, if within its plosive rate value drops on standard figures scope, the articulation type that user is described is normal; Otherwise there is voice disorder in excessive or too small all explanation of its plosive rate value, is specially articulation type and has obstacle.
Embodiment 5: the mode of supplying gas ratio measure
The mode ratio of supplying gas mainly refers to the ratio of aspirated sound duration and the duration of whole voice segments.The hybrid segmental phoneme data that harvester 1 gathers in user voice signal, it is carried out to the analysis of sound-type with accountant 2 and the calculating on duration and border obtains the duration of the mode of supplying gas and the duration of whole voice segments, thereby obtain supplying gas mode ratio, be a kind of parameter value of hybrid segmental phoneme data.
For example, the phonetic material that comprises aspirated stop that user is sent out " throw/handcuff " gathers, and accountant 2 carries out pure and impure sound differentiation to it, then carry out that sore throat relieving detects and aspirated sound to detect to obtain analysis result and the border of aspirated stop definite; Then calculate respectively the duration of aspirated stop and total duration of whole word, the duration by the former duration divided by the latter, its result is the mode rate value of supplying gas.
Judgment means 3 is compared user's the mode of supplying gas rate value with standard figures scope, within mode rate value drops on standard figures scope if it is supplied gas, illustrate that user's the mode of supplying gas ratio is normal; Otherwise there is voice disorder in excessive or too small all explanation of its mode rate value of supplying gas, is specially the mode of supplying gas and has obstacle.
Embodiment 6: cue is measured
The phonetic material that comprises sore throat relieving that user is sent out of take in the present embodiment " is taken " as example, Fig. 6 shows is the oscillogram of the diaphone bit data that collects by harvester 1, the formant trace diagram that wherein the second width coordinate diagram is corresponding sonogram, first three formant that it comprises voice signal, is followed successively by first, second, and third formant from down to up.Transverse axis is the time, and the longitudinal axis is corresponding resonance peak.Because the second formant mainly reflects structure and the shape in oral cavity, so reflect place of articulation with the second formant.
In the present embodiment, hybrid segmental phoneme data in the phonetic material of harvester 1 collection sore throat relieving and first mixture of tones, because sore throat relieving does not have formant, so accountant 2 is chosen the threshold value of the second formant as cue value, be a kind of parameter value of hybrid segmental phoneme data.In Fig. 6, cue value is the starting point frequency values of second horizontal line: 801Hz, the intensity level that the 3rd width coordinate diagram is corresponding formant.Because the standard figures scope of the cue value of " taking " is 1200Hz～2400Hz, and this user compares with normal standard figures scope, and its value is far smaller than standard figures, so judgement exists voice disorder, is specially cue abnormal.
Embodiment 7: place of articulation ratio measure
Place of articulation ratio mainly refers to the ratio of the duration of certain place of articulation and the duration of whole voice segments in input voice segments.Hybrid segmental phoneme data in harvester 1 collection user's voice signal, accountant 2 carries out the analysis of formant and the definite duration of its specific place of articulation and duration of whole voice segments of obtaining on place of articulation and border thereof to it, duration by the former duration divided by the latter, its result is place of articulation rate value, is a kind of parameter value of hybrid segmental phoneme data.
For example, user sends and comprises the phonetic material that the tip of the tongue plosive " is thrown/cover ", the hybrid segmental phoneme data that harvester 1 gathers in phonetic material, 2 pairs of hybrid segmental phoneme data of accountant are carried out formant extraction and trace obtains the frequency values of each section of formant and the frequency values of starting point thereof, and determine position and the border thereof of the tip of the tongue plosive according to it; Then calculate respectively the duration of the tip of the tongue plosive and total duration of whole word and compare, finally obtaining the place of articulation rate value of the tip of the tongue plosive, being a kind of parameter value of hybrid segmental phoneme data.
Judgment means 3 is compared place of articulation rate value with standard figures scope, if within its place of articulation rate value drops on standard figures scope, the place of articulation ratio that user is described is normal; Otherwise there is voice disorder in excessive or too small all explanation of its place of articulation rate value, is specially place of articulation and has obstacle.
With the following Examples 8, after hybrid segmental phoneme characteristic parameter is measured, user's different separate type segmental phoneme data analysis, differentiate the order of severity of user's separate type segmental phoneme obstacle.
Embodiment 8: nose flow measurement
Figure 7 shows that the oscillogram of the detection nose flow of an embodiment, wherein comprise two windows of main window and secondary window.Wherein three of main window oscillograms are respectively from top to bottom: nose flow diagram, nose voice signal sound wave, oral area voice signal sound wave.Secondary window is respectively nose voice signal sound wave, oral area voice signal sound wave from top to bottom, and their transverse axis is the time.Nose flow diagram is corresponding to sonogram below, and the nose flow value of different time place sound wave is different.
The phonetic material of above-mentioned parameter can be selected identical or different phonetic material as required.As user follows rhinophonia hypofunction, the voice content that selection comprises a large amount of rhinophonia is as the mouth and nose separate type acoustical signal of input, and analysis and measurement that it is carried out to above-mentioned each parameter, finally obtains the measurement result of its separate type segmental phoneme.
Supervising device 4 is mainly to utilize above-mentioned parameter in voice disorder during rehabilitation training to carry out periodic measurement with monitoring rehabilitation efficacy, monitors the effect of its training, to adjust in time training strategy; Mainly utilize check analysis and single tested technology to follow the trail of measurement.What Fig. 8 showed is the schematic diagram of the data result that in an embodiment, supervising device is measured, and wherein the check analysis of parameter periodic measurement result is comprised to autocorrelation test, significance test and regression analysis.By autocorrelation test, carry out the effectiveness of confirmatory measurement result, cross the diversity that significance test is verified intervention training fore-and-aft survey result, by regression analysis, carry out the variation tendency of confirmatory measurement result.
Protection content of the present invention is not limited to above embodiment.Do not deviating under the spirit and scope of inventive concept, variation and advantage that those skilled in the art can expect are all included in the present invention, and take appending claims as protection domain.
Claims (10)
1. the voice disorder measuring system based on real-time voice Conceptual Modeling technology, is characterized in that, comprising:
Harvester (1), it gathers phoneme data;
Accountant (2), calculates the parameter value of described phoneme data;
Judgment means (3), its standard figures scope by described parameter value and described phoneme data contrasts, if described parameter value, within the scope of described standard figures, represents not exist voice disorder; If described parameter value, below or above described standard figures scope, represents the voice disorder type that exists voice disorder judgement to exist.
2. the voice disorder measuring system based on real-time voice Conceptual Modeling technology according to claim 1, is characterized in that, described harvester (1) comprises omnidirectional microphone harvester separated with mouth and nose.
3. the voice disorder measuring system based on real-time voice Conceptual Modeling technology as claimed in claim 1, is characterized in that, described phoneme data comprise segmental phoneme data, supersegmental phoneme data, hybrid segmental phoneme data and separate type segmental phoneme data.
4. the voice disorder measuring system based on real-time voice Conceptual Modeling technology as claimed in claim 3, is characterized in that, described segmental phoneme data comprise speech intelligibility data.
5. the voice disorder measuring system based on real-time voice Conceptual Modeling technology as claimed in claim 3, is characterized in that, described supersegmental phoneme data comprise tonal variations rate data.
6. the voice disorder measuring system based on real-time voice Conceptual Modeling technology as claimed in claim 3, is characterized in that, described hybrid segmental phoneme data comprise pure and impure signal to noise ratio rate, articulation type ratio, place of articulation ratio, cue and the time ratio of supplying gas.
7. the voice disorder measuring system based on real-time voice Conceptual Modeling technology as claimed in claim 3, is characterized in that, described separate type segmental phoneme data comprise nose flow, nasal cavity linear prediction spectrum and nasal cavity power spectrum.
8. the voice disorder measuring system based on real-time voice Conceptual Modeling technology according to claim 1, it is characterized in that, further comprise supervising device (4), the rehabilitation efficacy of the front and back difference monitoring voice disorder of the described phoneme data that it obtains according to multi collect.
9. the voice disorder measuring method based on real-time voice Conceptual Modeling technology, is characterized in that, comprises the following steps:
Step 1: gather phoneme data;
Step 2: the parameter value that calculates described phoneme data;
Step 3: contrast described parameter value according to described standard figures scope, if described parameter value, within the scope of described standard figures, represents not exist voice disorder; If described parameter value, below or above described standard figures scope, represents the voice disorder type that exists voice disorder judgement to exist.
10. the voice disorder measuring method based on real-time voice Conceptual Modeling technology as claimed in claim 9, is characterized in that, further comprises step 4: according to the rehabilitation efficacy of the front and back difference monitoring voice disorder of the described phoneme data that repeatedly record.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310459854.1A CN103519785B (en) | 2013-09-24 | 2013-09-24 | Based on voice disorder multi-dimensional measuring system and the method thereof of real-time voice Conceptual Modeling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310459854.1A CN103519785B (en) | 2013-09-24 | 2013-09-24 | Based on voice disorder multi-dimensional measuring system and the method thereof of real-time voice Conceptual Modeling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103519785A true CN103519785A (en) | 2014-01-22 |
| CN103519785B CN103519785B (en) | 2016-01-27 |
Family
ID=49922403
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310459854.1A Active CN103519785B (en) | 2013-09-24 | 2013-09-24 | Based on voice disorder multi-dimensional measuring system and the method thereof of real-time voice Conceptual Modeling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103519785B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110870762A (en) * | 2019-06-27 | 2020-03-10 | 上海慧敏医疗器械有限公司 | Nasal cavity resonance rehabilitation instrument and method based on nasal flow measurement and audio-visual feedback technology |
| CN110876609A (en) * | 2019-07-01 | 2020-03-13 | 上海慧敏医疗器械有限公司 | Voice treatment instrument and method for frequency band energy concentration rate measurement and audio-visual feedback |
| CN111315302A (en) * | 2017-11-02 | 2020-06-19 | 松下知识产权经营株式会社 | Cognitive function assessment device, cognitive function assessment system, cognitive function assessment method and program |
| WO2020192230A1 (en) * | 2019-03-25 | 2020-10-01 | 中国科学院深圳先进技术研究院 | Array surface electromyography- based pronunciation function evaluation system |
| CN116473521A (en) * | 2023-06-21 | 2023-07-25 | 四川大学华西医院 | Voice frequency spectrum identification method and system for suspected cyprocoytenoid dislocation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102824162A (en) * | 2012-08-22 | 2012-12-19 | 泰亿格电子（上海）有限公司 | Nasal voice measuring instrument and measuring method thereof |
| CN102920433A (en) * | 2012-10-23 | 2013-02-13 | 泰亿格电子（上海）有限公司 | Rehabilitation system and method based on real-time audio-visual feedback and promotion technology for speech resonance |
| CN102920435A (en) * | 2012-10-23 | 2013-02-13 | 泰亿格电子（上海）有限公司 | Rehabilitation system and method based on real-time audio-visual feedback and promotion technology for speech production |
-
2013
- 2013-09-24 CN CN201310459854.1A patent/CN103519785B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102824162A (en) * | 2012-08-22 | 2012-12-19 | 泰亿格电子（上海）有限公司 | Nasal voice measuring instrument and measuring method thereof |
| CN102920433A (en) * | 2012-10-23 | 2013-02-13 | 泰亿格电子（上海）有限公司 | Rehabilitation system and method based on real-time audio-visual feedback and promotion technology for speech resonance |
| CN102920435A (en) * | 2012-10-23 | 2013-02-13 | 泰亿格电子（上海）有限公司 | Rehabilitation system and method based on real-time audio-visual feedback and promotion technology for speech production |
Non-Patent Citations (4)
| Title |
|---|
| 刘巧云 等: "《人工耳蜗儿童、助听器儿童与健听儿童音位对比识别能力比较研究》", 《中国特殊教育》, no. 2, 31 December 2011 (2011-12-31) * |
| 卢海丹 等: "不同测试语料下3～6岁健听与听障儿童音调比较研究", 《听力学及言语疾病杂志》, vol. 19, no. 3, 31 December 2011 (2011-12-31) * |
| 李欢: "构音障碍评估研究述评", 《中国特殊教育》, no. 6, 31 December 2010 (2010-12-31) * |
| 黄昭鸣: "言语障碍的定量测量及治疗对策", 《中华中医药学会耳鼻咽喉科分会第十二次学术研讨会暨嗓音言语听力医学专题学术研讨会论文集》, 31 December 2005 (2005-12-31) * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111315302A (en) * | 2017-11-02 | 2020-06-19 | 松下知识产权经营株式会社 | Cognitive function assessment device, cognitive function assessment system, cognitive function assessment method and program |
| CN111315302B (en) * | 2017-11-02 | 2023-04-25 | 松下知识产权经营株式会社 | Cognitive function evaluation device, cognitive function evaluation system, cognitive function evaluation method and program recording medium |
| WO2020192230A1 (en) * | 2019-03-25 | 2020-10-01 | 中国科学院深圳先进技术研究院 | Array surface electromyography- based pronunciation function evaluation system |
| CN110870762A (en) * | 2019-06-27 | 2020-03-10 | 上海慧敏医疗器械有限公司 | Nasal cavity resonance rehabilitation instrument and method based on nasal flow measurement and audio-visual feedback technology |
| CN110870762B (en) * | 2019-06-27 | 2022-03-29 | 上海慧敏医疗器械有限公司 | Nasal cavity resonance rehabilitation instrument and method based on nasal flow measurement and audio-visual feedback technology |
| CN110876609A (en) * | 2019-07-01 | 2020-03-13 | 上海慧敏医疗器械有限公司 | Voice treatment instrument and method for frequency band energy concentration rate measurement and audio-visual feedback |
| CN116473521A (en) * | 2023-06-21 | 2023-07-25 | 四川大学华西医院 | Voice frequency spectrum identification method and system for suspected cyprocoytenoid dislocation |
| CN116473521B (en) * | 2023-06-21 | 2023-08-18 | 四川大学华西医院 | Voice frequency spectrum identification method and system for suspected cyprocoytenoid dislocation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103519785B (en) | 2016-01-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Matos et al. | Detection of cough signals in continuous audio recordings using hidden Markov models | |
| Silva et al. | Jitter estimation algorithms for detection of pathological voices | |
| US20180214061A1 (en) | Systems for speech-based assessment of a patient's state-of-mind | |
| Lee et al. | An investigation of vocal tract characteristics for acoustic discrimination of pathological voices | |
| AU2013274940B2 (en) | Cepstral separation difference | |
| CN103405217B (en) | System and method for multi-dimensional measurement of dysarthria based on real-time articulation modeling technology | |
| CN103505183B (en) | Based on the speech disorder multi-dimensional measuring system of real-time verbal Conceptual Modeling | |
| EP3954278A1 (en) | Apnea monitoring method and device | |
| CN103519785B (en) | Based on voice disorder multi-dimensional measuring system and the method thereof of real-time voice Conceptual Modeling | |
| US20110178799A1 (en) | Methods and systems for identifying speech sounds using multi-dimensional analysis | |
| Fletcher et al. | Predicting intelligibility gains in dysarthria through automated speech feature analysis | |
| Song et al. | AHI estimation of OSAHS patients based on snoring classification and fusion model | |
| Taib et al. | Voice feature analysis for early detection of voice disability in children | |
| Dubey et al. | Pitch-Adaptive Front-end Feature for Hypernasality Detection. | |
| CN104622432A (en) | Sleep snore monitoring method and system based on bass ratio | |
| CN106782616A (en) | A kind of method that respiratory tract is detected by voice analysis | |
| Adiga et al. | Zero Frequency Filter Based Analysis of Voice Disorders. | |
| CN113948109A (en) | A system for recognizing physiological phenomena based on sound | |
| Lester-Smith et al. | Characterization of source-filter interactions in vocal vibrato using a neck-surface vibration sensor: a pilot study | |
| Reilly et al. | Voice Pathology Assessment Based on a Dialogue System and Speech Analysis. | |
| Whitehill et al. | Instrumental analysis of resonance in speech impairment | |
| CN110870762B (en) | Nasal cavity resonance rehabilitation instrument and method based on nasal flow measurement and audio-visual feedback technology | |
| Sahoo et al. | Analyzing the vocal tract characteristics for out-of-breath speech | |
| CN103505313B (en) | Speech disorder based on real-time rhythm Conceptual Modeling reads interfering system and method thereof again | |
| CN110870765A (en) | Voice treatment instrument and method adopting glottis closing real-time measurement and audio-visual feedback technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| CB02 | Change of applicant information |
Address after: 201611 Shanghai Songjiang District Chedun town car No. 180 Jia Building 2 5A Applicant after: SHANGHAI TIGER REHABILITATION THERAPY SCIENCE AND TECHNOLOGY Co.,Ltd. Address before: 201611 Shanghai city Songjiang District students Pioneering Park No. 228 Applicant before: Tiger Electronics (Shanghai) Co.,Ltd. |
|
| CB03 | Change of inventor or designer information |
Inventor after: Gao Dong Inventor after: Huang Zhaoming Inventor after: Li Ning Inventor after: Wang Hengmin Inventor after: Mo Sixia Inventor before: Huang Zhaoming Inventor before: Li Ning |
|
| COR | Change of bibliographic data | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20151224 Address after: 200000 Shanghai city Yangpu District State Road No. 1688 Bay Valley Science and Technology Park A5 Building 9 floor Applicant after: SHANGHAI TAIYIGE REHABILITATION SERVICE Co.,Ltd. Address before: 201611 Shanghai Songjiang District Chedun town car No. 180 Jia Building 2 5A Applicant before: SHANGHAI TIGER REHABILITATION THERAPY SCIENCE AND TECHNOLOGY Co.,Ltd. |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |