CN106157212A - A kind of dysphonia Chinese appraisal procedure based on EMA - Google Patents
A kind of dysphonia Chinese appraisal procedure based on EMA Download PDFInfo
- Publication number
- CN106157212A CN106157212A CN201610521815.3A CN201610521815A CN106157212A CN 106157212 A CN106157212 A CN 106157212A CN 201610521815 A CN201610521815 A CN 201610521815A CN 106157212 A CN106157212 A CN 106157212A
- Authority
- CN
- China
- Prior art keywords
- patient
- testing material
- dysphonia
- normal person
- tongue
- 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.)
- Pending
Links
- 206010013952 Dysphonia Diseases 0.000 title claims abstract description 67
- 238000000034 method Methods 0.000 title claims abstract description 17
- 239000000463 material Substances 0.000 claims abstract description 147
- 238000012360 testing method Methods 0.000 claims abstract description 147
- 210000005182 tip of the tongue Anatomy 0.000 claims abstract description 72
- 230000007547 defect Effects 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims abstract description 4
- 150000001875 compounds Chemical class 0.000 claims description 34
- 230000003068 static effect Effects 0.000 claims description 18
- 208000011293 voice disease Diseases 0.000 claims description 6
- 230000007704 transition Effects 0.000 claims description 5
- 241001672694 Citrus reticulata Species 0.000 claims description 3
- XXXSILNSXNPGKG-ZHACJKMWSA-N Crotoxyphos Chemical compound COP(=O)(OC)O\C(C)=C\C(=O)OC(C)C1=CC=CC=C1 XXXSILNSXNPGKG-ZHACJKMWSA-N 0.000 claims description 3
- 239000005367 kimax Substances 0.000 claims description 3
- 239000005364 simax Substances 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 abstract description 3
- 230000001575 pathological effect Effects 0.000 abstract description 2
- 208000018737 Parkinson disease Diseases 0.000 description 3
- 201000006417 multiple sclerosis Diseases 0.000 description 2
- 210000000056 organ Anatomy 0.000 description 2
- 230000035479 physiological effects, processes and functions Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000005236 sound signal Effects 0.000 description 2
- 206010013887 Dysarthria Diseases 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 208000035475 disorder Diseases 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000003292 glue Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000001771 impaired effect Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 210000001584 soft palate Anatomy 0.000 description 1
- 230000000153 supplemental effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/22—Social work or social welfare, e.g. community support activities or counselling services
Landscapes
- Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Child & Adolescent Psychology (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The present invention relates to pronunciation evaluation technical field, be based particularly on the dysphonia assessment technology field of EMA.A kind of dysphonia Chinese appraisal procedure based on EMA, type according to different dysphonia determines testing material, pronounce the slope of formant trajectory equation for parameter Criterion data base with normal person's the tip of the tongue Euclidean distance, normal person's lips folding distance, normal person's duration, normal person, with the pronounce slope of formant trajectory equation of patient's the tip of the tongue Euclidean distance, patient's lips folding distance, patient's duration, patient as reduced parameter, fuzzy membership functions concept is selected to judge the defect level of dysphonia patient.The present invention combines kinetics and acoustic information, it is possible to more comprehensively assess dysphonia patient, provides theoretical basis and technical support for pathological study.
Description
Technical field
The present invention relates to pronunciation evaluation technical field, be based particularly on the dysphonia assessment technology field of EMA.
Background technology
The assessment of dysphonia is to be analyzed by acoustic signal mostly, using normal person as referential, studies its difference also
Evaluate.Common method has formant to extract the degree of accuracy contrast of contrast, duration contrast and vowel-consonant.Tested mother tongue is many
For English, the pronunciation research to Chinese Chinese is less.
Author Michal Novotn et al. is at document " Automatic Evaluation of Articulatory
Disorders in Parkinson ' s Disease " in propose an assessment Parkinsonian and pronounce the method for defect,
The method is automatically to be assessed by acoustic method based on pronunciation character.This experiment recruit 24 Parkinsonians and 22 of the same age
Normal person as reference group, it is desirable to tested quickly repetition reads aloud syllable/pa/ ,/ta/ ,/ka/.It is used for describing the feature of pronunciation
Including tonequality, throat's degrees of coordination, sound channel motion, the motion of the accuracy of consonant articulation, tongue, degree of engagement and duration of speaking, this
A little features are also by the factor as assessment.With support vector cassification algorithm based on pronunciation character distinguish Parkinsonian and
Normal person, this detecting algorithm rate of accuracy reached is to 80%.First tested audio signal is labeled as initial point of articulation (initial
Burst), vowel starting point (vowel onset) and halt (occlusion), then commented by above-mentioned six pronunciation characters
Estimate the grade of pronunciation defect.The method can evaluate the defect rank of dysphonia patient, but assessment reference factor is from
The analysis of audio signal, does not has the dynamic information that actual tongue moves, so appraisal procedure is the most comprehensive.
Author Kris Tjaden et al. is at document " Vowel Acoustics in Parkinson ' s Disease and
Multiple Sclerosis: Comparison of Clear, Loud, and Slow Speaking Conditions》
In with normal artificial reference, compared for Parkinsonian and multiple sclerosis patients from definition, loudness, slow degree
Vowel articulation, finally wishes to find the lifting intelligibility of speech from comparative study, increase the tip of the tongue displacement, raising tongue movement velocity
Therapeutic Method.Literary composition is mentioned and causes dysarthric main cause to have the following aspects: be the vowel of deformation, inaccurate auxiliary
Sound, not accurate and degree of irregularity.But being only acoustic signal to be extracted formant be analyzed, contrast condition is more single.
Author Vincent Martel Sauvageau et al. is at document " Impact of the LSVT on vowel
Articulation and coarticulation in Parkinson ' s disease " use equation of locus (locus
Equation) intelligibility of pronunciation is measured.Equation of locus describes pronunciation the second formant starting point and the line of midpoint relation
Property model, the pronunciation situation of enunciator can effectively be assessed by this model.But only with being estimated ignoring to acoustic features
The defect of patient itself.
Summary of the invention
The technical problem to be solved is: the how kinematics information to dysphonia patient (dysarthria)
Carry out the pronunciation situation of comparative evaluation impaired patients with normal person (healthy controls) with acoustic information simultaneously.
The technical solution adopted in the present invention is: a kind of dysphonia Chinese appraisal procedure based on EMA, according to following step
Suddenly carry out:
Step one, type according to different dysphonia determine testing material, and testing material is one or more, each test
Language material is the standard according to the combination of Chinese Pin Yin pseudonym simple or compound vowel of a Chinese syllable, an initial consonant and all simple or compound vowel of a Chinese syllable that can pronounce with the combination of this initial consonant
All combining forms, select more than several mandarin level second-rank first class and the normal person without pronunciation medical history to read each survey successively
Examination language material, ready reading when gathering their reading test language material with EMA instrument but the seat of the tip of the tongue present position when not reading
Mark i.e. normal person's the tip of the tongue static frames coordinate, read during each testing material static with normal person's the tip of the tongue in the coordinate of the tip of the tongue place
The maximum value i.e. normal person's the tip of the tongue Euclidean distance of frame coordinate Euclidean distance, read upper lip and lower lip opening and closing degree in each testing material
Maximum i.e. normal person's lips folding distance, read duration used by each testing material i.e. normal person's duration, normal person pronunciation
The slope of formant trajectory equation, with normal person's the tip of the tongue Euclidean distance, normal person's lips folding distance, normal person's duration, normal
The pronounce slope of formant trajectory equation of people is parameter Criterion data base, during normal person reads each testing material,
Initial consonant being total to the starting point of simple or compound vowel of a Chinese syllable transition that one initial consonant of the testing material collected with EMA instrument combines with all simple or compound vowel of a Chinese syllable
Peak frequency of shaking is ordinate value, and the formant frequency at the midpoint of simple or compound vowel of a Chinese syllable is abscissa value, is formed and simple or compound vowel of a Chinese syllable quantity equal number
Discrete point, these discrete points are linear and tight clusters, and the slope of the discrete point fitting a straight line asked is normal person's pronunciation
The slope of formant trajectory equation;
Step 2, patient to be tested, according to dysphonia type selecting read test language material, gather patient to be tested with EMA instrument
Ready reading when reading each testing material but when not reading the i.e. patient's the tip of the tongue static frames of coordinate of the tip of the tongue present position sit
Mark, read value maximum with patient's the tip of the tongue static frames coordinate Euclidean distance in the coordinate of the tip of the tongue place during each testing material i.e.
Patient's the tip of the tongue Euclidean distance, read in each testing material the maximum of upper lip and lower lip opening and closing degree i.e. patient's lips folding away from
From, read duration used by each testing material i.e. patient's duration, with patient's the tip of the tongue Euclidean distance, patient's lips folding distance, suffer from
The pronounce slope of formant trajectory equation of person's duration, patient is reduced parameter, during patient reads each testing material, with
The initial consonant that any one initial consonant of the testing material that EMA instrument collects and all simple or compound vowel of a Chinese syllable combine is total to the starting point of simple or compound vowel of a Chinese syllable transition
Peak frequency of shaking is ordinate value, and the formant frequency at the midpoint of simple or compound vowel of a Chinese syllable is abscissa value, is formed and simple or compound vowel of a Chinese syllable quantity equal number
Discrete point, these discrete points are linear and tight clusters, and the slope of the discrete point fitting a straight line asked is patient and pronounces altogether
Shake the slope of peak equation of locus;
Step 3, selection fuzzy membership functions concept judge the defect level of dysphonia patient, to i-th testing material, i
For natural number, in all normal persons, normal person's the tip of the tongue Euclidean distance maximum is SiMax, normal person's the tip of the tongue Euclidean distance is minimum
Value is SiMin, experience obtains patient the tip of the tongue Euclidean distance maximum Smax, patient's the tip of the tongue Euclidean distance Si, work as SiWhen=0, i-th
Testing material patient apical articulation obstacle SziIt is 0, as 0 < Si<SiDuring min, i-th testing material patient apical articulation obstacle Szi
For Si/SiMin, works as Simin≦Si≦SiDuring max, i-th testing material patient apical articulation obstacle SziIt is 1, works as Simax<Si<
During Smax, i-th testing material patient apical articulation obstacle SziFor (Simax- Si)/(Smax-SiMax), work as Si≥Smax
Time, i-th testing material patient apical articulation obstacle SziIt is 0;In all normal persons, normal person's lips folding distance maximum
For ZiMax, normal person's lips folding distance minima is Z imin, experience obtains patient's lips folding distance maximum Zmax, suffers from
Person's lips folding distance is Zi, work as ZiWhen=0, i-th testing material patient face dysphonia ZziIt is 0, as 0 < Zi<ZiDuring min,
I-th testing material patient face dysphonia ZziFor Zi/ZiMin, works as Zimin≦Zi≦ZiDuring max, i-th testing material is suffered from
Person face dysphonia ZziIt is 1, works as Zimax<Zi< during Zmax, i-th testing material patient face dysphonia ZziFor
(Zimax- Zi)/(Zmax-ZiMax), work as ZiDuring >=Zmax, i-th testing material patient face dysphonia ZziIt is 0;Institute
Having in normal person, normal person's duration maximum is JiMax, normal person's duration minima is J imin, experience obtains patient's duration
It is worth greatly Jmax, a length of J during patienti, work as JiWhen=0, i-th testing material patient duration dysphonia JziIt is 0, as 0 < Ji<Jimin
Time, i-th testing material patient duration dysphonia JziFor Ji/JiMin, works as Jimin≦Ji≦JiDuring max, i-th test language
Material patient duration dysphonia JziIt is 1, works as Jimax<Ji< during Jmax, i-th testing material patient duration dysphonia JziFor
(Jimax- Ji)/(Jmax-JiMax), work as JiDuring >=Jmax, i-th testing material patient duration dysphonia JziIt is 0;Institute
Having in normal person, the pronounce gradient maxima of formant trajectory equation of normal person is KiMax, normal person pronounces formant trajectory side
The slope minima of journey is KiMin, experience obtains patient and pronounces the gradient maxima Kmax of formant trajectory equation, and experience obtains
Patient pronounces slope minima Kmin of formant trajectory equation, and Patients Patients to be measured pronounces the slope of formant trajectory equation
Ki, work as KiDuring Kmin, i-th testing material patient slope obstacle KziIt is 0, as Kmin < Ki<KiDuring min, i-th testing material
Patient slope obstacle KziFor (Ki-Kmin)/(KiMin-Kmin), work as Kimin≦Ki≦KiDuring max, i-th testing material patient
Duration dysphonia KziIt is 1, works as Kimax<Ki< during Kmax, i-th testing material patient duration dysphonia KziFor (Kimax-
Ki)/(Kmax-KiMax), work as KiDuring >=Kmax, i-th testing material patient duration dysphonia KziIt is 0;I-th test language
Material patient obstacle Ui=0.4*Szi+0.1*Zzi+0.1*Jzi+0.4*Kzi;
Step 4, Patient Global pronounce voice disorder U=| 1-U1|+...+|1-Ui|+...+|1-Un|, U1It is the 1st test language
Material patient's obstacle, UiFor i-th testing material patient's obstacle, UnBeing n-th testing material patient's obstacle, n is the total of testing material
Quantity belongs to natural number.
As a kind of optimal way: in step 3, experience obtains patient the tip of the tongue Euclidean distance maximum Smax and refers to doctor
Maximum in the patient's all data of the tip of the tongue Euclidean distance collected, experience obtains patient's lips folding distance maximum Zmax
Referring to patient's lips folding that doctor collects maximum in all data, experience obtains patient duration maximum Jmax
Referring to the maximum in all data of patient's duration that doctor collects, experience obtains patient and pronounces the oblique of formant trajectory equation
Rate maximum Kmax refer to the patient that doctor collects pronounce formant trajectory equation all data of slope in maximum, warp
Test and obtain pronounce slope minima Kmin of formant trajectory equation of patient and refer to that the patient that doctor collects pronounces formant rail
Minima in all data of slope of mark equation, and refer to the collection to patients different under same testing material situation.
The invention has the beneficial effects as follows: the exercise data gathered by EMA can pass through MATLAB drawing three-dimensional coordinate diagram, directly
Seeing and effectively compare with normal person, the method, from the angle of physiology, improves the accuracy of assessment, contrasts more intuitively
Dysphonia patient and the pronunciation difference of normal person.Equation of locus pronunciation model is based on neuroscience, is used for assessing voice
Stability and the method for particularity, will have breakthrough to domestic pathology voice study.The present invention combines kinetics and acoustics letter
Breath, it is possible to more accurately dysphonia patient is comprehensively assessed, provide theoretical basis and technology for pathological study
Support.
Detailed description of the invention
The present invention is with Windows7 system as operating environment, and MATLAB R2010b is data processing platform (DPP).The following is concrete
Operational approach:
Step one, type according to different dysphonia determine testing material, choose testing material and follow the feature of Chinese speech pronunciation
And rule, it is possible to adjusting testing material according to the type of different dysphonia, testing material is one or more, each test
Language material is the standard according to the combination of Chinese Pin Yin pseudonym simple or compound vowel of a Chinese syllable, an initial consonant and all simple or compound vowel of a Chinese syllable that can pronounce with the combination of this initial consonant
All combining forms, the present embodiment is for assessing the pronunciation patient lifting obstacle on tongue, owing to tongue cannot normally lift contact
Soft palate and upper tooth, cause some initial consonant cacologies of patient true, such as/l/ ,/d/ ,/t/ ,/s/ ,/ch/ etc..The present embodiment chooses survey
Examination language material is initial consonant/d/ ,/l/ ,/ch/, selects more than 10 mandarin level second-rank first class and the normal person without pronunciation medical history to depend on
The each testing material of secondary reading, ready reading when gathering their reading test language material with EMA instrument but the tip of the tongue when not reading
The coordinate of present position i.e. normal person's the tip of the tongue static frames coordinate, read during each testing material in the coordinate of the tip of the tongue place with just
The maximum value i.e. normal person's the tip of the tongue Euclidean distance of ordinary person's the tip of the tongue static frames coordinate Euclidean distance, read upper lip in each testing material
With the i.e. normal person's lips folding of maximum of lower lip opening and closing degree distance, read the i.e. normal person of duration used by each testing material time
Long, normal person pronounces the slope of formant trajectory equation, with normal person's the tip of the tongue Euclidean distance, normal person's lips folding distance, just
The pronounce slope of formant trajectory equation of ordinary person's duration, normal person is parameter Criterion data base, and normal person reads each survey
During examination language material, the initial consonant of an initial consonant of the testing material collected with EMA instrument and the combination of all simple or compound vowel of a Chinese syllable is to simple or compound vowel of a Chinese syllable mistake
The formant frequency of the starting point crossed is ordinate value, and the formant frequency at the midpoint of simple or compound vowel of a Chinese syllable is abscissa value, is formed and simple or compound vowel of a Chinese syllable
The discrete point of quantity equal number, these discrete points are linear and tight clusters, the slope of the discrete point fitting a straight line asked
Be normal person to pronounce the slope of formant trajectory equation, in the present invention normal person pronounce formant trajectory equation slope involved by
And coordinate be plane right-angle coordinate coordinate, other coordinate is 3 D stereo coordinate, and 3 D stereo coordinate is with each reader
Left and right directions is X-axis and direction is to be incremented by from right to left, with each reader's fore-and-aft direction as Y-axis and direction is from forward direction
Rear incremental;With each reader's above-below direction as Z axis and direction is to be incremented by from bottom to top.The present embodiment use INSTRUMENT MODEL is
AG501, records articulation with the sample rate that 200 frames are per second, gathers the fortune of each organ while tester produces voice
Dynamic data, and record synchronous voice data.With physiology glue, sensor (sensor) is adhered to the tip of the tongue of tester, upper lip
Change with these site location of synchro measure in the middle of centre, lower lip.
As a example by a wherein bit test person pronounces initial consonant/d/, first gathering enunciator's static frames data with EMA, pronunciation is dynamic
Making the static frames of data and refer to mute and without obvious articulation a Frame, tongue now and upperlip etc. are sent out
Sound organ is in relaxation state, and corresponding voice data is quiet section of speech waveform.Collecting test person pronunciation again/d/
Time motion trace data, choose the key frame of pronunciation/d/, due to pronunciation time the tip of the tongue be directly connected to pronunciation definition, I
The key frame of the primary study the tip of the tongue;
In order to study the pronunciation character of Chinese phoneme, need to go out can mark from three-dimensional articulation extracting data complicated and changeable
Know a frame of this phoneme or a few frame to characterize its personal characteristics, referred to as key frame, select the tip of the tongue Europe relative to the tip of the tongue static frames
The maximum frame of formula distance as in the tip of the tongue place coordinate during the tip of the tongue key frame, and then the reading test language material/d/ asked with
Value i.e. normal person's the tip of the tongue Euclidean distance that the tip of the tongue static frames coordinate Euclidean distance is maximum;
Table 1 one bit test person's static frames and the position of key frame
X-axis (mm) | Y-axis (mm) | Z axis (mm) | |
The static frames the tip of the tongue (T1) | 9.40 | 32.54 | 90.18 |
The key frame the tip of the tongue (T1) | 9.89 | 33.27 | 94.17 |
Extract the formant information gathering enunciator pronunciation/d/.According to the standard of Chinese Pin Yin pseudonym simple or compound vowel of a Chinese syllable combination, initial consonant/d/
With simple or compound vowel of a Chinese syllable combination have 18 kinds of forms, be respectively/da/ ,/duo/ ,/de/ ,/di/ ,/du/ ,/dai/ ,/dui/ ,/dao/ ,/
Dou/ ,/diu/ ,/die/ ,/dan/ ,/din/ ,/dun/ ,/dang/ ,/deng/ ,/ding/ ,/dong/, gather above group respectively
The pronunciation information closed, by the second formant initial consonant to the starting point (F2 of simple or compound vowel of a Chinese syllable transitiononset) and the midpoint of the second formant simple or compound vowel of a Chinese syllable
(F2mid) draw, these discrete points are linear and tight clusters.Discrete point obeys unary linear regression equation, according to research
Show that the slope of equation of locus can reflect the speech quality of speaker, therefore can judge the pronunciation feelings of enunciator from slope
Condition;
By calculating the equation of locus of a bit test person pronunciation/d/ it is
F2onset =0.416* F2mid+ 1288.316, k=0.416, the pronunciation duration J=0.67s of record enunciator.
The pronunciation data of same method 10 normal persons of collection, and set up the data base of normal articulation, as shown in table 2.
Wherein S represents normal person's the tip of the tongue Euclidean distance, and Z represents lips maximum folding distance, and K represents the slope of pronunciation equation of locus, J table
Show pronunciation duration.
The data of 2 10 health adult hair's speech mother/d/ of table
S(mm) | Z(mm) | K | J(s) | |
1 | 4.09 | 13.44 | 0.416 | 0.67 |
2 | 4.36 | 13.23 | 0.423 | 0.62 |
3 | 3.97 | 12.98 | 0.396 | 0.58 |
4 | 4.12 | 13.25 | 0.425 | 0.64 |
5 | 4.16 | 13.56 | 0.414 | 0.68 |
6 | 4.01 | 12.89 | 0.403 | 0.59 |
7 | 3.99 | 13.46 | 0.419 | 0.63 |
8 | 4.06 | 13.11 | 0.428 | 0.61 |
9 | 4.04 | 13.54 | 0.423 | 0.71 |
10 | 4.26 | 13.24 | 0.410 | 0.70 |
Step 2, patient to be tested, according to dysphonia type selecting read test language material, gather patient to be tested with EMA instrument
Ready reading when reading each testing material but when not reading the i.e. patient's the tip of the tongue static frames of coordinate of the tip of the tongue present position sit
Mark, read value maximum with patient's the tip of the tongue static frames coordinate Euclidean distance in the coordinate of the tip of the tongue place during each testing material i.e.
Patient's the tip of the tongue Euclidean distance, read in each testing material the maximum of upper lip and lower lip opening and closing degree i.e. patient's lips folding away from
From, read duration used by each testing material i.e. patient's duration, with patient's the tip of the tongue Euclidean distance, patient's lips folding distance, suffer from
The pronounce slope of formant trajectory equation of person's duration, patient is reduced parameter, during patient reads each testing material, with
The initial consonant that any one initial consonant of the testing material that EMA instrument collects and all simple or compound vowel of a Chinese syllable combine is total to the starting point of simple or compound vowel of a Chinese syllable transition
Peak frequency of shaking is ordinate value, and the formant frequency at the midpoint of simple or compound vowel of a Chinese syllable is abscissa value, is formed and simple or compound vowel of a Chinese syllable quantity equal number
Discrete point, these discrete points are linear and tight clusters, and the slope of the discrete point fitting a straight line asked is patient and pronounces altogether
Shake the slope of peak equation of locus;
Gather the pronunciation data of patient to be measured, as a comparison supplemental characteristic, as shown in table 3.Wherein S ' represents patient's the tip of the tongue Euclidean
Distance, Z ' represents lips maximum folding distance, and K ' represents the slope of pronunciation equation of locus, and J ' represents pronunciation duration.
The pronunciation data of table 3 patient to be measured
S’(mm) | Z’(mm) | K’ | J’(s) | |
1 | 3.82 | 12.85 | 0.392 | 0.52 |
Step 3, selection fuzzy membership functions concept judge the defect level of dysphonia patient, to i-th testing material, i
For natural number, in all normal persons, normal person's the tip of the tongue Euclidean distance maximum is SiMax, normal person's the tip of the tongue Euclidean distance is minimum
Value is SiMin, experience obtains patient the tip of the tongue Euclidean distance maximum Smax, patient's the tip of the tongue Euclidean distance Si, work as SiWhen=0, i-th
Testing material patient apical articulation obstacle SziIt is 0, as 0 < Si<SiDuring min, i-th testing material patient apical articulation obstacle Szi
For Si/SiMin, works as Simin≦Si≦SiDuring max, i-th testing material patient apical articulation obstacle SziIt is 1, works as Simax<Si<
During Smax, i-th testing material patient apical articulation obstacle SziFor (Simax- Si)/(Smax-SiMax), work as Si≥Smax
Time, i-th testing material patient apical articulation obstacle SziIt is 0;In all normal persons, normal person's lips folding distance maximum
For ZiMax, normal person's lips folding distance minima is ZiMin, experience obtains patient's lips folding distance maximum Zmax, suffers from
Person's lips folding distance is Zi, work as ZiWhen=0, i-th testing material patient face dysphonia ZziIt is 0, as 0 < Zi<ZiDuring min,
I-th testing material patient face dysphonia ZziFor Zi/ZiMin, works as Zimin≦Zi≦ZiDuring max, i-th testing material is suffered from
Person face dysphonia ZziIt is 1, works as Zimax<Zi< during Zmax, i-th testing material patient face dysphonia ZziFor
(Zimax- Zi)/(Zmax-ZiMax), work as ZiDuring >=Zmax, i-th testing material patient face dysphonia ZziIt is 0;Institute
Having in normal person, normal person's duration maximum is JiMax, normal person's duration minima is J imin, experience obtains patient's duration
It is worth greatly Jmax, a length of J during patienti, work as JiWhen=0, i-th testing material patient duration dysphonia JziIt is 0, as 0 < Ji<Jimin
Time, i-th testing material patient duration dysphonia JziFor Ji/JiMin, works as Jimin≦Ji≦JiDuring max, i-th test language
Material patient duration dysphonia JziIt is 1, works as Jimax<Ji< during Jmax, i-th testing material patient duration dysphonia JziFor
(Jimax- Ji)/(Jmax-JiMax), work as JiDuring >=Jmax, i-th testing material patient duration dysphonia JziIt is 0;Institute
Having in normal person, the pronounce gradient maxima of formant trajectory equation of normal person is KiMax, normal person pronounces formant trajectory side
The slope minima of journey is KiMin, experience obtains patient and pronounces the gradient maxima Kmax of formant trajectory equation, and experience obtains
Patient pronounces slope minima Kmin of formant trajectory equation, and Patients Patients to be measured pronounces the slope of formant trajectory equation
Ki, work as KiDuring Kmin, i-th testing material patient slope obstacle KziIt is 0, as Kmin < Ki<KiDuring min, i-th testing material
Patient slope obstacle KziFor (Ki-Kmin)/(KiMin-Kmin), work as Kimin≦Ki≦KiDuring max, i-th testing material patient
Duration dysphonia KziIt is 1, works as Kimax<Ki< during Kmax, i-th testing material patient duration dysphonia KziFor (Kimax-
Ki)/(Kmax-KiMax), work as KiDuring >=Kmax, i-th testing material patient duration dysphonia KziIt is 0;I-th test language
Material patient obstacle Ui=0.4*Szi+0.1*Zzi+0.1*Jzi+0.4*Kzi;
Illustrate using/d/ as first testing material, by step 2 and step 3 it is recognised that in all normal persons,
Normal person's the tip of the tongue Euclidean distance maximum is S1Max=4.26, normal person's the tip of the tongue Euclidean distance minima is S1Min=3.97, experience
Obtain patient the tip of the tongue Euclidean distance maximum Smax=4.55, patient's the tip of the tongue Euclidean distance S1=3.82, first testing material is suffered from
Person apical articulation obstacle Sz1For S1/S1Min=0.962, in all normal persons, normal person's lips folding distance maximum is Z1max
=13.56, normal person's lips folding distance minima is Z1Min=12.89, experience obtains patient's lips folding distance maximum
Zmax=14.15, patient's lips folding distance is Z1=12.85, first testing material patient face dysphonia Zz1For Z1/
Z1Min=0.997, in all normal persons, normal person's duration maximum is J1Max=0.71, normal person's duration minima is J 1min=
0.58, experience obtains patient duration maximum Jmax=0.82, a length of J during patient1=0.52, first testing material patient's duration
Dysphonia Jz1For J1/J1Min=0.897, in all normal persons, normal person pronounces the gradient maxima of formant trajectory equation
For K1Max=0.428, the pronounce slope minima of formant trajectory equation of normal person is K1Min=0.396, experience obtains patient and sends out
The gradient maxima Kmax=0.498 of sound formant trajectory equation, experience obtain patient pronounce formant trajectory equation slope
Little value Kmin=0.223, Patients Patients to be measured pronounces the slope K of formant trajectory equation1=0.392, first testing material is suffered from
Person slope obstacle Kz1For (K1-Kmin)/(K1Min-Kmin)=0.169/0.173=0.977, first testing material patient barrier
Hinder U1=0.4*Sz1+0.1*Zz1+0.1*Jz1+0.4*Kz1=0.920, same method obtains second testing material (/l/) and suffers from
Person obstacle U2=0.933, the 3rd testing material (/ch/) patient obstacle U3=0.893, second testing material in the present embodiment (/
L/) detailed process of patient's obstacle and the 3rd testing material (/ch/) patient's obstacle is complete in first testing material patient's obstacle
Universal class seemingly, does not the most add explanation.
Step 4, Patient Global pronounce voice disorder U=| 1-U1|+...+|1-Ui|+...+|1-Un|, U1It is the 1st survey
Examination language material patient's obstacle, UiFor i-th testing material patient's obstacle, UnBeing n-th testing material patient's obstacle, n is testing material
Total quantity.
The present embodiment, Patient Global pronounces voice disorder U=| 1-0.920 |+| 1-0.933 |+| 1-0.893 |=0.254, suffers from
Person's voice disorder value of comprehensively pronouncing is the biggest, illustrates that patient's voice disorder that pronounces is the biggest.
Claims (2)
1. a dysphonia Chinese appraisal procedure based on EMA, it is characterised in that carry out in accordance with the following steps:
Step one, type according to different dysphonia determine testing material, and testing material is one or more, each test
Language material is the standard according to the combination of Chinese Pin Yin pseudonym simple or compound vowel of a Chinese syllable, an initial consonant and all simple or compound vowel of a Chinese syllable that can pronounce with the combination of this initial consonant
All combining forms, select more than several mandarin level second-rank first class and the normal person without pronunciation medical history to read each survey successively
Examination language material, ready reading when gathering their reading test language material with EMA instrument but the seat of the tip of the tongue present position when not reading
Mark i.e. normal person's the tip of the tongue static frames coordinate, read during each testing material static with normal person's the tip of the tongue in the coordinate of the tip of the tongue place
The maximum value i.e. normal person's the tip of the tongue Euclidean distance of frame coordinate Euclidean distance, read upper lip and lower lip opening and closing degree in each testing material
Maximum i.e. normal person's lips folding distance, read duration used by each testing material i.e. normal person's duration, normal person pronunciation
The slope of formant trajectory equation, with normal person's the tip of the tongue Euclidean distance, normal person's lips folding distance, normal person's duration, normal
The pronounce slope of formant trajectory equation of people is parameter Criterion data base, during normal person reads each testing material,
Initial consonant being total to the starting point of simple or compound vowel of a Chinese syllable transition that one initial consonant of the testing material collected with EMA instrument combines with all simple or compound vowel of a Chinese syllable
Peak frequency of shaking is ordinate value, and the formant frequency at the midpoint of simple or compound vowel of a Chinese syllable is abscissa value, is formed and simple or compound vowel of a Chinese syllable quantity equal number
Discrete point, these discrete points are linear and tight clusters, and the slope of the discrete point fitting a straight line asked is normal person's pronunciation
The slope of formant trajectory equation;
Step 2, patient to be tested, according to dysphonia type selecting read test language material, gather patient to be tested with EMA instrument
Ready reading when reading each testing material but when not reading the i.e. patient's the tip of the tongue static frames of coordinate of the tip of the tongue present position sit
Mark, read value maximum with patient's the tip of the tongue static frames coordinate Euclidean distance in the coordinate of the tip of the tongue place during each testing material i.e.
Patient's the tip of the tongue Euclidean distance, read in each testing material the maximum of upper lip and lower lip opening and closing degree i.e. patient's lips folding away from
From, read duration used by each testing material i.e. patient's duration, with patient's the tip of the tongue Euclidean distance, patient's lips folding distance, suffer from
The pronounce slope of formant trajectory equation of person's duration, patient is reduced parameter, during patient reads each testing material, with
The initial consonant that any one initial consonant of the testing material that EMA instrument collects and all simple or compound vowel of a Chinese syllable combine is total to the starting point of simple or compound vowel of a Chinese syllable transition
Peak frequency of shaking is ordinate value, and the formant frequency at the midpoint of simple or compound vowel of a Chinese syllable is abscissa value, is formed and simple or compound vowel of a Chinese syllable quantity equal number
Discrete point, these discrete points are linear and tight clusters, and the slope of the discrete point fitting a straight line asked is patient and pronounces altogether
Shake the slope of peak equation of locus;
Step 3, selection fuzzy membership functions concept judge the defect level of dysphonia patient, to i-th testing material, i
For natural number, in all normal persons, normal person's the tip of the tongue Euclidean distance maximum is SiMax, normal person's the tip of the tongue Euclidean distance is minimum
Value is SiMin, experience obtains patient the tip of the tongue Euclidean distance maximum Smax, patient's the tip of the tongue Euclidean distance Si, work as SiWhen=0, i-th
Testing material patient apical articulation obstacle SziIt is 0, as 0 < Si<SiDuring min, i-th testing material patient apical articulation obstacle Szi
For Si/SiMin, works as Simin≦Si≦SiDuring max, i-th testing material patient apical articulation obstacle SziIt is 1, works as Simax<Si<
During Smax, i-th testing material patient apical articulation obstacle SziFor (Simax- Si)/(Smax-SiMax), work as Si≥Smax
Time, i-th testing material patient apical articulation obstacle SziIt is 0;In all normal persons, normal person's lips folding distance maximum
For ZiMax, normal person's lips folding distance minima is Z imin, experience obtains patient's lips folding distance maximum Zmax, suffers from
Person's lips folding distance is Zi, work as ZiWhen=0, i-th testing material patient face dysphonia ZziIt is 0, as 0 < Zi<ZiDuring min,
I-th testing material patient face dysphonia ZziFor Zi/ZiMin, works as Zimin≦Zi≦ZiDuring max, i-th testing material is suffered from
Person face dysphonia ZziIt is 1, works as Zimax<Zi< during Zmax, i-th testing material patient face dysphonia ZziFor
(Zimax- Zi)/(Zmax-ZiMax), work as ZiDuring >=Zmax, i-th testing material patient face dysphonia ZziIt is 0;Institute
Having in normal person, normal person's duration maximum is JiMax, normal person's duration minima is J imin, experience obtains patient's duration
It is worth greatly Jmax, a length of J during patienti, work as JiWhen=0, i-th testing material patient duration dysphonia JziIt is 0, as 0 < Ji<Jimin
Time, i-th testing material patient duration dysphonia JziFor Ji/JiMin, works as Jimin≦Ji≦JiDuring max, i-th test language
Material patient duration dysphonia JziIt is 1, works as Jimax<Ji< during Jmax, i-th testing material patient duration dysphonia JziFor
(Jimax- Ji)/(Jmax-JiMax), work as JiDuring >=Jmax, i-th testing material patient duration dysphonia JziIt is 0;Institute
Having in normal person, the pronounce gradient maxima of formant trajectory equation of normal person is KiMax, normal person pronounces formant trajectory side
The slope minima of journey is KiMin, experience obtains patient and pronounces the gradient maxima Kmax of formant trajectory equation, and experience obtains
Patient pronounces slope minima Kmin of formant trajectory equation, and Patients Patients to be measured pronounces the slope of formant trajectory equation
Ki, work as KiDuring Kmin, i-th testing material patient slope obstacle KziIt is 0, as Kmin < Ki<KiDuring min, i-th testing material
Patient slope obstacle KziFor (Ki-Kmin)/(KiMin-Kmin), work as Kimin≦Ki≦KiDuring max, i-th testing material patient
Duration dysphonia KziIt is 1, works as Kimax<Ki< during Kmax, i-th testing material patient duration dysphonia KziFor (Kimax-
Ki)/(Kmax-KiMax), work as KiDuring >=Kmax, i-th testing material patient duration dysphonia KziIt is 0;I-th test language
Material patient obstacle Ui=0.4*Szi+0.1*Zzi+0.1*Jzi+0.4*Kzi;
Step 4, Patient Global pronounce voice disorder U=| 1-U1|+...+|1-Ui|+...+|1-Un|, U1It is the 1st testing material
Patient's obstacle, UiFor i-th testing material patient's obstacle, UnBeing n-th testing material patient's obstacle, n is the sum of testing material
Amount.
A kind of dysphonia Chinese appraisal procedure based on EMA the most according to claim 1, it is characterised in that: step 3
In, experience obtains patient the tip of the tongue Euclidean distance maximum Smax and refers to patient's all data of the tip of the tongue Euclidean distance that doctor collects
In maximum, experience obtain patient's lips folding distance maximum Zmax refer to patient's lips folding distance that doctor collects
Maximum in all data, experience obtains patient duration maximum Jmax and refers to all data of patient's duration that doctor collects
In maximum, experience obtains the pronounce gradient maxima Kmax of formant trajectory equation of patient and refers to the patient that doctor collects
Maximum in all data of slope of pronunciation formant trajectory equation, experience obtains patient and pronounces the oblique of formant trajectory equation
Rate minima Kmin refer to the patient that doctor collects pronounce formant trajectory equation all data of slope in minima, and
And refer to the collection to patients different under same testing material situation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610521815.3A CN106157212A (en) | 2016-07-05 | 2016-07-05 | A kind of dysphonia Chinese appraisal procedure based on EMA |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610521815.3A CN106157212A (en) | 2016-07-05 | 2016-07-05 | A kind of dysphonia Chinese appraisal procedure based on EMA |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106157212A true CN106157212A (en) | 2016-11-23 |
Family
ID=58061640
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610521815.3A Pending CN106157212A (en) | 2016-07-05 | 2016-07-05 | A kind of dysphonia Chinese appraisal procedure based on EMA |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106157212A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107452370A (en) * | 2017-07-18 | 2017-12-08 | 太原理工大学 | A kind of application method of the judgment means of Chinese vowel followed by a nasal consonant dysphonia patient |
CN108630225A (en) * | 2018-03-29 | 2018-10-09 | 太原理工大学 | Barrier children's vowel appraisal procedure is listened based on fuzzy overall evaluation |
CN109360645A (en) * | 2018-08-01 | 2019-02-19 | 太原理工大学 | A kind of statistical classification method of dysarthrosis pronunciation movement spatial abnormal feature |
CN110379221A (en) * | 2019-08-09 | 2019-10-25 | 陕西学前师范学院 | A kind of pronunciation of English test and evaluation system |
-
2016
- 2016-07-05 CN CN201610521815.3A patent/CN106157212A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107452370A (en) * | 2017-07-18 | 2017-12-08 | 太原理工大学 | A kind of application method of the judgment means of Chinese vowel followed by a nasal consonant dysphonia patient |
CN108630225A (en) * | 2018-03-29 | 2018-10-09 | 太原理工大学 | Barrier children's vowel appraisal procedure is listened based on fuzzy overall evaluation |
CN109360645A (en) * | 2018-08-01 | 2019-02-19 | 太原理工大学 | A kind of statistical classification method of dysarthrosis pronunciation movement spatial abnormal feature |
CN109360645B (en) * | 2018-08-01 | 2021-06-11 | 太原理工大学 | Statistical classification method for dysarthria pronunciation and movement abnormal distribution |
CN110379221A (en) * | 2019-08-09 | 2019-10-25 | 陕西学前师范学院 | A kind of pronunciation of English test and evaluation system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11786171B2 (en) | Method and system for articulation evaluation by fusing acoustic features and articulatory movement features | |
Lam et al. | Acoustics of clear speech: Effect of instruction | |
Gallena et al. | Effects of levodopa on laryngeal muscle activity for voice onset and offset in Parkinson disease | |
Lee et al. | Relationship between tongue positions and formant frequencies in female speakers | |
Laganaro et al. | Sensitivity and specificity of an acoustic-and perceptual-based tool for assessing motor speech disorders in French: The MonPaGe-screening protocol | |
CN106073706B (en) | A kind of customized information and audio data analysis method and system towards Mini-mental Status Examination | |
CN106157212A (en) | A kind of dysphonia Chinese appraisal procedure based on EMA | |
Wang et al. | Individual articulator's contribution to phoneme production | |
Ribeiro et al. | Speaker-independent classification of phonetic segments from raw ultrasound in child speech | |
Kuruvilla-Dugdale et al. | An exploratory model of speech intelligibility for healthy aging based on phonatory and articulatory measures | |
CN108962397B (en) | Pen and voice-based cooperative task nervous system disease auxiliary diagnosis system | |
Fivela et al. | Italian Vowel and Consonant (co) articulation in Parkinson’s Disease: extreme or reduced articulatory variability? | |
Sy et al. | A statistical causal model for the assessment of dysarthric speech and the utility of computer-based speech recognition | |
Rong et al. | Speech intelligibility loss due to amyotrophic lateral sclerosis: the effect of tongue movement reduction on vowel and consonant acoustic features | |
Engwall | Tongue talking: studies in intraoral speech synthesis | |
Carignan | A network-modeling approach to investigating individual differences in articulatory-to-acoustic relationship strategies | |
Mizoguchi | Articulation of the Japanese moraic nasal: Place of articulation, assimilation, and L2 transfer | |
Zhao et al. | Visemes of Chinese Shaanxi Xi’an Dialect Talking Head [J] | |
Mielke et al. | Development of a new vowel feature from coarticulation: Biomechanical modeling of rhotic vowels in Kalasha | |
Swartz | Exploring vowel space metrics and quality of life measures in adolescents with typical speech, residual speech sound disorder, and childhood apraxia of speech | |
Wang et al. | Contribution of tongue lateral to consonant production | |
Middag et al. | DIA: a tool for objective intelligibility assessment of pathological speech. | |
Dokovova et al. | Tongue shape complexity in children with and without speech sound disorders | |
Koenig et al. | Studying articulatory variability using functional data analysis | |
Byeon | The acoustic characteristics of Korean diphthongs in speakers with flaccid and hypokinetic dysarthria |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161123 |
|
RJ01 | Rejection of invention patent application after publication |