CN107610691A - English vowel sounding error correction method and device - Google Patents
English vowel sounding error correction method and device Download PDFInfo
- Publication number
- CN107610691A CN107610691A CN201710803552.XA CN201710803552A CN107610691A CN 107610691 A CN107610691 A CN 107610691A CN 201710803552 A CN201710803552 A CN 201710803552A CN 107610691 A CN107610691 A CN 107610691A
- Authority
- CN
- China
- Prior art keywords
- english
- vowel
- measurand
- sounding
- typing
- 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
Landscapes
- Electrically Operated Instructional Devices (AREA)
Abstract
The present invention relates to a kind of English vowel sounding error correction method, including：Step 1, the English vowel sounding acoustic model of pre-stored criteria；Step 2, the English Phonetics of typing measurand；Step 3, the vowel in the English Phonetics of measurand is identified；Step 4, typing measurand reads aloud the voice of the vowel identified；Step 5, the voice for reading aloud measurand the vowel identified carries out English vowel sounding acoustic analysis；Step 6, the English vowel sounding acoustic analyze data of measurand obtains first degree of deviation compared with the English vowel sounding acoustic model of standard；Step 7, error correction is carried out to the English vowel sounding of measurand according to first degree of deviation.By analyzing the English vowel sounding acoustic of measurand, and compared with the English vowel sounding acoustic model of the standard with prestoring, so as to be corrected to the English vowel sounding of measurand, to cause the English vowel sounding of measurand more accurate.
Description
Technical field
The present invention relates to technical field of voice recognition, and in particular to English vowel sounding error correction method and device.
Background technology
Language, it is the bridge of Human communication, the diversity of language, and the obstacle of exchange.English, as frequency of use compared with
High common language is dominated." variety of English " of the different pronunciation characteristics of country variant turns into the obstacle in exchange in the world.For
The research of English feature under different Mother Languages is to break through communication disorder, improves the key of speaker's pronunciation of English.To speaking
The identification of people's phonetic feature, it is the unavoidable process to the identification of speaker's voice content.To improve the English language of speaker
Sound pronounces, and improves the communication disorder between the people of different nationalities, when speaker's voice deviates target language speech criterion certain limit
When, the pronunciation of English to speaker, which is corrected, just seems increasingly important.
From the air-flow oral cavity that acoustic resonator acts on by of lung's exhalation during pronunciation, send drag minimization and have no friction sound
Voice.Although vocal cords all vibrate when sending out vowel in general, it can also vocal cords is not vibrated, send out into voiceless sound or whisper in sb.'s ear
Sound.
From the viewpoint of articulatory phonetics, vowel is generally classified by the position of tongue and the shape of lips, high vowel hair
Lingual surface arches upward during sound, and against palate, the position of tongue is relatively low flat during low vowel articulation, and the distance of lingual surface and palate is slightly larger, in
The position of tongue is in intermediateness during vowel articulation, and high, medium and low vowel is also classified by front and rear row.The position of lingual surface and the shape of lip
Shape is a standard of vowel classification.
According to articulatory phonetics, vowel is airstream vibration vocal cords, a kind of pronunciation for hindering and being formed is not received in oral cavity,
Different oval shapes forms different vowels；And consonant is that air-flow is hindered in oral cavity and formed, different pronunciations
Position or manner of articulation form different consonants.
The content of the invention
The purpose of the present invention is when based on English being target language, the English vowel sonification model pair of measurand and standard
Than the error correction of the measurand English vowel sounding of realization.
One aspect of the present invention provides a kind of English vowel sounding error correction method, including：Step 1, the English member of pre-stored criteria
Sound sounding acoustic model；Step 2, the English Phonetics of typing measurand；Step 3, the English Phonetics of the measurand is identified
In vowel；Step 4, measurand described in typing reads aloud the voice of the vowel identified；Step 5, to described tested pair
The voice of the vowel identified as described in reading aloud carries out English vowel sounding acoustic analysis；Step 6, the English of the measurand
Vowel sounding acoustic analyze data obtains first degree of deviation compared with the English vowel sounding acoustic model of the standard；Step
7, error correction is carried out to the English vowel sounding of the measurand according to first degree of deviation.
The step 1 includes：The English Phonetics of the multiple standard English sample objects of typing；Identify the multiple standard English
Vowel in the English Phonetics of sample object；English vowel sounding acoustic analysis is carried out to the vowel of each sample object respectively；
The English vowel sounding acoustic model of the standard is generated according to the English vowel sounding acoustic analysis result.
The step 2 includes：The phonetic material is provided according to the nationality of the measurand, and tested pair described in typing
English Phonetics as reading aloud the phonetic material.
The step 3 includes：The member in the English Phonetics of the measurand is identified according to the resonance peak of the vowel
Sound.
The step 3 includes：Described tested pair according to the identification of the duration of the resonance peak of the vowel and the vowel
Vowel in the English Phonetics of elephant.
The resonance peak includes the first resonance peak and the second resonance peak.
The step 7 includes：The data and the English vowel of the standard analyzed according to the English vowel sounding acoustic
Sounding acoustic model, with the sounding of measurand English vowel described in visual Image Adjusting.
After the step 7, in addition to：Measurand described in typing reads aloud the voice of the vowel again again；To described
The voice that measurand is read aloud again carries out English vowel sounding acoustic analysis；The English for the voice that the measurand is read aloud again
Language vowel sounding acoustic analyze data obtains second degree of deviation compared with the English vowel sounding acoustic model of the standard；Root
The English vowel sounding that the measurand is exported according to first degree of deviation and second degree of deviation evaluates text.
The English vowel sounding acoustic analysis includes：Measure the resonance peak of the English vowel sounding of the typing；Survey
Measure the duration of the English vowel sounding of the typing；According to the resonance peak of the English vowel sounding of the typing and it is described when
The long English vowel sounding acoustic analyze data for generating the typing.
The resonance peak includes the first resonance peak and the second resonance peak.
The present invention also provides a kind of storage device, wherein being stored with a plurality of instruction, the instruction is suitable to be loaded by processor
And perform and be：Step 1, the English vowel sounding acoustic model of pre-stored criteria；Step 2, the English Phonetics of typing measurand；
Step 3, the vowel in the English Phonetics of the measurand is identified；Step 4, measurand described in typing reads aloud described identify
Vowel voice；Step 5, the voice for reading aloud the measurand vowel identified carries out English vowel sounding sound
Credit is analysed；Step 6, the English vowel sounding sound of the English Phonetics sounding acoustic analyze data of the measurand and the standard
Learn model to compare, obtain first degree of deviation；Step 7, the English vowel of the measurand is sent out according to first degree of deviation
Sound carries out error correction.
The step 1 includes：The English Phonetics of the multiple standard English sample objects of typing；Identify the multiple standard English
Vowel in the English Phonetics of sample object；English vowel sounding acoustic analysis is carried out to the vowel of each sample object respectively；
The English vowel sounding acoustic model of the standard is generated according to the English vowel sounding acoustic analysis result.
The step 2 includes：Phonetic material is provided according to the nationality of the measurand, and measurand is bright described in typing
Read the English Phonetics of the phonetic material.
The step 3 includes：The member in the English Phonetics of the measurand is identified according to the resonance peak of the vowel
Sound.
The step 3 also includes：It is described tested according to the identification of the duration of the resonance peak of the vowel and the vowel
Vowel in the English Phonetics of object.
The resonance peak includes the first resonance peak and the second resonance peak.
The step 7 includes：The data and the English vowel of the standard analyzed according to the English vowel sounding acoustic
Sounding acoustic model, with the sounding of measurand English vowel described in visual Image Adjusting.
After the step 7, in addition to：Measurand described in typing reads aloud the language of the vowel identified again again
Sound；The voice read aloud again the measurand carries out English vowel sounding acoustic analysis；The measurand is read aloud again
Voice English vowel sounding acoustic analyze data compared with the English vowel sounding acoustic model of the standard, obtain second
The degree of deviation；The English vowel sounding that the measurand is exported according to first degree of deviation and second degree of deviation evaluates text
This.
The English vowel sounding acoustic analysis includes：Measure the resonance peak of the English vowel sounding of the typing；Survey
Measure the duration of the vowel sounding of the typing；According to the resonance peak of the English vowel sounding of the typing and it is described when it is long-living
Into the English vowel sounding acoustic analyze data of the typing.
The resonance peak includes the first resonance peak and the second resonance peak.
The present invention also provides a kind of English vowel sounding error correction device, including：Processor, it is adapted for carrying out each instruction；And
Storage device, suitable for storing a plurality of instruction, the instruction is suitable for being loaded and being performed by the processor：Step 1, pre-stored criteria
English vowel sounding acoustic model；Step 2, the English Phonetics of typing measurand；Step 3, the measurand is identified
Vowel in English Phonetics；Step 4, measurand described in typing reads aloud the voice of the vowel identified；Step 5, to institute
State the voice progress English vowel sounding acoustic analysis that measurand reads aloud the vowel identified；Step 6, described tested pair
The English vowel sounding acoustic of elephant is analyzed compared with the English vowel sounding acoustic model of the standard, obtains first degree of deviation；
Step 7, error correction is carried out to the English vowel sounding of the measurand according to first degree of deviation.
The step 1 includes：The English Phonetics of the multiple standard English sample objects of typing；Identify the multiple standard English
Vowel in the English Phonetics of sample object；English vowel sounding acoustic analysis is carried out to the vowel of each sample object respectively；
The English vowel sounding acoustic model of the standard is generated according to the English vowel sounding acoustic analysis result.
The step 2 includes：Phonetic material is provided according to the nationality of the sample object, and sample object is bright described in typing
Read the English Phonetics of the phonetic material.
The step 3 includes：The member in the English Phonetics of the measurand is identified according to the resonance peak of the vowel
Sound.
The step 3 includes：Described tested pair according to the identification of the duration of the resonance peak of the vowel and the vowel
Vowel in the English Phonetics of elephant.
The resonance peak includes the first resonance peak and the second resonance peak.
The step 7 includes：The data and the English vowel of the standard analyzed according to the English vowel sounding acoustic
Sounding acoustic model, with the sounding of measurand English vowel described in visual Image Adjusting.
After the step 7, in addition to：Measurand described in typing reads aloud the language of the vowel identified again again
Sound；The voice read aloud again the measurand carries out English vowel sounding acoustic analysis；The measurand is read aloud again
English vowel sounding acoustic analyze data compared with the English vowel sounding acoustic model of the standard, obtain the second deviation
Degree；The English vowel sounding that the measurand is exported according to first degree of deviation and second degree of deviation evaluates text.
The English vowel sounding acoustic analysis includes：Measure the resonance peak of the English vowel sounding of the typing；Survey
Measure the duration of the vowel sounding of the typing；According to the resonance peak of the English vowel sounding of the typing and it is described when it is long-living
Into the English vowel sounding acoustic analyze data of the typing.
The resonance peak includes the first resonance peak and the second resonance peak.
The beneficial effects of the present invention are, by analyzing the English vowel sounding acoustic of measurand, and with it is pre-
The English vowel sounding acoustic model for the standard deposited compares, so as to be corrected to the English vowel sounding of measurand, with
So that the English vowel sounding of measurand is more accurate.
Brief description of the drawings
Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing：
Fig. 1 is the flow chart of the English vowel sounding error correction method 100 of one embodiment of the invention；
Fig. 2 is the method 200 for forming selection sample object in standard English vowel sonification model of one embodiment of the invention
Flow chart；
Fig. 3 is the flow chart of the method 300 of the formation standard English vowel sonification model of one embodiment of the invention；
Fig. 4 is the visualization English vowel sounding error correction figure of the English vowel sounding error correction method of one embodiment of the invention；
Fig. 5 is the flow chart of the English vowel sounding error correction method 500 of further embodiment of this invention；
Fig. 6 is the flow chart of the English vowel sounding analysis method 600 of one embodiment of the invention；
Fig. 7 is the difference generated using the analysis method of English vowel sounding acoustic shown in Fig. 6 600 of one embodiment of the invention
The English vowel sounding acoustic analysis chart of national men and women's measurand；
Fig. 8 is the standard generated using the analysis method of English vowel sounding acoustic shown in Fig. 6 600 of one embodiment of the invention
English vowel sonification model figure.
Embodiment
In conjunction with accompanying drawing, presently preferred embodiments of the present invention is elaborated.
As shown in figure 1, the flow chart of the English vowel sounding error correction method 100 of one embodiment of the invention.
Step 101, the English vowel sonification model of pre-stored criteria first.For example, the sample of English native background can be preserved
English vowel sonification model of the vowel sonification model of object as standard.Heretofore described sample object refers to being used for
Selected speaker during the English vowel sonification model of formation standard.
Step 103, the English Phonetics of typing measurand.In a specific embodiment, it is possible to provide phonetic material is for tested
Object is read aloud, and other any English words or sentence etc. can be also read aloud by measurand.When selection is read aloud according to phonetic material,
Identical phonetic material when can provide and establish the English vowel sonification model of standard, it can also provide and establish the English of standard
Different phonetic material during vowel sonification model.
Step 105, the vowel in the English Phonetics of the measurand of typing is identified.
Step 107, typing measurand reads aloud the voice of the vowel identified.
Step 109, the voice for the vowel identified read aloud the measurand of institute's typing does English vowel sounding acoustic
Analysis, can obtain the vowel sounding acoustic analyze data.
Step 111, by the English vowel sounding of resulting English vowel sounding acoustic analyze data and the standard to prestore
Model compares, and obtains first degree of deviation.
Step 113, error correction is carried out to the English vowel sounding of measurand according to first degree of deviation.
In a specific embodiment, after the English Phonetics that a certain measurand of typing is read aloud, the English is identified first
Vowel in voice, the voice of the vowel identified is read aloud in typing measurand, reads aloud measurand the identification
The voice of the vowel gone out carries out English vowel sounding acoustic analysis, obtained English vowel sounding acoustic analyze data and prestores
The English vowel sonification model of standard is compared, and the English vowel pronunciation and the English vowel of standard that can obtain measurand are sent out
The difference of sound, then the English vowel sounding of measurand can be corrected according to the difference.
In one embodiment, it is necessary to which the sample object of selection standard English goes to establish the English vowel sounding acoustic of standard
Model, it can be completed using Fig. 2 method.
Step 201, phonetic material is selected, the present invention is for English as target language, and using vowel as main object, language material relates to
And presentation of all vowels of English in the structure of word and the structure of sentence.Sentence structure includes all simple sentence sentence patterns 5 of English
Kind.Semanteme, which is related to, can predict sentence and unpredictable sentence.It can predict in sentence comprising the high sentence of predictability with that can predict
The low sentence of property.All words are high frequency words, but including all English vowels.Phonetic material can be set by world-renowned phonetician
Meter.Can predict with unpredictable sentence, can have an impact to the perception of vowel, such as：Complete sentence is " spread some
Butter on the bread ", though sentence for " spread some butter on the ", although do not say " bread ", or
Person does not identify " butter ", but it is " butter/bread " to be generally known that, that is to say, that sentence can be predicted, so
It can identify.
Step 203, selection standard English sample object reads aloud phonetic material, establishes sound bank.In actual applications, it is optional
The native country English adult in California is selected as standard English sample object, after selecting these people, is read aloud respectively selected
The phonetic material selected, and record and form sound bank.
Step 205, the selection that standard English sample object represents, can be chosen in the U.S. identical with sample object language setting
Music-listener, carry out mid-term perception, selection perceive in be in median men and women speaker, they are most A representative Samples
Object.
Step 207, overall English vowel sounding sound is carried out to the most representational sample object perceived by mid-term
Credit is analysed, and forms the English vowel sounding acoustic model of standard.
In one embodiment, by method shown in Fig. 2 choose most representational standard English pronounce crowd it
Afterwards, the English vowel sounding acoustic model of the method generation standard shown in Fig. 3 can be used.
Step 301, the English Phonetics of the multiple standard English sample objects of typing, after phonetic entry, sound materials are formed,
Vowel more than one in natural recreation area, so form cardinal vowels data.
Step 303, the vowel in the English Phonetics of the multiple standard English sample object is identified, in an embodiment
In, the vowel in voice can be identified according to the resonance peak of vowel, or it is further to improve identification accuracy, by altogether
The duration of peak value combination vowel of shaking identifies the vowel in voice.In a specific embodiment, resonance peak can be vowel
First resonance peak and the second resonance peak.Wherein, the first resonance peak F1 represent be lip dimension, that is, the dimension up and down pronounced
Degree, the second resonance peak F2 represent be tongue dimension, that is, the front and rear dimension pronounced.
Step 305, English vowel sounding acoustic analysis is carried out to the vowel in the voice of each standard English sample object.
Step 307, the English vowel sounding acoustic analysis number of the multiple standard English sample objects obtained according to step 305
According to the English vowel sounding acoustic model of generation standard.In a specific embodiment, the single vowel feature of each sample object
Meeting difference and other speakers, but forward and backward, the high and low vowel of a people can be in certain scope, according to this particularity
The vowel of speaker is carried out overall measurement, the essential characteristic of speaker is then determined according to vowel scope.
In one embodiment, phonetic material can be provided according to the nationality of measurand, and typing measurand reads aloud institute
State the English Phonetics of phonetic material.For example, for Chinese, the higher vowel of error rate is：/i:~/ ,/ε~/ and/
u:~/, then when providing phonetic material, the phonetic material for including above-mentioned vowel can be provided, with more to measurand more
The vowel often to malfunction is corrected so that the correction of English vowel sounding is more targeted.
In one embodiment, carrying out error correction to the English vowel sounding of measurand according to first degree of deviation can wrap
Include：The data and the English vowel sounding acoustic model of standard analyzed according to English vowel sounding acoustic, are adjusted with visual image
The sounding of whole measurand English vowel.As shown in figure 4, be the English vowel sounding error correction method of one embodiment of the invention can
Depending on changing English vowel sounding error correction figure, for example, when carrying out error correction to the English vowel sounding of a Chinese, there is provided voice material
Material is " a good looking woman ", by using the English vowel sounding error correction method pair of the present inventionPronunciation is carried out
Error correction, what is obtained first is the first formant and the second formant of the vowel sounding for the Chinese that Fig. 4 orbicular spots represent
Coordinate position, Fig. 4 intermediate cams representStandard the first formant of sounding and the coordinate position of the second formant, pass through
Such a visualization contrast, can more intuitively allow speaker to know the English vowel sounding for how adjusting oneself.
Fig. 5 is the flow chart of the English vowel sounding error correction method 500 of further embodiment of this invention, wherein, step 501 to
513 with the step 101 in Fig. 1 to 113 identical, step 515, typing measurand reads aloud the voice of the vowel identified again；
Step 517, the voice read aloud again measurand carries out English vowel sounding acoustic analysis；Step 519, step 517 is obtained
To English vowel sounding acoustic analysis data compared with the English vowel sounding acoustic model of standard, obtain the second deviation
Degree；Step 521, the English vowel sounding for measurand being exported according to first degree of deviation and the second deviation evaluates text.
In a specific embodiment, English vowel sounding evaluation text can include the original English vowel hair of measurand
The information such as sound acoustic analysis figure and the English vowel sounding acoustic analysis chart after correction, to facilitate measurand to understand itself
Pronunciation of English situation and the existing articulation problems for needing to correct so that measurand can carry out mesh according to own characteristic
English pronunciation practice, to improve the English sounding of measurand itself.
In the English vowel sounding acoustic model of generation standard or English vowel sounding acoustic analysis is carried out to measurand
When, the English vowel sounding acoustic analysis method shown in Fig. 6 can be used.
Step 601 typing English Phonetics first.When the English vowel sounding acoustic model of the standard of generation, typing is mark
The voice of the sample object of quasi- pronunciation of English；When carrying out the analysis of English vowel sounding acoustic to measurand, typing be by
Survey the voice of object.
Vowel in step 603 identification English Phonetics.In one embodiment, can be identified according to the resonance peak of vowel
Vowel in voice, or it is further to improve identification accuracy, language identified by the duration of resonance peak combination vowel
Vowel in sound.In a specific embodiment, resonance peak can be the first resonance peak and the second resonance peak of vowel.
Step 605 measures the first resonance peak F1 and the second resonance peak F2 of vowel, the F1 and F2 of vowel formant value
And non-linear relation, in a specific embodiment, the hertz value of formant can be converted into Bark (Bark) value, specific conversion is public
Formula is：
Bark=[(26.81 x F)/(1960+F)] -0.53.
Step 607 measures to the duration of vowel.
The data that step 609 measures according to step 605 and step 607 generate English vowel sounding acoustic analyze data.
It is the sample object progress to multiple standard English when the English vowel sounding acoustic model for generating standard
English vowel sounding acoustic is analyzed, and finally generates standard according to the English vowel sounding acoustic analyze data of multiple sample objects
English vowel sounding acoustic model.
Fig. 7 is the different nationality men and women measurands generated using English vowel sounding acoustic analysis method shown in Fig. 6 600
English vowel sounding acoustic analysis chart, be tested pair obtained by the vowel sounding acoustic analysis done for some speaker
Onomatopoeia characteristic pattern, left side are male, and right side is women, and abscissa is the F2 values of the formant of vowel second, and ordinate is vowel
The F1 values of first formant, and resonance peak F1 and F2 is converted into Bark (Bark) value by hertz value.
Top layer is the acoustic feature for the measurand that Chinese is mother tongue in figure, and the vowel sounding of measurand has no obvious pine
The differentiation of strength vowel, the interference of obvious Chinese vowel is embodied, there is obvious Chinese accent feature.Middle level is that Dutch is in figure
The acoustic feature of the measurand of mother tongue, the indivedual sounds of measurand vowel sounding have point of elastic sound, and indivedual sounds substantially have Holland
Voice negative transfer effect.Bottom is the acoustic feature of Amerenglish mother tongue measurand in figure, and the vowel sounding of measurand has
There is the differentiation of obvious elastic vowel, this acoustic feature of embodiment English native can be as the English vowel sounding acoustic mould of standard
Type.
In one embodiment, as shown in figure 8, being English vowel sounding sound shown in use Fig. 6 of one embodiment of the invention
Analysis method 600 is learned, and chooses sample object of the American as standard English, the English vowel sounding mould of the standard generated
Type figure.
It should be appreciated that the present invention does not limit the execution sequence of each step in English vowel sounding error correction method, can root
The execution sequence of each step is adjusted according to actual demand, technical scheme can be achieved.
The those skilled in the art are appreciated that each step in the English vowel sounding error correction method of the present invention
Suddenly system, method or computer program product be can be implemented as.Therefore, various aspects of the invention can be implemented as with
Lower form, i.e.,：Complete hardware embodiment, complete Software Implementation (including firmware, resident software, microcode etc.),
Or the embodiment combined in terms of hardware and software.
It should be appreciated that in each square frame and flow chart of above-mentioned flow chart each square frame combination, can be by computer
Programmed instruction is realized.These computer program instructions can be supplied to all-purpose computer, special-purpose computer or other programmable numbers
According to the processor of processing unit, so as to produce a kind of machine so that these instructions are passing through computer or other programmable numbers
According to processing unit computing device when, generate the dress of function specified in one or more of implementation process figure square frame
Put.Can being combined to write the operation for performing various aspects of the invention with one or more programming languages
Computer program code, described program design language includes object oriented program language-such as Java, C++ and C
The similar programming language such as language.
These computer program instructions can also be stored in computer-readable medium, these instruct cause computer,
Other programmable data processing units or other equipment work in a specific way, so as to be stored in computer-readable medium
Instruction just produces the instruction for including function/action specified in one or more of implementation process figure and/or block diagram square frame
Manufacture.Computer-readable recording medium can be portable computer diskette, hard disk, random access memory (RAM), read-only
Memory (ROM), erasable programmable read only memory (EPROM or flash memory), portable compact disc read only memory (CD-
ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.Computer-readable recording medium can be
It is any to include or the tangible medium of storage program, the program can be commanded execution system, device either device using or with
It is used in combination.
Computer program instructions can also be loaded into computer, other programmable data processing units or other equipment
On so that series of operation steps is performed in computer, other programmable devices or other equipment, to produce computer realization
Process so that the instruction that performs on the computer or other programmable apparatus is provided in implementation process figure and/or block diagram
One or more square frames specified in function/action process.
It should be appreciated that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations, to ability
For field technique personnel, the technical scheme described in above-described embodiment can be modified, or it is special to which part technology
Sign carries out equivalent substitution；And all such modifications and replacement, it should all belong to the protection domains of appended claims of the present invention.
Claims (30)
- A kind of 1. English vowel sounding error correction method, it is characterised in that including：Step 1, the English vowel sounding acoustic model of pre-stored criteria；Step 2, the English Phonetics of typing measurand；Step 3, the vowel in the English Phonetics of the measurand is identified；Step 4, measurand described in typing reads aloud the voice of the vowel identified；Step 5, the voice for reading aloud the measurand vowel identified carries out English vowel sounding acoustic analysis；Step 6, the English vowel sounding acoustic of the English vowel sounding acoustic analyze data of the measurand and the standard Model compares, and obtains first degree of deviation；Step 7 carries out error correction according to first degree of deviation to the English vowel sounding of the measurand.
- 2. English vowel sounding error correction method as claimed in claim 1, it is characterised in that the step 1 includes：The English Phonetics of the multiple standard English sample objects of typing；Identify the vowel in the English Phonetics of the multiple standard English sample object；English vowel sounding acoustic analysis is carried out to the vowel of each sample object respectively；The English vowel sounding acoustic model of the standard is generated according to the English vowel sounding acoustic analysis result.
- 3. English vowel sounding error correction method as claimed in claim 1, it is characterised in that the step 2 includes：According to described The nationality of measurand provides phonetic material, and measurand described in typing reads aloud the English Phonetics of the phonetic material.
- 4. English vowel sounding error correction method as claimed in claim 1, it is characterised in that the step 3 includes：According to described The resonance peak of vowel identifies the vowel in the English Phonetics of the measurand.
- 5. English vowel sounding error correction method as claimed in claim 1, it is characterised in that the step 3 includes：According to described The duration of the resonance peak of vowel and the vowel identifies the vowel in the English Phonetics of the measurand.
- 6. the English vowel sounding error correction method as described in claim 4 or 5, it is characterised in that the resonance peak includes the One resonance peak and the second resonance peak.
- 7. English vowel sounding error correction method as claimed in claim 1, it is characterised in that the step 7 includes：According to described The data of English vowel sounding acoustic analysis and the English vowel sounding acoustic model of the standard, with visual image rectification institute State the sounding of the English vowel of measurand.
- 8. English vowel sounding error correction method as claimed in claim 1, it is characterised in that after the step 7, in addition to：Measurand described in typing reads aloud the voice of the vowel identified again again；The voice read aloud again the measurand carries out English vowel sounding acoustic analysis；The English vowel sounding acoustic analyze data for the voice that the measurand is read aloud again and the English vowel of the standard Sounding acoustic model compares, and obtains second degree of deviation；The English vowel sounding that the measurand is exported according to first degree of deviation and second degree of deviation evaluates text.
- 9. the English vowel sounding error correction method as described in claim 1,2 or 8, it is characterised in that the English vowel sounding Acoustic analysis includes：Measure the resonance peak of the English vowel sounding of the typing；Measure the duration of the English vowel sounding of the typing；The English member of the measurand is generated according to the resonance peak of the English vowel sounding of the typing and the duration Sound sounding acoustic analyze data.
- 10. English vowel sounding error correction method as claimed in claim 9, it is characterised in that the resonance peak includes first Resonance peak and the second resonance peak.
- 11. a kind of storage device, wherein being stored with a plurality of instruction, the instruction is suitable for being loaded by processor and being performed：Step 1, the English vowel sounding acoustic model of pre-stored criteria；Step 2, the English Phonetics of typing measurand；Step 3, the vowel in the English Phonetics of the measurand is identified；Step 4, measurand described in typing reads aloud the voice of the vowel identified；Step 5, the voice for reading aloud the measurand vowel identified carries out English vowel sounding acoustic analysis；Step 6, the English vowel sounding acoustic of the English vowel sounding acoustic analyze data of the measurand and the standard Model compares, and obtains first degree of deviation；Step 7, error correction is carried out to the English vowel sounding of the measurand according to first degree of deviation.
- 12. storage device as claimed in claim 11, it is characterised in that the step 1 includes：The English Phonetics of the multiple standard English sample objects of typing；Identify the vowel in the English Phonetics of the multiple standard English sample object；English vowel sounding acoustic analysis is carried out to the vowel of each sample object respectively；The English vowel sounding acoustic model of the standard is generated according to the English vowel sounding acoustic analysis result.
- 13. storage device as claimed in claim 11, it is characterised in that the step 2 includes：According to the measurand Nationality provides phonetic material, and measurand described in typing reads aloud the English Phonetics of the phonetic material.
- 14. storage device as claimed in claim 11, it is characterised in that the step 3 includes：According to the resonance of the vowel Peak value identifies the vowel in the English Phonetics of the measurand.
- 15. storage device as claimed in claim 11, it is characterised in that the step 3 includes：According to the resonance of the vowel The duration of peak value and the vowel identifies the vowel in the English Phonetics of the measurand.
- 16. the storage device as described in claims 14 or 15, it is characterised in that the resonance peak includes the first resonance peak With the second resonance peak.
- 17. storage device as claimed in claim 11, it is characterised in that the step 7 includes：Sent out according to the English vowel The English vowel sounding acoustic model of the data of sound acoustic analysis and the standard, with measurand described in visual Image Adjusting The sounding of English vowel.
- 18. storage device as claimed in claim 11, it is characterised in that after the step 7, in addition to：Measurand described in typing reads aloud the voice of the vowel identified again again；The voice read aloud again the measurand carries out English vowel sounding acoustic analysis；The English vowel sounding acoustic analyze data for the voice that the measurand is read aloud again and the English vowel of the standard Sounding acoustic model compares, and obtains the second deviation；The English vowel sounding that the measurand is exported according to first degree of deviation and second deviation evaluates text.
- 19. the storage device as described in claim 11,12 or 18, it is characterised in that the English vowel sounding acoustic analysis Including：Measure the resonance peak of the English vowel sounding of the typing；Measure the duration of the vowel sounding of the typing；The English vowel that the typing is generated according to the resonance peak of the English vowel sounding of the typing and the duration is sent out Sound acoustic analysis data.
- 20. storage device as claimed in claim 19, it is characterised in that the resonance peak includes the first resonance peak and the Two resonance peaks.
- 21. a kind of English vowel sounding error correction device, including：Processor, it is adapted for carrying out each instruction；AndStorage device, suitable for storing a plurality of instruction, the instruction is suitable for being loaded and being performed by the processor：Step 1, the English vowel sounding acoustic model of pre-stored criteria；Step 2, the English Phonetics of typing measurand；Step 3, the vowel in the English Phonetics of the measurand is identified；Step 4, measurand described in typing reads aloud the voice of the vowel identified；Step 5, the voice for reading aloud the measurand vowel identified carries out English vowel sounding acoustic analysis；Step 6, the English vowel sounding acoustic of the English vowel sounding acoustic analyze data of the measurand and the standard Model compares, and obtains first degree of deviation；Step 7, error correction is carried out to the English vowel sounding of the measurand according to first degree of deviation.
- 22. English vowel sounding error correction device as claimed in claim 21, it is characterised in that the step 1 includes：The English Phonetics of the multiple standard English sample objects of typing；Identify the vowel in the English Phonetics of the multiple standard English sample object；English vowel sounding acoustic analysis is carried out to the vowel of each sample object respectively；The English vowel sounding acoustic model of the standard is generated according to the English vowel sounding acoustic analysis result.
- 23. English vowel sounding error correction device as claimed in claim 21, it is characterised in that the step 2 includes：According to institute The nationality for stating measurand provides phonetic material, and measurand described in typing reads aloud the English Phonetics of the phonetic material.
- 24. English vowel sounding error correction device as claimed in claim 21, it is characterised in that the step 3 includes：According to institute State the vowel in the English Phonetics of the resonance peak identification measurand of vowel.
- 25. English vowel sounding error correction device as claimed in claim 21, it is characterised in that the step 3 includes：According to institute The duration of the resonance peak and the vowel of stating vowel identifies the vowel in the English Phonetics of the measurand.
- 26. the English vowel sounding error correction device as described in claim 24 or 25, it is characterised in that the resonance peak includes First resonance peak and the second resonance peak.
- 27. English vowel sounding error correction device as claimed in claim 21, it is characterised in that the step 7 includes：According to institute The data of English vowel sounding acoustic analysis and the English vowel sounding acoustic model of the standard are stated, with visual Image Adjusting The sounding of the measurand English vowel.
- 28. English vowel sounding error correction device as claimed in claim 21, it is characterised in that after the step 7, in addition to：Measurand described in typing reads aloud the voice of the vowel identified again again；The voice read aloud again the measurand carries out English vowel sounding acoustic analysis；The English vowel sounding acoustic analyze data for the voice that the measurand is read aloud again and the English vowel of the standard Sounding acoustic model compares, and obtains the second deviation；The English vowel sounding that the speaker is exported according to first degree of deviation and second deviation evaluates text.
- 29. the English vowel sounding error correction device as described in claim 21,22 or 28, it is characterised in that the English vowel Sounding acoustic analysis includes：Measure the resonance peak of the English vowel sounding of the typing；Measure the duration of the vowel sounding of the typing；The English vowel that the typing is generated according to the resonance peak of the English vowel sounding of the typing and the duration is sent out Sound acoustic analysis data.
- 30. English vowel sounding error correction device as claimed in claim 29, it is characterised in that the resonance peak includes first Resonance peak and the second resonance peak.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710803552.XA CN107610691B (en) | 2017-09-08 | 2017-09-08 | English vowel sounding error correction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710803552.XA CN107610691B (en) | 2017-09-08 | 2017-09-08 | English vowel sounding error correction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107610691A true CN107610691A (en) | 2018-01-19 |
| CN107610691B CN107610691B (en) | 2021-07-06 |
Family
ID=61062775
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710803552.XA Active CN107610691B (en) | 2017-09-08 | 2017-09-08 | English vowel sounding error correction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107610691B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108766059A (en) * | 2018-05-21 | 2018-11-06 | 重庆交通大学 | A kind of cloud service English teaching equipment and teaching method |
| CN113053364A (en) * | 2019-12-27 | 2021-06-29 | 北京搜狗科技发展有限公司 | Voice recognition method and device for voice recognition |
| EP4332965A1 (en) * | 2022-08-31 | 2024-03-06 | Beats Medical Limited | System and method configured for analysing acoustic parameters of speech to detect, diagnose, predict and/or monitor progression of a condition, disorder or disease |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06110494A (en) * | 1992-09-08 | 1994-04-22 | Osaka Gas Co Ltd | Pronounciation learning device |
| CN101292283A (en) * | 2005-10-20 | 2008-10-22 | 日本电气株式会社 | Voice judging system, voice judging method, and program for voice judgment |
| CN104143332A (en) * | 2013-05-08 | 2014-11-12 | 卡西欧计算机株式会社 | VOICE PROCESSING DEVICE, and VOICE PROCESSING METHOD |
| CN106023986A (en) * | 2016-05-05 | 2016-10-12 | 河南理工大学 | Voice identification method based on sound effect mode detection |
-
2017
- 2017-09-08 CN CN201710803552.XA patent/CN107610691B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06110494A (en) * | 1992-09-08 | 1994-04-22 | Osaka Gas Co Ltd | Pronounciation learning device |
| CN101292283A (en) * | 2005-10-20 | 2008-10-22 | 日本电气株式会社 | Voice judging system, voice judging method, and program for voice judgment |
| CN104143332A (en) * | 2013-05-08 | 2014-11-12 | 卡西欧计算机株式会社 | VOICE PROCESSING DEVICE, and VOICE PROCESSING METHOD |
| CN106023986A (en) * | 2016-05-05 | 2016-10-12 | 河南理工大学 | Voice identification method based on sound effect mode detection |
Non-Patent Citations (1)
| Title |
|---|
| 王红岩: "英语元音声学分析——汉语、荷兰语及美国英语背景下的英语元音发生研究", 《外国语言文学( 季刊) 》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108766059A (en) * | 2018-05-21 | 2018-11-06 | 重庆交通大学 | A kind of cloud service English teaching equipment and teaching method |
| CN113053364A (en) * | 2019-12-27 | 2021-06-29 | 北京搜狗科技发展有限公司 | Voice recognition method and device for voice recognition |
| EP4332965A1 (en) * | 2022-08-31 | 2024-03-06 | Beats Medical Limited | System and method configured for analysing acoustic parameters of speech to detect, diagnose, predict and/or monitor progression of a condition, disorder or disease |
| WO2024047102A1 (en) * | 2022-08-31 | 2024-03-07 | Beats Medical Limited | System and method configured for analysing acoustic parameters of speech to detect, diagnose, predict and/or monitor progression of a condition, disorder or disease |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107610691B (en) | 2021-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Boë et al. | Evidence of a vocalic proto-system in the baboon (Papio papio) suggests pre-hominin speech precursors | |
| CN105206258B (en) | The generation method and device and phoneme synthesizing method and device of acoustic model | |
| Tran et al. | Improvement to a NAM-captured whisper-to-speech system | |
| Felps et al. | Foreign accent conversion through concatenative synthesis in the articulatory domain | |
| KR20150076128A (en) | System and method on education supporting of pronunciation ussing 3 dimensional multimedia | |
| Birkholz et al. | The contribution of phonation type to the perception of vocal emotions in German: An articulatory synthesis study | |
| JP2014038282A (en) | Prosody editing apparatus, prosody editing method and program | |
| CN107610691A (en) | English vowel sounding error correction method and device | |
| Kim et al. | A kinematic study of critical and non-critical articulators in emotional speech production | |
| US9484045B2 (en) | System and method for automatic prediction of speech suitability for statistical modeling | |
| JP4381404B2 (en) | Speech synthesis system, speech synthesis method, speech synthesis program | |
| JP5726822B2 (en) | Speech synthesis apparatus, method and program | |
| Ling et al. | Articulatory control of HMM-based parametric speech synthesis driven by phonetic knowledge | |
| KR102484006B1 (en) | Voice self-practice method for voice disorders and user device for voice therapy | |
| Steiner et al. | Synthesis of tongue motion and acoustics from text using a multimodal articulatory database | |
| Ling et al. | HMM-based text-to-articulatory-movement prediction and analysis of critical articulators | |
| Jones | Development of kinematic templates for automatic pronunciation assessment using acoustic-to-articulatory inversion | |
| CN112951208B (en) | Method and device for speech recognition | |
| Athanasopoulos et al. | 3D immersive karaoke for the learning of foreign language pronunciation | |
| Krňoul et al. | Training of coarticulation models using dominance functions and visual unit selection methods for audio-visual speech synthesis | |
| CN107633843A (en) | Identify the method and device of people nationality | |
| Kim et al. | Estimation of the movement trajectories of non-crucial articulators based on the detection of crucial moments and physiological constraints. | |
| Murphy | Controlling the voice quality dimension of prosody in synthetic speech using an acoustic glottal model | |
| Johnson | Speech production patterns in producing linguistic contrasts are partly determined by individual differences in anatomy | |
| i Barrobes | Voice Conversion applied to Text-to-Speech systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |