CN107610691B - English vowel pronunciation error correction method and device - Google Patents
English vowel pronunciation error correction method and device Download PDFInfo
- Publication number
- CN107610691B CN107610691B CN201710803552.XA CN201710803552A CN107610691B CN 107610691 B CN107610691 B CN 107610691B CN 201710803552 A CN201710803552 A CN 201710803552A CN 107610691 B CN107610691 B CN 107610691B
- Authority
- CN
- China
- Prior art keywords
- english
- vowel
- sounding
- tested object
- voice
- 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.)
- Active
Links
Images
Landscapes
- Electrically Operated Instructional Devices (AREA)
Abstract
本发明涉及一种英语元音发声纠错方法，包括：步骤1，预存标准的英语元音发声声学模型；步骤2，录入被测对象的英语语音；步骤3，识别被测对象的英语语音中的元音；步骤4，录入被测对象朗读所述识别出的元音的语音；步骤5，对被测对象朗读所述识别出的元音的语音进行英语元音发声声学分析；步骤6，被测对象的英语元音发声声学分析数据与标准的英语元音发声声学模型比较，得到第一偏差度；步骤7，根据第一偏差度对被测对象的英语元音发声进行纠错。通过对被测对象的英语元音发声声学进行分析，并与预存的标准的英语元音发声声学模型相比较，从而对被测对象的英语元音发声进行纠正，以使得被测对象的英语元音发声更准确。
The invention relates to an error correction method for English vowel pronunciation, comprising: step 1, pre-storing a standard acoustic model of English vowel pronunciation; step 2, inputting the English voice of a tested object; step 3, identifying the English voice of the tested object in Step 4, enter the voice of the identified vowel that is read aloud by the tested object; Step 5, carry out the acoustic analysis of English vowel vocalization to the voice of the tested object reading the identified vowel; Step 6, Compare the English vowel vocalization acoustic analysis data of the test object with the standard English vowel vocalization acoustic model to obtain a first degree of deviation; step 7, correct the English vowel vocalization of the test object according to the first degree of deviation. By analyzing the English vowel vocalization acoustics of the test object and comparing it with the pre-stored standard English vowel vocalization acoustic model, the English vowel vocalization of the test object is corrected, so that the English vowel sound of the test object is correct. sound is more accurate.
Description
Technical Field
The invention relates to the technical field of voice recognition, in particular to a method and a device for correcting pronunciation of English vowels.
Background
Language is a bridge for human communication, and the diversity of language is also a barrier to communication. English is the leading common language with high frequency of use. The "english variant" of different pronunciation features in different countries of the world becomes a barrier in communication. The research on the English characteristics under different mother language backgrounds is the key for breaking through communication barriers and improving the English pronunciation of speakers. The recognition of the voice characteristics of the speaker is an irrevocable process for recognizing the voice content of the speaker. In order to improve the pronunciation of the speaker in English and improve the communication barriers between people of different nationalities, it is very important to correct the pronunciation of the speaker in English when the speaker's pronunciation deviates from the target pronunciation specification by a certain range.
When in pronunciation, the airflow exhaled from the lung passes through the oral cavity which plays the role of a resonator, and the voice with extremely low resistance and no friction sound is produced. Although the vocal cords vibrate in the case of a vowel sound in a general case, the vocal cords may not vibrate and may be uttered as unvoiced or aural sounds.
From the point of pronunciation phonetics, vowels are generally classified by tongue position and lip shape, high vowels are pronounced with their tongue surfaces arched against the palate, low vowels are pronounced with their tongue portions relatively flat, the tongue surfaces are slightly spaced from the palate, middle vowels are pronounced with their tongue positions in the middle, and high, middle and low vowels are also classified in front and rear rows. The position of the lingual surface and the shape of the lips are one criterion for vowel classification.
According to pronunciation phonetics, vowels are a class of pronunciation formed by airflow vibration vocal cords and no obstruction in the oral cavity, and different vowels are formed by different oral cavity shapes; the consonants are formed by blocking airflow in the oral cavity, and different consonants are formed at different pronunciation parts or pronunciation methods.
Disclosure of Invention
The invention aims to realize the error correction of English vowel sounding of a tested object based on the comparison between the tested object and a standard English vowel sounding model when English is a target language.
The invention provides an English vowel sounding error correction method on one hand, which comprises the following steps: step 1, pre-storing a standard English vowel sounding acoustic model; step 2, inputting English voice of the tested object; step 3, identifying vowels in English voices of the tested object; step 4, inputting the voice of the recognized vowel read by the tested object; step 5, performing English vowel sounding acoustic analysis on the voice of the identified vowel read by the tested object; step 6, comparing the English vowel sounding acoustic analysis data of the tested object with the standard English vowel sounding acoustic model to obtain a first deviation degree; and 7, correcting the English vowel sounding of the tested object according to the first deviation degree.
The step 1 comprises the following steps: inputting English voices of a plurality of standard English sample objects; identifying vowels in the English speech of the plurality of standard English sample objects; performing English vowel sounding acoustic analysis on vowels of each sample object respectively; and generating the standard English vowel sounding acoustic model according to the English vowel sounding acoustic analysis result.
The step 2 comprises the following steps: and providing the voice material according to the nationality of the tested object, and recording English voice of the tested object for reading the voice material.
The step 3 comprises the following steps: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel.
The step 3 comprises the following steps: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel and the duration of the vowel.
The resonance peaks include a first resonance peak and a second resonance peak.
The step 7 comprises the following steps: and adjusting the sound production of the English vowel of the tested object by a visual image according to the data of the English vowel sound production acoustic analysis and the standard English vowel sound production acoustic model.
After the step 7, the method further comprises the following steps: recording the voice of the vowel read again by the tested object; performing acoustic analysis on the pronunciation of the measured object read again on the English vowel; comparing the English vowel sounding acoustic analysis data of the voice read again by the tested object with the standard English vowel sounding acoustic model to obtain a second deviation degree; and outputting an English vowel sounding evaluation text of the tested object according to the first deviation degree and the second deviation degree.
The acoustic analysis of the English vowel sounding comprises the following steps: measuring the resonance peak value of the input English vowel sounding; measuring the sounding duration of the input English vowel; and generating the acoustic analysis data of the input English vowel sounding according to the resonance peak value of the input English vowel sounding and the time length.
The resonance peaks include a first resonance peak and a second resonance peak.
The present invention also provides a memory device having stored therein a plurality of instructions adapted to be loaded and executed by a processor to: step 1, pre-storing a standard English vowel sounding acoustic model; step 2, inputting English voice of the tested object; step 3, identifying vowels in English voices of the tested object; step 4, inputting the voice of the recognized vowel read by the tested object; step 5, performing English vowel sounding acoustic analysis on the voice of the identified vowel read by the tested object; step 6, comparing the English voice sounding acoustic analysis data of the tested object with the standard English vowel sounding acoustic model to obtain a first deviation degree; and 7, correcting the English vowel sounding of the tested object according to the first deviation degree.
The step 1 comprises the following steps: inputting English voices of a plurality of standard English sample objects; identifying vowels in the English speech of the plurality of standard English sample objects; performing English vowel sounding acoustic analysis on vowels of each sample object respectively; and generating the standard English vowel sounding acoustic model according to the English vowel sounding acoustic analysis result.
The step 2 comprises the following steps: and providing a voice material according to the nationality of the tested object, and recording English voice of the tested object for reading the voice material.
The step 3 comprises the following steps: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel.
The step 3 further comprises: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel and the duration of the vowel.
The resonance peaks include a first resonance peak and a second resonance peak.
The step 7 comprises the following steps: and adjusting the sound production of the English vowel of the tested object by a visual image according to the data of the English vowel sound production acoustic analysis and the standard English vowel sound production acoustic model.
After the step 7, the method further comprises the following steps: recording the voice of the vowel read again by the tested object; performing acoustic analysis on the pronunciation of the measured object read again on the English vowel; comparing the English vowel sounding acoustic analysis data of the voice read again by the tested object with the standard English vowel sounding acoustic model to obtain a second deviation degree; and outputting an English vowel sounding evaluation text of the tested object according to the first deviation degree and the second deviation degree.
The acoustic analysis of the English vowel sounding comprises the following steps: measuring the resonance peak value of the input English vowel sounding; measuring the sounding duration of the input vowel; and generating the acoustic analysis data of the input English vowel sounding according to the resonance peak value of the input English vowel sounding and the time length.
The resonance peaks include a first resonance peak and a second resonance peak.
The invention also provides an English vowel sounding error correction device, which comprises: a processor adapted to implement instructions; and a storage device adapted to store a plurality of instructions, the instructions adapted to be loaded and executed by the processor to: step 1, pre-storing a standard English vowel sounding acoustic model; step 2, inputting English voice of the tested object; step 3, identifying vowels in English voices of the tested object; step 4, inputting the voice of the recognized vowel read by the tested object; step 5, performing English vowel sounding acoustic analysis on the voice of the identified vowel read by the tested object; step 6, comparing the acoustic analysis of the English vowel sounding of the tested object with the acoustic model of the standard English vowel sounding to obtain a first deviation degree; and 7, correcting the English vowel sounding of the tested object according to the first deviation degree.
The step 1 comprises the following steps: inputting English voices of a plurality of standard English sample objects; identifying vowels in the English speech of the plurality of standard English sample objects; performing English vowel sounding acoustic analysis on vowels of each sample object respectively; and generating the standard English vowel sounding acoustic model according to the English vowel sounding acoustic analysis result.
The step 2 comprises the following steps: and providing voice materials according to nationality of the sample object, and inputting English voices of the sample object for reading the voice materials.
The step 3 comprises the following steps: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel.
The step 3 comprises the following steps: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel and the duration of the vowel.
The resonance peaks include a first resonance peak and a second resonance peak.
The step 7 comprises the following steps: and adjusting the sound production of the English vowel of the tested object by a visual image according to the data of the English vowel sound production acoustic analysis and the standard English vowel sound production acoustic model.
After the step 7, the method further comprises the following steps: recording the voice of the vowel read again by the tested object; performing acoustic analysis on the pronunciation of the measured object read again on the English vowel; comparing the English vowel sounding acoustic analysis data read again by the tested object with the standard English vowel sounding acoustic model to obtain a second deviation degree; and outputting an English vowel sounding evaluation text of the tested object according to the first deviation degree and the second deviation degree.
The acoustic analysis of the English vowel sounding comprises the following steps: measuring the resonance peak value of the input English vowel sounding; measuring the sounding duration of the input vowel; and generating the acoustic analysis data of the input English vowel sounding according to the resonance peak value of the input English vowel sounding and the time length.
The resonance peaks include a first resonance peak and a second resonance peak.
The method has the advantages that the English vowel sounding of the tested object is analyzed and compared with the pre-stored standard English vowel sounding acoustic model, so that the English vowel sounding of the tested object is corrected, and the English vowel sounding of the tested object is more accurate.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flow chart of a method 100 for correcting spoken English vowels in accordance with an embodiment of the present invention;
FIG. 2 is a flow diagram of a method 200 for selecting a sample object in a standard English vowel sounds model in accordance with an embodiment of the present invention;
FIG. 3 is a flow diagram of a method 300 of forming a standard English vowel sound generation model in accordance with an embodiment of the present invention;
FIG. 4 is a visual English vowel utterance correction map of the English vowel utterance correction method of an embodiment of the present invention;
FIG. 5 is a flowchart of an English vowel utterance correction method 500 according to yet another embodiment of the present invention;
FIG. 6 is a flow diagram of an English vowel utterance analysis method 600 in accordance with an embodiment of the present invention;
FIG. 7 is an acoustic analysis diagram of English vowel sounds generated by the method 600 of generating English vowel sounds of FIG. 6 for male and female subjects in different countries according to an embodiment of the present invention;
fig. 8 is a diagram of a standard english vowel utterance model generated using the english vowel utterance acoustic analysis method 600 shown in fig. 6 according to an embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
As shown in fig. 1, a flowchart of an english vowel utterance correction method 100 according to an embodiment of the present invention is shown.
And 103, inputting English voices of the tested object. In a specific implementation, a voice material may be provided for the tested object to read, and the tested object may also read any other english word or sentence. When speaking according to the phonetic material is selected, the same phonetic material as when the standard english vowel sound production model is established may be provided, and a phonetic material different from when the standard english vowel sound production model is established may be provided.
And 105, identifying the vowel in the English voice of the recorded tested object.
And step 107, recording the voice of the vowel read by the tested object.
And step 109, performing English vowel vocal acoustic analysis on the recorded voice of the identified vowel read by the tested object, and obtaining the vowel vocal acoustic analysis data.
And step 111, comparing the obtained acoustic analysis data of the English vowel sounds with a pre-stored standard English vowel sound production model to obtain a first deviation degree.
And 113, correcting the English vowel sounding of the tested object according to the first deviation degree.
In a specific embodiment, after inputting english voice read aloud by a certain tested object, firstly, identifying vowel in the english voice, inputting the voice of the recognized vowel read aloud by the tested object, reading the voice of the recognized vowel aloud by the tested object, performing english vowel vocal analysis on the voice of the recognized vowel, comparing obtained english vowel vocal analysis data with a pre-stored standard english vowel vocal model, obtaining the difference between english vowel pronunciation of the tested object and standard english vowel pronunciation, and correcting the english vowel pronunciation of the tested object according to the difference.
In one embodiment, a standard english sample object is selected to establish a standard english vowel acoustic model, which can be accomplished by using the method of fig. 2.
In step 205, selection of the standard english sample object representation can select listeners with the same language background as the sample object in the united states for intermediate sensing, and select male and female speakers with middle perception, which are the most representative sample objects.
And step 207, performing overall English vowel vocal acoustic analysis on the most representative sample object which is perceived in the middle stage to form a standard English vowel vocal acoustic model.
In one embodiment, after the most representative population of standard english utterances is selected by the method shown in fig. 2, a standard english vowel utterance acoustic model can be generated by the method shown in fig. 3.
In one embodiment, the voice material can be provided according to the nationality of the tested object, and the English voice of the tested object reading the voice material is recorded. For example, for Chinese, the higher error rate vowels are: e: >
/、/ε～
E/and/u: >
If the voice material is provided, more voice materials containing the vowels can be provided, so that the vowels which are frequently wrong in the tested object can be corrected more, and the correction of the pronunciation of the English vowel is more targeted.
In one embodiment, correcting the english vowel sounds of the subject according to the first degree of deviation may include: and adjusting the sounding of the English vowel of the tested object according to the data of the English vowel sounding acoustic analysis and the standard English vowel sounding acoustic model by using a visual image. As shown in fig. 4, the visual english vowel utterance correction chart of the english vowel utterance correction method according to the embodiment of the present invention is shown, for example, when an english vowel utterance of a chinese is corrected, a provided speech material is "a good speaking mean", and the english vowel utterance correction method according to the present invention is used to correct an english vowel utterance of a chinese
The pronunciation is corrected by obtaining the coordinate positions of the first formant and the second formant of the vowel pronunciation of the Chinese shown by the dots in FIG. 4, and the triangle in FIG. 4
The coordinate positions of the first formant and the second formant of the standard vocalization, through the visual comparison, the speaker can know how to adjust the pronunciation of the English vowel more intuitively.
Fig. 5 is a flowchart of an english vowel utterance correction method 500 according to another embodiment of the present invention, where steps 501 to 513 are the same as steps 101 to 113 in fig. 1, and step 515 is to re-enter the voice of the vowel spoken by the subject; 517, performing acoustic analysis of English vowel pronunciation on the pronunciation read again by the tested object; step 519, comparing the data of the acoustic analysis of the English vowel sounds obtained in the step 517 with a standard acoustic model of the English vowel sounds to obtain a second degree of deviation; and step 521, outputting an English vowel sounding evaluation text of the tested object according to the first deviation degree and the second deviation.
In a specific implementation mode, the English vowel sounding evaluation text can include information such as an original English vowel sounding acoustic analysis diagram of the tested object and a corrected English vowel sounding acoustic analysis diagram, so that the tested object can know the English sounding condition of the tested object and the sounding problem needing to be corrected, the tested object can carry out purposeful English sounding practice according to characteristics of the tested object, and the English sounding of the tested object is improved.
When generating a standard acoustic model for english vowel vocalization or performing acoustic analysis of english vowel vocalization on a test object, the acoustic analysis method for english vowel vocalization shown in fig. 6 may be used.
Step 601 first enters english language speech. When a standard English vowel sounding acoustic model is generated, the voice of a sample object with standard English pronunciation is input; when the acoustic analysis of English vowel vocalization is performed on the tested object, the voice of the tested object is recorded.
Step 603 identifies vowels in the english language. In one embodiment, vowels in speech may be identified based on their formants, or may be identified based on the combination of the formants and the durations of the vowels in order to further improve the accuracy of the identification. In a specific embodiment, the resonance peak may be a first resonance peak and a second resonance peak of a vowel.
Step 605 measures the first resonance peak F1 and the second resonance peak F2 of the vowel, and F1 and F2 of the resonance peaks of the vowel are not linear, and in an embodiment, the hertz value of the resonance peaks can be converted into Bark (Bark) value by the following formula:
Bark＝[(26.81 x F)/(1960+F)]–0.53。
step 607 measures the duration of the vowel.
Step 609 generates acoustic analysis data of the english vowel utterance from the data measured in steps 605 and 607.
When the method is used for generating the standard English vowel sounding acoustic model, English vowel sounding acoustic analysis is carried out on a plurality of standard English sample objects, and finally the standard English vowel sounding acoustic model is generated according to English vowel sounding acoustic analysis data of the plurality of sample objects.
Fig. 7 is an english vowel vocal analysis chart of the male and female subjects of different nationalities generated by the english vowel vocal analysis method 600 shown in fig. 6, which is a subject acoustic feature chart obtained by the vowel vocal analysis performed on a speaker, the left side is a male, the right side is a female, the abscissa is the F2 value of the second formant of the vowel, the ordinate is the F1 value of the first formant of the vowel, and the formants F1 and F2 are converted from hertz values to Bark (Bark) values.
The top layer in the figure is the acoustic characteristics of the tested object with Chinese as the mother language, the vowel sounding of the tested object is not distinguished by obvious loose vowel, the obvious interference of Chinese vowel is reflected, and the Chinese accent characteristics are obvious. The middle layer in the figure is the acoustic characteristics of a tested object with Dutch as a mother language, the vowel sounding of the tested object has the difference of tightness, and the individual tone obviously has the Dutch voice negative migration effect. The bottom layer in the figure is the acoustic characteristics of the American English native language measured object, the vowel sounding of the measured object has the distinction of obvious elasticity vowels, the acoustic characteristics of the English native language are embodied, and the acoustic characteristics can be used as a standard English vowel sounding acoustic model.
In one embodiment, as shown in fig. 8, a standard english vowel vocal model diagram is generated by using the english vowel vocal acoustic analysis method 600 shown in fig. 6 and selecting americans as sample objects in standard english according to an embodiment of the present invention.
It should be understood that the present invention does not limit the execution sequence of each step in the english vowel sound generation correction method, and the execution sequence of each step can be adjusted according to actual requirements, so that the technical solution of the present invention can be implemented.
As will be appreciated by one skilled in the art, each of the steps of the English vowel voicing correction methods of the present invention may be embodied as a system, method, or computer program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining hardware and software aspects.
It will be understood that each block of the flowchart illustrations, and combinations of blocks in the flowchart illustrations, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart block or blocks. Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or C.
These computer program instructions may also be stored in a computer readable medium that can direct a computer, other programmable data processing apparatus, or other devices to function in a particular manner, such that the instructions stored in the computer readable medium produce an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks. The computer readable storage medium may be a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. A computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
It should be understood that the above embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same, and those skilled in the art can modify the technical solutions described in the above embodiments, or make equivalent substitutions for some technical features; and all such modifications and alterations are intended to fall within the scope of the appended claims.
Claims (21)
1. An English vowel sounding error correction method is characterized by comprising the following steps:
step 1, pre-storing a standard English vowel sounding acoustic model, wherein the English vowel sounding acoustic model is generated according to formants and duration of vowel sounding in English voices of a plurality of standard English sample objects;
step 2, inputting English voice of the tested object;
step 3, identifying vowels in English voices of the tested object;
step 4, inputting the voice of the vowel recognized in the step 3 read by the tested object;
step 5, performing English vowel sounding acoustic analysis on the voice of the vowel read by the tested object, wherein the method comprises the following steps:
measuring the resonance peak value of the input English vowel sounding;
measuring the sounding duration of the input English vowel;
generating English vowel sounding acoustic analysis data of the tested object according to the recorded resonance peak value and the time length of the English vowel sounding;
step 6, comparing the English vowel sounding acoustic analysis data of the tested object with the standard English vowel sounding acoustic model to obtain a first deviation degree;
step 7, correcting the sound production of English vowels of the tested object by a visual image according to the first deviation degree;
step 8, recording the voice of the vowel read again by the tested object; performing acoustic analysis on the pronunciation of the measured object read again on the English vowel; comparing the English vowel sounding acoustic analysis data of the voice read again by the tested object with the standard English vowel sounding acoustic model to obtain a second deviation degree; and outputting an English vowel sounding evaluation text of the tested object according to the first deviation degree and the second deviation degree, wherein the English vowel sounding evaluation text comprises an original English vowel sounding acoustic analysis graph and a corrected English vowel sounding acoustic analysis graph of the tested object.
2. The english vowel sound utterance correction method according to claim 1, wherein the step 2 includes: and providing a voice material according to the nationality of the tested object, and recording English voice of the tested object for reading the voice material.
3. The english vowel sound utterance correction method according to claim 1, wherein the step 3 includes: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel.
4. The english vowel sound utterance correction method according to claim 1, wherein the step 3 includes: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel and the duration of the vowel.
5. The english vowel sound correction method according to claim 3 or 4, wherein the formant values include a first formant value and a second formant value.
6. The english vowel utterance correction method of claim 1, wherein the step 8 of the acoustic analysis of the english vowel utterance comprises:
measuring the resonance peak value of the input English vowel sounding;
measuring the sounding duration of the input English vowel;
and generating English vowel sounding acoustic analysis data of the tested object according to the recorded resonance peak value of the English vowel sounding and the time length.
7. The english vowel sound correction method of claim 6, wherein the formant values include a first formant value and a second formant value.
8. A memory device having stored therein a plurality of instructions adapted to be loaded and executed by a processor to:
step 1, pre-storing a standard English vowel sounding acoustic model, wherein the English vowel sounding acoustic model is generated according to formants and duration of vowel sounding in English voices of a plurality of standard English sample objects;
step 2, inputting English voice of the tested object;
step 3, identifying vowels in English voices of the tested object;
step 4, inputting the voice of the vowel recognized in the step 3 read by the tested object;
step 5, performing English vowel sounding acoustic analysis on the voice of the vowel read by the tested object, wherein the method comprises the following steps:
measuring the resonance peak value of the input English vowel sounding;
measuring the sounding duration of the input English vowel;
generating English vowel sounding acoustic analysis data of the tested object according to the recorded resonance peak value and the time length of the English vowel sounding;
step 6, comparing the English vowel sounding acoustic analysis data of the tested object with the standard English vowel sounding acoustic model to obtain a first deviation degree;
step 7, correcting the sounding of English vowels of the tested object by a visual image according to the first deviation degree;
step 8, recording the voice of the vowel read again by the tested object; performing acoustic analysis on the pronunciation of the measured object read again on the English vowel; comparing the English vowel sounding acoustic analysis data of the voice read again by the tested object with the standard English vowel sounding acoustic model to obtain a second deviation; and outputting an English vowel sounding evaluation text of the tested object according to the first deviation degree and the second deviation, wherein the English vowel sounding evaluation text comprises an original English vowel sounding acoustic analysis chart of the tested object and a corrected English vowel sounding acoustic analysis chart.
9. The storage device of claim 8, wherein the step 2 comprises: and providing a voice material according to the nationality of the tested object, and recording English voice of the tested object for reading the voice material.
10. The storage device of claim 8, wherein the step 3 comprises: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel.
11. The storage device of claim 8, wherein the step 3 comprises: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel and the duration of the vowel.
12. The memory device of claim 10 or 11, wherein the resonance peak comprises a first resonance peak and a second resonance peak.
13. The storage device of claim 8, wherein the acoustic analysis of english vowel sounds in step 8 comprises:
measuring the resonance peak value of the input English vowel sounding;
measuring the sounding duration of the input vowel;
and generating the acoustic analysis data of the input English vowel sounding according to the resonance peak value of the input English vowel sounding and the time length. Is ═ i
14. The memory device of claim 13, wherein the resonance peak comprises a first resonance peak and a second resonance peak.
15. An English vowel sound generation error correction device, comprising:
a processor adapted to implement instructions; and
a storage device adapted to store a plurality of instructions, the instructions adapted to be loaded and executed by the processor to:
step 1, pre-storing a standard English vowel sounding acoustic model, wherein the English vowel sounding acoustic model is generated according to formants and duration of vowel sounding in English voices of a plurality of standard English sample objects;
step 2, inputting English voice of the tested object;
step 3, identifying vowels in English voices of the tested object;
step 4, inputting the voice of the vowel recognized in the step 3 read by the tested object;
step 5, performing English vowel sounding acoustic analysis on the voice of the vowel read by the tested object, wherein the method comprises the following steps:
measuring the resonance peak value of the input English vowel sounding;
measuring the sounding duration of the input English vowel;
generating English vowel sounding acoustic analysis data of the tested object according to the recorded resonance peak value and the time length of the English vowel sounding;
step 6, comparing the English vowel sounding acoustic analysis data of the tested object with the standard English vowel sounding acoustic model to obtain a first deviation degree;
step 7, correcting the sounding of English vowels of the tested object by a visual image according to the first deviation degree;
step 8, recording the voice of the vowel read again by the tested object; performing acoustic analysis on the pronunciation of the measured object read again on the English vowel; comparing the English vowel sounding acoustic analysis data of the voice read again by the tested object with the standard English vowel sounding acoustic model to obtain a second deviation; and outputting an English vowel sounding evaluation text of the tested object according to the first deviation degree and the second deviation, wherein the English vowel sounding evaluation text comprises an original English vowel sounding acoustic analysis chart of the tested object and a corrected English vowel sounding acoustic analysis chart.
16. The english vowel sound generation correction apparatus of claim 15, wherein said step 2 comprises: and providing a voice material according to the nationality of the tested object, and recording English voice of the tested object for reading the voice material.
17. The english vowel sound generation correction apparatus of claim 15, wherein said step 3 comprises: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel.
18. The english vowel sound generation correction apparatus of claim 15, wherein said step 3 comprises: and identifying the vowel in the English voice of the tested object according to the resonance peak value of the vowel and the duration of the vowel.
19. The english vowel sound generation correction apparatus of claim 17 or 18, wherein the resonance peak includes a first resonance peak and a second resonance peak.
20. The english vowel sound correction apparatus of claim 15, wherein said step 8 of analyzing the acoustic analysis of the english vowel sound includes:
measuring the resonance peak value of the input English vowel sounding;
measuring the sounding duration of the input vowel;
and generating the acoustic analysis data of the input English vowel sounding according to the resonance peak value of the input English vowel sounding and the time length.
21. The english vowel sound generation correction apparatus of claim 20, wherein the resonance peak includes a first resonance peak and a 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 pronunciation 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 pronunciation error correction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107610691A CN107610691A (en) | 2018-01-19 |
| CN107610691B true 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 pronunciation error correction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107610691B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108766059B (en) * | 2018-05-21 | 2020-09-01 | 重庆交通大学 | Cloud service English teaching equipment and teaching method |
| CN113053364B (en) * | 2019-12-27 | 2024-11-15 | 北京搜狗科技发展有限公司 | A speech recognition method, device and device for speech 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 |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06110494A (en) * | 1992-09-08 | 1994-04-22 | Osaka Gas Co Ltd | Pronounciation learning device |
| WO2007046267A1 (en) * | 2005-10-20 | 2007-04-26 | Nec Corporation | Voice judging system, voice judging method, and program for voice judgment |
| JP6171544B2 (en) * | 2013-05-08 | 2017-08-02 | カシオ計算機株式会社 | Audio processing apparatus, audio processing method, and program |
| CN106023986B (en) * | 2016-05-05 | 2019-08-30 | 河南理工大学 | A Speech Recognition Method Based on Sound Effect Pattern Detection |
-
2017
- 2017-09-08 CN CN201710803552.XA patent/CN107610691B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN107610691A (en) | 2018-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wright et al. | Factors of lexical competition in vowel articulation | |
| Kochanski et al. | Loudness predicts prominence: Fundamental frequency lends little | |
| CN101292283B (en) | Voice discrimination system and voice discrimination method | |
| KR20190008137A (en) | Apparatus for deep learning based text-to-speech synthesis using multi-speaker data and method for the same | |
| JP2007122004A (en) | Pronunciation diagnosis apparatus, pronunciation diagnosis method, recording medium, and pronunciation diagnosis program | |
| Mertens | Polytonia: a system for the automatic transcription of tonal aspects in speech corpora | |
| US9147392B2 (en) | Speech synthesis device and speech synthesis method | |
| Li et al. | Speaker verification based on the fusion of speech acoustics and inverted articulatory signals | |
| Gao et al. | A study on robust detection of pronunciation erroneous tendency based on deep neural network. | |
| CN104835493A (en) | Speech synthesis dictionary generation apparatus and speech synthesis dictionary generation method | |
| KR20160122542A (en) | Method and apparatus for measuring pronounciation similarity | |
| CN107610691B (en) | English vowel pronunciation error correction method and device | |
| Etman et al. | American dialect identification using phonotactic and prosodic features | |
| JP2014066779A (en) | Voice recognition device and method, and semiconductor integrated circuit device | |
| JP2006171750A (en) | Feature vector extracting method for speech recognition | |
| JP5754141B2 (en) | Speech synthesis apparatus and speech synthesis program | |
| Alqadasi et al. | Improving Automatic Forced Alignment for Phoneme Segmentation in Quranic Recitation | |
| Ali et al. | Development and analysis of speech emotion corpus using prosodic features for cross linguistics | |
| WO2007114346A1 (en) | Speech recognition device | |
| Bouferroum et al. | CV coarticulation, locus and locus equation perspective on the invariance issue involving Algerian Arabic consonants | |
| JP2011180308A (en) | Voice recognition device and recording medium | |
| Barczewska et al. | Detection of disfluencies in speech signal | |
| Jones | Development of kinematic templates for automatic pronunciation assessment using acoustic-to-articulatory inversion | |
| CN107633843B (en) | Method and device for identifying nationality of a person | |
| Lertwongkhanakool et al. | An automatic real-time synchronization of live speech with its transcription approach |
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 | ||
| OL01 | Intention to license declared | ||
| OL01 | Intention to license declared |