Method for generating pronunciation dictionary according to Vietnamese written text
A text and dictionary technology, applied in the field of generating pronunciation dictionaries based on Vietnamese written texts, can solve the problems affecting the quality, accuracy and applicability of pronunciation dictionaries, and achieve improved accuracy and applicability, improved applicability, and reduced usage. effect of difficulty
Active Publication Date: 2021-10-15
成都启英泰伦科技有限公司
View PDF8 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Therefore, the phoneme labeling method of the above pronunciation dictionary has obvious deficiencies in accuracy and applicability, which affects the quality of the pronunciation dictionary
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0018] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0019] The Vietnamese pronunciation coding method of the present invention includes a method of generating a pronunciation dictionary based on Vietnamese writing text, including the following steps:
[0020] Decompose the writing text to at least two types of phones, which are characterized by two characteristics of the rhyme, including three types of sounds, rhyme, and tone;
[0021] In Vietnamese, the same vowels or consonant symbols appear in different locations, and their actual pronunciation may have a clear difference: Type Tất as an example, if the phoneme is divided according to the vowel and consonant, the TấT is marked as: t â5 t, but actually 1 as a parential T and the second TA actual pronunciation of T act as a rhyme. The present invention is divided by a sound mother and a rhyme (TấT label: t-t5), the first T is the sound mother, the second T is the rhyme of the rhyme. The method can be the same as the writin...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
A method for generating a pronunciation dictionary according to a Vietnamese written text comprises the following steps that the Vietnamese written text is decomposed into phonemes at least comprising two characteristics of vowels and tones, and phonemes at most comprise the characteristics of initial consonants, vowels and tones; each phoneme is expressed by phoneme symbols which are not mutually repeated; initial consonants or vowels with different written texts but the same pronunciation are represented by the same phoneme symbols; and after all the written texts are decomposed, a corresponding relation between the written texts and phoneme symbols are generated as a pronunciation dictionary. According to the invention, the corresponding relation between the Vietnamese character writing text and the phonemes of the pronunciation dictionary is constructed on the basis of actual pronunciation, the method is suitable for model training of corpora of different sizes, and the accuracy and applicability of the pronunciation dictionary are improved. According to the pronunciation dictionary construction method adopted by the invention, special letters and tone symbols contained in Vietnamese are represented by new coded symbols so that the use difficulty of technicians in the field is reduced, and the applicability of the pronunciation dictionary is improved.
Description
Technical field [0001] The present invention belongs to the technical field of speech recognition, and more particularly to a method of generating a pronunciation dictionary in accordance with Vietnamese writing text. Background technique [0002] Voice is an important means of interacting with humans and human interactions. Since the early 1950 years, speech recognition technology has realized commercial and gradually integrated into people's daily life. But at present, there are not many research on Vietnamese language, which is limited to professional knowledge, corpus size and other factors, and the development of Vietnamese voice recognition is slow. [0003] In speech recognition technology, the pronunciation dictionary is an important part of the speech recognition system, its accuracy and applicability have an important impact on the improvement of speech recognition. The pronunciation dictionary contains mappings from words to phonemes that are used to connect acoustic m...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G10L13/02G10L13/08G10L13/10
CPCG10L13/02G10L13/08G10L13/10
Inventor 孙春玲
Owner 成都启英泰伦科技有限公司