Supercharge Your Innovation With Domain-Expert AI Agents!

A phonetic symbol decomposition and its synthesis method

A synthesis method and technology of phonetic symbols, applied in speech synthesis, speech analysis, instruments, etc., can solve the problems of uncoordinated tones, decreased voice naturalness, high calculation cost, etc., and achieve improved quality, less resource occupation, and simple calculation methods Effect

Inactive Publication Date: 2008-09-10
SUNPLUS TECH CO LTD +1
View PDF1 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] 1) This method simply decomposes phonetic symbols based on vowels and consonants in phonetics, resulting in incongruous tones, decreased naturalness of the voice, and even inconsistencies between the front and rear sounds due to inconsistent waveform frequencies and lack of transitional sounds in the later stage of waveform splicing and synthesis. ;
[0006] 2) This method sets a large number of rules to decompose the synthetic phonetic symbols, and these rules are unique to the English K.K. phonetic symbol and are not applicable to other phonetic symbols or languages;
[0007] 3) The calculation method involved in this method is relatively complicated, and the calculation cost is relatively high

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A phonetic symbol decomposition and its synthesis method
  • A phonetic symbol decomposition and its synthesis method
  • A phonetic symbol decomposition and its synthesis method

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0079] Embodiment 1 is described by taking English speech as an example. figure 1 It is a flow chart of decomposition and synthesis of phonetic symbols. Such as figure 1 As shown, the phonetic symbol decomposition and synthesis process includes the following steps:

[0080] Step 1 is syllable splitting: convert phonetic phonetic symbols into text phonetic codes, decompose according to waveform characteristics, and form segmentation units;

[0081] Step 2 is naming the syllable: according to the decomposed segmentation unit, name the segmentation unit according to the rules to form a pronunciation pronunciation unit;

[0082] Step 3 is speech synthesis: the phoneme waveforms corresponding to the speech pronunciation units are concatenated for speech synthesis.

[0083] Such as figure 2 As shown, step 1 syllable splitting further includes three steps:

[0084] Step 11 is preprocessing: convert known phonetic symbol symbols into text phonetic symbol code, form text phonetic...

specific Embodiment approach 2

[0155] Embodiment 2 is described by taking Chinese speech as an example. The process of decomposing and synthesizing the phonetic symbols of Chinese synthetic vocabulary and short sentences is basically consistent with the process of decomposing and synthesizing the phonetic symbols of English phonetic symbols in the specific embodiment 1, except that:

[0156] The basis for adding the segmentation mark at the corresponding position of the text phonetic symbol code sequence is different. According to the phoneme waveform features, the phonetic codes of Chinese speech can be divided into two categories:

[0157] The first type of collection, non-periodic waveform characteristic consonants, including pinyin are:

[0158] / b / , / p / , / f / , / d / , / t / , / g / , / k / , / h / , / j / , / q / , / x / , / zh / , / ch / , / sh / , / z / , / c / , / s / , a total of 17;

[0159] The second type of collection includes periodic waveform characteristic consonants and periodic waveform characteristic vowels, wherein the period...

specific Embodiment approach 3

[0168] Embodiment 3 is described by taking Japanese voice as an example. The process of decomposing and synthesizing the phonetic symbols of Japanese synthetic words, phrases or short sentences is basically consistent with the process of decomposing and synthesizing English phonetic symbols in the specific embodiment 1, except that:

[0169] The basis for adding the segmentation mark at the corresponding position of the text phonetic symbol code sequence is different. According to the phoneme waveform characteristics, the phonetic codes of Japanese speech can be divided into two categories:

[0170] The first type of collection, the head pronunciation is hiragana with non-periodic waveform features, including:

[0171] か(ka) line, さ(sa) line, た(ta) line, ほ(ha) line and their corresponding voiced, semi-voiced and obtrusive lines, totaling 72;

[0172] The second type of collection, the head pronunciation is hiragana with periodic waveform features, including:

[0173] な(na) li...

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

No PUM Login to View More

Abstract

The invention discloses a method for decomposing and synthesizing phonetic alphabets. At first, the phonetic alphabet of a phonetic sound is decomposed by utilizing cutting marks to form subunits; then the subunits are converted into pronunciation units of the phonetic sound; finally, a phoneme waveform corresponding to the pronunciation units of the phonetic sound is connected in series to realize the speech synthesis, wherein, the phonetic alphabet of the phonetic sound is decomposed by utilizing the cutting marks to carry out cutting processing to consonants with nonperiodic waveform characteristics. The invention can decompose the phonetic alphabet by category by analyzing the waveform characteristics corresponding to each phoneme of the phonetic sound to reduce obvious connection trace which is generated by the conventional speech synthesis so as to lead the effect of the synthesized phonetic sound to be similar to the voice of a true person to the greatest extent.

Description

technical field [0001] The invention relates to the technical field of computer speech processing, in particular to phonetic symbol decomposition and synthesis technology. Background technique [0002] Language is a bridge for human beings to communicate with each other and the most convenient way to obtain information. Therefore, learning the unique languages ​​of human beings and realizing fast and effective human-computer interaction are important research topics in the field of speech. [0003] In recent years, various electronic dictionary products have led the trend of language learning. In order to improve the competitiveness of their products, various manufacturers have put a lot of effort into the quality and naturalness of voice, competing to promote the functions of human pronunciation and massive vocabulary. At present, the "human pronunciation" function of various electronic dictionaries on the market is mainly realized through two methods: whole word compress...

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
no application Login to View More
IPC IPC(8): G10L13/00G10L13/02G10L13/08G10L13/10
Inventor 赵妍刘扬山
Owner SUNPLUS TECH CO LTD
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com