A Front-end Processing Method of Mongolian Speech Synthesis Based on HMM

A speech synthesis and front-end processing technology, applied in speech synthesis, speech analysis, speech recognition, etc., can solve the problems of limited range of pitch period and duration adjustment, high synthesis cost, etc., and achieve the effect of maintaining natural fluency and reducing costs

Active Publication Date: 2016-08-17
INNER MONGOLIA UNIVERSITY
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] The purpose of the embodiments of the present invention is to provide a front-end processing method for Mongolian speech synthesis based on HMM, aiming to solve the problems of high synthesis cost and limited range of pitch period and duration adjustment in the existing Mongolian speech synthesis method

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 Front-end Processing Method of Mongolian Speech Synthesis Based on HMM
  • A Front-end Processing Method of Mongolian Speech Synthesis Based on HMM

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] In order to make the object, technical solution and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with the examples. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0033] The application principle of the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0034] Such as figure 1 Shown, the method for the Mongolian speech synthesis based on HMM of the embodiment of the present invention comprises the following steps:

[0035] S101: Firstly, analyze the speech data in the sound bank, and extract the corresponding speech parameters;

[0036] S102: According to the extracted speech parameters, the observation vector of HMM can be divided into two parts: spectrum and fundamental frequency, and summarize the context a...

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

A method for front-end processing of Mongolian speech synthesis based on HMM. First, analyze the speech data in the sound library; summarize the context attribute set and the problem set used for decision tree clustering; The training result; the corresponding clustering state model is obtained to form a clustering state model sequence; the dynamic characteristics of the parameters are used to generate the target acoustic parameter sequence, and the synthesized speech is obtained through the STRAIGHT synthesizer; the front-end processing method is: the input Mongolian Processing special characters; transcribing Mongolian to Latin; tagging Mongolian parts of speech; using G2P model to transcribe Latin to phonemes; using rules to segment syllables; synthesizing data in the format required for prosody prediction; performing prosody prediction ; Process to get the text format required for synthesis. The invention can freely adjust the time length and pitch period parameters of the synthesized speech, maintain the natural and smoothness of the synthesized speech, and reduce the synthesis cost.

Description

technical field [0001] The invention belongs to the technical field of speech synthesis, and in particular relates to a front-end processing method for Mongolian speech synthesis based on HMM. Background technique [0002] Speech synthesis is a key technology to realize human-machine voice communication. It involves multiple disciplines such as acoustics, linguistics, digital signal processing, and computer science. It is a cutting-edge technology in the field of information processing. The main problem to be solved is how to integrate The conversion of text information into audible sound information is to make the machine speak like a human. [0003] Today's mainstream speech synthesis method is the speech synthesis technology based on HMM. Compared with the waveform splicing technology based on large corpus, it can automatically build a new system in a short time without manual intervention, and the whole The training process basically does not depend on the advantages of...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G10L13/08G10L15/14
Inventor 飞龙高光来赵建东张学良
Owner INNER MONGOLIA UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap