Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Chinese rhythm boundary prediction method based on graph-to-sequence

A prosodic boundary prediction, prosodic boundary technology, applied in speech synthesis, speech analysis, instruments, etc., can solve the problem of not combining temporal information and spatial information.

Pending Publication Date: 2020-11-17
TIANJIN UNIV
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, the study of prosodic boundary prediction in Chinese speech synthesis does not combine the temporal information and spatial information in the text

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
  • Chinese rhythm boundary prediction method based on graph-to-sequence
  • Chinese rhythm boundary prediction method based on graph-to-sequence
  • Chinese rhythm boundary prediction method based on graph-to-sequence

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The present invention will be further described and proved in detail below in combination with the experimental process and experimental results.

[0025] Based on the basic framework of the current general sequence prediction Bilstm-CRF, the present invention proposes the representation of text space information from the perspective of text analysis, and on this basis, combines the time series information and space information of the text for the first time To improve the results of prosodic boundary prediction in speech synthesis. The main points of the specific technical plan are divided into the following three parts:

[0026] (1) Basic structure of sequence prediction

[0027] At present, in the prosody prediction module of speech synthesis, the most common method in the industry is Bilstm-CRF. Among them, the text embedding vector input by Bilstm outputs the features extracted in the time domain. The output of Bilstm is also the input of CRF. According to the c...

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 Chinese rhythm boundary prediction method based on graph-to-sequence, which specifically comprises the following four parts: (1) word embedding representation characteristics: converting the characteristics into digital representation, so that the technology of mapping words into real number field vectors is called word embedding; (2) a text time sequence feature extraction model, wherein annotation of rhythm boundaries is sequence annotation in the time dimension; (3) text space information: inputting a text sequence, processing the text sequence into a graphic structure, and processing a dependency relationship between rhythm boundaries by adding space information; and (4) combining spatial and temporal features. Time sequence information and space information of the text are combined together to serve as a new feature, and the accuracy of the final rhythm boundary is improved.

Description

technical field [0001] The invention belongs to the field of speech synthesis, and mainly relates to a technique for improving the accuracy rate of text prosodic boundary prediction in speech synthesis, so as to provide better conditions for subsequent speech synthesis to synthesize natural speech. Background technique [0002] Text-to-speech (TTS) is a text-to-speech technology designed to enable machines to generate artificial speech. A classic statistical parametric speech synthesis (SPSS) system usually consists of three modules, including: a front-end module (converting text into language features), an acoustic model (mapping language features into acoustic features) and a vocoder (generating voice waveform). Over the past few decades, statistical parametric speech synthesis has been complicated to use due to the artificially designed nature and cumbersome inter-module communication. However, with the development of deep learning in recent years, end-to-end speech syn...

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): G10L13/02G10L13/08G10L25/30
CPCG10L13/02G10L13/08G10L25/30
Inventor 陈帅婷王龙标本多清志
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com