Large-length voice full-automatic segmentation method

A fully automatic, voice technology, applied in the direction of voice analysis, voice recognition, instruments, etc., can solve the problems of reducing the amount of labeling, the accuracy cannot be guaranteed, etc., and achieve the effect of reducing loss and calculation cost

Inactive Publication Date: 2013-10-09
张巍
View PDF5 Cites 49 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the amount of labeling can be greatly

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
  • Large-length voice full-automatic segmentation method
  • Large-length voice full-automatic segmentation method
  • Large-length voice full-automatic segmentation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0053] The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0054] Proposal of Sentence Segmentation Algorithm Based on HMM of Spectral Parameters and Prosodic Parameters

[0055] Introduction to Fully Automatic Sentence Segmentation Algorithms

[0056] First of all, we regard the unlabeled automatic sentence segmentation algorithm and the sentence segmentation algorithm based on semi-supervised learning with minimal annotation as two subsystems of the entire sentence automatic segmentation system framework, and regard the former definition based on Force-alignment The algorithm's unlabeled sentence segmentation algorithm is used as a labeling system to provide a small amount of fine-label data for training, and it is renamed Sub-Labeling System (ZLSS). Then use the minimally labeled sentence automatic segmentation algorithm defined by the latter as our classification system, and rena...

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 relates to a large-length voice full-automatic segmentation method which is a zero-labeling sentence automatic segmentation algorithm having higher accuracy. The algorithm enables a Force-alignment non-supervision algorithm and a semi-supervised learning method based on an HMM to be blended, automatic expansion is carried out on a few precise labeling sets provided for the zero-labeling sentence segmentation algorithm by a semi-supervised learning minimization labeling sentence segmentation algorithm through establishment of an iteration mechanism based on a timer shaft, the purpose of the maximization of the precise labeling sets is realized, and then according to obtained correct periods, voice of an original length is cut into smaller paragraphs or sets of sentences. According to the method, a Force-aligned method under the HMM and a Co_training method in semi-supervised learning are blended together, so the facts that in a large-length voice sentence segmentation process, manual intervention is not needed, and segmentation accuracy is high are guaranteed. The large-length voice full-automatic segmentation method can be applied to rapid and automatic construction of a voice corpus.

Description

technical field [0001] The invention belongs to the technical fields of speech synthesis, speech recognition, speech retrieval and labeling, and relates to a long-length speech automatic segmentation method. Background technique [0002] At present, there are two mainstream speech synthesis methods in the world. One is the trainable (Trainable TTS) speech synthesis method based on HMM, such as CLUSTERGEN of Carnegie Mellon University (CMU) in the United States, and HTS developed by Nagoya Institute of Technology in Japan. Speech synthesis engines, they all use a method based on parametric statistical (Parametric Statistical) synthesis; the other is a speech synthesis method based on a large speech corpus (corpus-based TTS (text to speech)), for example, the Chinese Academy of Sciences Acoustics The proposed KX-PSOLA (1993), and the speech synthesis technology adopted by Xunfei on the telecommunications platform, these are based on the technology of unit selection and wavefor...

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): G10L15/04
Inventor 张巍王永远张志楠
Owner 张巍
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