Noise adaptation system of speech model, noise adaptation method, and noise adaptation program for speech recognition

An adaptive system and noise technology, applied in speech recognition, speech analysis, instruments, etc., can solve problems such as characteristic deviation of noise

Inactive Publication Date: 2004-11-03
NTT DOCOMO INC +1
View PDF0 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The problem with this approach is that there is a bias between the features of the noise during clustering and the features in the noisy speech model during model learning

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
  • Noise adaptation system of speech model, noise adaptation method, and noise adaptation program for speech recognition
  • Noise adaptation system of speech model, noise adaptation method, and noise adaptation program for speech recognition
  • Noise adaptation system of speech model, noise adaptation method, and noise adaptation program for speech recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] Embodiments of the present invention will be described below with reference to the accompanying drawings. The same elements are marked with the same reference numerals in the drawings referred to in the following description.

[0023] In the present invention, the noisy speech model space is constructed into a tree structure according to the signal-to-noise ratio (SNR) and the sound quality. Representing noise characteristics through a tree structure provides a model in which the global characteristics of the noise are represented at higher levels and local characteristics are represented at lower levels. The optimal segmentation space can be selected by searching the tree structure from the root down in a top-down manner, thereby selecting the optimal model.

[0024] Since the noisy speech is always used in the clustering process and the model learning process, the noisy speech model providing the highest likelihood can be learned, and improvement in recognition accur...

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

An object of the present invention is to enable optimal clustering for many types of noise data and to improve the accuracy of estimation of a speech model sequence of input speech. Noise is added to speech in accordance with noise-to-signal ration conditions to generated noise-added speech (step S1), the mean value of speech cepstral is subtracted from the generated, noise-added speech (step S2), a Gaussian distribution model of each piece of noise-added speech is created (step S3), the likelihoods of the pieces of noise-added speech are calculated to generate a likelihood matrix (step S4) to obtain a clustering result. An optimum model is selected (step S7) and linear transformation is performed to provide a maximized likelihood (step S8). Because noise-added speech is consistently used both in clustering and model learning, clustering for many types of noise data and an accurate estimation of a speech model sequence can be achieved.

Description

technical field [0001] The invention relates to a speech model noise adaptive system, a noise adaptive method and a speech recognition noise adaptive program. Specifically, the present invention relates to adapting a clean speech model generated by modeling speech features using a Hidden Markov Model (HMM) with the noisy speech to be recognized, thereby improving speech recognition in noisy environments. Speech model noise adaptive system, noise adaptive method and speech recognition noise adaptive program of the recognition rate. Background technique [0002] A tree-structured piecewise linear transformation method is described in Non-Patent Document 1 below. According to the method disclosed in this document, the noise is clustered, and according to the result of the clustering, a tree structure noisy speech model space is produced, and the speech feature parameters of the input noisy speech to be recognized are extracted, from the tree structure The optimal model is sel...

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 Applications(China)
IPC IPC(8): G10L15/06G10L15/02G10L15/065G10L15/14G10L15/20G10L17/00G10L25/00
CPCG10L15/20
Inventor 张志鹏大辻清太杉村利明古井贞熙
Owner NTT DOCOMO INC
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