English  |  正體中文  |  简体中文  |  Items with full text/Total items : 25391/25986 (98%)
Visitors : 6128364      Online Users : 368
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version

    Please use this identifier to cite or link to this item: http://ir.lib.ksu.edu.tw/handle/987654321/7015

    Title: 白色雜訊特性未知下,訊號子空間語音加強方法之研究
    Signal Subspace-Based Speech Enhancement without A Priori Knowledge of White Noise
    Authors: 郭崇仁
    Keywords: 訊號子空間
    signal subspace processing
    noise subspace
    sliding window
    orthogonal projection
    white noise
    recursive covariance matrix
    sliding window covariance matrix
    Date: 2003-07-31
    Issue Date: 2009-12-31 17:29:17 (UTC+8)
    Abstract: 本計畫提出一個新的非參數型訊號子空間處理語音訊號加強法。此訊號子空間處理技術利用滑動視窗及正交投射的原理來預測無雜訊語音訊號,此種方法並不需要運用到”事先白色雜訊特性 (a priori knowledge of noise) ”。兩個短期共變異數矩陣-- 迴歸共變異數矩陣及滑動視窗共變異數矩陣,被設計來預測語音訊號的特徵値及特徵向量。原含雜訊語音訊號空間可被分離成兩個子空間 -- 原含雜訊訊號子空間及雜訊子空間。再運用重要正交投射,將原含雜訊的語音訊號投射到重要子空間上,借以去掉大部分的雜訊而獲得較清析的語音訊號因而提高SNR。
    A novel nonparametric speech enhancement approach based on signal subspace processing has been proposed in this program. This signal subspace processing technique utilizes a sliding window and orthogonal projection principles without a priori knowledge of white noise to estimate the clean speech signal. Two short-term covariance matrices, the recursive covariance matrix and the sliding window covariance matrix have been designed for estimating the eigenvalues and eigenvectors of speech signal. The entire space can be partitioned the noisy signal space into the signal-plus-noise subspace and noise subspace. Using the dominant orthogonal eigen-projection, more speech signal but less noise will be projected onto this dominant subspace, thereby increasing the SNR.
    Appears in Collections:[電機工程系所] 研究計畫

    Files in This Item:

    File Description SizeFormat
    912218E168002.pdf203KbAdobe PDF297View/Open

    All items in KSUIR are protected by copyright, with all rights reserved.

    ©Kun Shan University Library and Information Center
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback