计算机工程与应用
計算機工程與應用
계산궤공정여응용
COMPUTER ENGINEERING AND APPLICATIONS
2013年
18期
217-222
,共6页
语音端点检测%语带频谱%最大特征值%鲁棒性
語音耑點檢測%語帶頻譜%最大特徵值%魯棒性
어음단점검측%어대빈보%최대특정치%로봉성
voice activity detection%speech bandwidth spectrum%maximum eigenvalue%robustness
针对在低信噪比条件下语音端点检测问题，提出了一种基于Toeplitz最大特征值的去噪语音端点检测方法。该方法用语带频谱自相关序列构造一个对称Toeplitz矩阵，利用该矩阵最大特征值的信息量对语音信号进行双门限端点检测。新算法经过实验，能够有效地区分语音和噪声，在不同的低噪声环境条件下具有良好的鲁棒性。与新近的信号递归度分析方法比较，准确率较高。该算法计算代价小，实时性好，简洁易实现。
針對在低信譟比條件下語音耑點檢測問題，提齣瞭一種基于Toeplitz最大特徵值的去譟語音耑點檢測方法。該方法用語帶頻譜自相關序列構造一箇對稱Toeplitz矩陣，利用該矩陣最大特徵值的信息量對語音信號進行雙門限耑點檢測。新算法經過實驗，能夠有效地區分語音和譟聲，在不同的低譟聲環境條件下具有良好的魯棒性。與新近的信號遞歸度分析方法比較，準確率較高。該算法計算代價小，實時性好，簡潔易實現。
침대재저신조비조건하어음단점검측문제，제출료일충기우Toeplitz최대특정치적거조어음단점검측방법。해방법용어대빈보자상관서렬구조일개대칭Toeplitz구진，이용해구진최대특정치적신식량대어음신호진행쌍문한단점검측。신산법경과실험，능구유효지구분어음화조성，재불동적저조성배경조건하구유량호적로봉성。여신근적신호체귀도분석방법비교，준학솔교고。해산법계산대개소，실시성호，간길역실현。
A Toeplitz de-noising method using the maximum eigenvalue is proposed for the voice activity detection at low SNR scenarios. This method uses the self-correlation sequence of speech bandwidth spectrum to construct a new symmetric Toeplitz matrix and to compute the largest eigenvalue, and the double decision thresholds in the largest eigenvalue are applied in the deci-sion framewok. Simulation results show that the presented algorithm is more effective in distinguishing speech from noise and has better robustness under various noisy environments. Compared with novel method of recurrence rate analysis, this algorithm shows lower wrong decision rate. The algorithm is of low computational complexity and is simple in real-time realization.
