Paper Title:
A New Method Based on HMMs and K-Means Algorithms for Noise-Robust Voice Activity Detector
  Abstract

In this paper, we proposed left-right hidden Markov models (HMMs) combination with k-means threshold of Likelihood ratio test (LRT) to identify the start and end of the speech. This method builds two models of non-speech and speech but not two states, i.e. each model could conclude several states. In the experiments we present the Voice Activity Detection (VAD) results between two states hidden semi-Markov model (HSMM) and proposed algorithm. We also compare accuracy and robust between the k-means threshold and the adaptive threshold in high signal to noise rate in the background noise. It presents that k-means threshold is more effective than the adaptive threshold and the proposed method also make a better performance than two states HSMM based VAD, especially in the low signal-to-noise ratio (SNR) environment.

  Info
Periodical
Chapter
Chapter 3: Sensor, Test and Signal Processing
Edited by
Zhixiang Hou
Pages
461-464
DOI
10.4028/www.scientific.net/AMM.128-129.461
Citation
B. Luo, Z. Pei, L. Xu, D. L. Hu, "A New Method Based on HMMs and K-Means Algorithms for Noise-Robust Voice Activity Detector", Applied Mechanics and Materials, Vols. 128-129, pp. 461-464, 2012
Online since
October 2011
Export
Price
$32.00
Share

In order to see related information, you need to Login.

In order to see related information, you need to Login.

Authors: Zhong Qu, Qing Wei Ma
Abstract:In this paper, we extract slight movement object of real-time video images by using skin color detection and clustering methods. The...
603
Authors: Zheng Hong Xiao, Zhi Gang Chen, Xiao Heng Deng
Chapter 7: Computer Application in Design and Manufacturing (1)
Abstract:Based on the principle that the same class is adjacent, an anomaly intrusion detection method based on K-means and Support Vector Machine...
3745
Authors: Guan Hua Liu, Liang Zhong Yi, Zheng Pei, Da Li Hu
Chapter 3: Sensor, Test and Signal Processing
Abstract:The traditional algorithm of searching a new radio signal is not from the nature of the signal, but from the simple statistics and four...
714
Authors: Yan Jun Cui, Yan Dong Ma, Jie Li, Zheng Zhao
Chapter 11: Artificial Intelligence
Abstract:A new algorithm for training radial basis function neural network (RBFNN) is presented in this paper. This algorithm is based on the dynamic...
1593
Authors: Yang Pan, An Hua Chen, Ling Li Jiang
Chapter 3: Mechanical Transmission, Vibration and Noise
Abstract:According to the selection difficulties of initial clustering center of k-means clustering algorithm, this paper proposes a method that is to...
250