Statistical Voice Activity Detection Using Continues Hidden Markov Model
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This paper presents a novel method for identifying regions of speech based on the Continues Hidden Markov Model. Probably the most difficult problem faced when using HMMs is that of specifying the model parameters. The Baum-Welch algorithm is a method for estimating model parameters. Unfortunately, for large HMMs, multiplying many probabilities always yields very small numbers that will give underflow errors on any computer. We solve this problem by applying the Baum-Welch algorithm on the small segments of the observed noisy speech vector, based on the statistical approaches. The proposed algorithm estimates model parameters, recursively.Emissions probability distributions are assumed to be Variance Gamma (VG) and Gaussian distributions for speech and non-speech states, respectively.The simulation results show that the proposed Voice Activity Detection (VAD) is able to operate down to -5 dB and in nonstationary environments.
Abstract
This paper presents a novel method for identifying regions of speech based on the Continues Hidden Markov Model. Probably the most difficult problem faced when using HMMs is that of specifying the model parameters. The Baum-Welch algorithm is a method for estimating model parameters. Unfortunately, for large HMMs, multiplying many probabilities always yields very small numbers that will give underflow errors on any computer. We solve this problem by applying the Baum-Welch algorithm on the small segments of the observed noisy speech vector, based on the statistical approaches. The proposed algorithm estimates model parameters, recursively.Emissions probability distributions are assumed to be Variance Gamma (VG) and Gaussian distributions for speech and non-speech states, respectively.The simulation results show that the proposed Voice Activity Detection (VAD) is able to operate down to -5 dB and in nonstationary environments.
참고문헌 (15)
- A. Rezayee, "An adaptive KLT approach for speech enhancement", IEEE Trans. Speech Audio Processing, 9, pp.87~95 (2001)
- N. R. Garner, "Robust noise detection for speech detection and enhancement", Electron. Lett, 33(4), pp.270~271 (1997)
- F. Beritelli, "A robust voice activity detector for wireless communications using soft computing", IEEE J. Select. Areas Commun, 16, pp.1818~1829 (1998)
- J. Ramírez, "Statistical Voice Activity Detection Using a Multiple Observation Likelihood Ratio Test", IEEE Signal Process. Lett, 12(10), pp.~692 (2005)
- H. Othman, "A semi continuous state transition probability HMM based voice activity detection", IEEE Int. Conf. Acoustics, Speech, Signal Processing, pp.17~21 (2004)
- A. Davis, "Statistical Voice Activity Detection Using Low Variance Spectrum Estimation ans an Adaptive Threshold", IEEE Trans. Audio, Speech and Language Processing, 통권 : 14 (2006)
- R. Tahmasbi, "A Soft Voice Activity Detection Using GARCH Filter and Variance Gamma Distribution", IEEE Trans. Audio, Speech and Language Processing
- R. J. Mackay, "Hidden Markov Models,In Multiple Processes and Model Selection", 박사, pp.2003
- R. Rabiner, "A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition", Proceedings of the IEEE, 77(2) (1989)
- E. Seneta, "Fitting the Variance Gamma to Financial Data", Journal of Applied Probability, 41, pp.177~187 (2004)
- S. Gazor, "Speech probability distribution", IEEE Signal Processing Lett, 10, pp.204~207 (2003)
- R. Tahmasbi, "A soft voice activity detection using GARCH filter and variance gamma distribution", IEEE Trans. Audio Speech Lang, 15, pp.1129~1134 (2007)
- R. Tahmasbi, "Change Point detection in GARCH Models for Voice Activity Detection", IEEE Trans. Audio Speech Lang. Process, 16, pp.1036~1046 (2008)
- S. Rezaei, "A fast and soft voice activating detector GARCH Model and Morkov Model Chain", Amirkabir, 18(16), pp.65~72 (2007)
- S. Rezaei, "A robust statistical approach for detecting of speech and non speech interval", Amirkabir Journal of Science and Technology, 19(68), pp.65~72 (2008)