Voice Activity Detection with Decision Trees in Noisy Environments

Abstract:

An improved project based on double thresholds method in noisy environments is proposed for robust endpoints detection. Firstly, in this method, the distribution of zero crossing rate (ZCR) on the preprocessed signal is taken into account, and then the speech signal is divided into different parts to obtain appropriate thresholds with decision trees on the basis of the ZCR distribution. Finally, the double thresholds method, focusing on different importance of the energy and ZCR, is taken in the corresponding situation to determine the input segment is speech or non-speech. Simulation results indicate that the proposed method with decision trees obtains more accurate data than the traditional double thresholds method.

• Periodical: Applied Mechanics and Materials (Volumes 128-129)
• Pages: 749-752
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.128-129.749
• Online since: October 2011
• Authors: Da Li Hu, Liang Zhong Yi, Zheng Pei, Bing Luo
• Keywords: Density, Distribution, Threshold, Voice Activity Detection (VAD), Zero Crossing Rate (ZCR)

References

[1] Rabiner L. R and Sambur M. R, in: An algorithm for determining the endpoints of isolated utterances [J]. Bell System Technical Journal, 1975, 54: 297-315.

DOI: 10.1002/j.1538-7305.1975.tb02840.x

[2] Misra Hemant, Ikbal Shajith etc, in: Multi-resolution spectral entropy feature for robust ASR. ICASSPC05, 2005, 1: 253-256.

[3] Wu Bingfei and Wang Kunching, in: Robust endpoint detection algorithm based on the adaptive band-partitioning spectral entropy in adverse environments. IEEE Trans on Speech Processing, 2005, 13: 762-775.

DOI: 10.1109/tsa.2005.851909

[4] Long Hainan and Zhang Cuigai, in: An improved method for robust speech endpoint detection, Proceedings of the Eighth International Conference on Machine Learning and Cybernetics, 2009: 2067-(2071).

DOI: 10.1109/icmlc.2009.5212154

[5] Martin Arnaud and Mauuary Laurent, in: Robust speech/non-speech detection based on LDA-derived parameter and voicing parameter for speech recognition in noisy environments, Speech Communication, 2006, 48: 191-206.

DOI: 10.1016/j.specom.2005.07.005

[6] Hsieh Cheng-Hsiung, Feng Ting-Yu and Huang Po-Chin, in: Energy-based VAD with grey magnitude spectral subtraction. Speech Communication, 2009, 51: 810-819.

DOI: 10.1016/j.specom.2008.08.005

[7] Liu Qingsheng, Xu Xiaopeng and Huang Wenhao, in: Research on a speech endpoint method, Computer Engineering, 2003, 29: 120-121.