Long-range speech acquirement and enhancement with dual-point laser Doppler vibrometers

This paper presents a long-range speech acquirement and speech enhancement system by utilizing two laser Doppler vibrometers (LDV) to measure two separate points on only one vibration object or two different vibration objects. This proposed LDV system can provide dual-channel synchronous signals, and thus the coherent-to-diffused ratio-based and the multi-channel linear prediction-based algorithms can be introduced to reduce the reverberation and the noise of the acquired speech signals. The two algorithms are combined together to further improve the speech enhancement performance. Experimental results show that the proposed algorithm can significantly improve the PESQ scores.

[1]  Emanuel A. P. Habets,et al.  Signal-to-reverberant ratio estimation based on the complex spatial coherence between omnidirectional microphones , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[2]  Walter Kellermann,et al.  Coherent-to-Diffuse Power Ratio Estimation for Dereverberation , 2015, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[3]  Thomas S. Huang,et al.  LDV Remote Voice Acquisition and Enhancement , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[4]  Tao Wang,et al.  Vibration Characteristics of Various Surfaces Using an LDV for Long-Range Voice Acquisition , 2011, IEEE Sensors Journal.

[5]  Huaguo Zang,et al.  Laser Doppler vibrometer for real-time speech-signal acquirement , 2009 .

[6]  Jun Du,et al.  Auxiliary Features from Laser-Doppler Vibrometer Sensor for Deep Neural Network Based Robust Speech Recognition , 2018, J. Signal Process. Syst..

[7]  Israel Cohen,et al.  Speech measurements using a laser Doppler vibrometer sensor: Application to speech enhancement , 2011, 2011 Joint Workshop on Hands-free Speech Communication and Microphone Arrays.

[8]  Biing-Hwang Juang,et al.  Speech Dereverberation Based on Variance-Normalized Delayed Linear Prediction , 2010, IEEE Transactions on Audio, Speech, and Language Processing.

[9]  Shoji Makino,et al.  Exponentially weighted stepsize NLMS adaptive filter based on the statistics of a room impulse response , 1993, IEEE Trans. Speech Audio Process..

[10]  Xiaodong Li,et al.  Bandwidth extension for speech acquired by laser Doppler vibrometer with an auxiliary microphone , 2015, 2015 10th International Conference on Information, Communications and Signal Processing (ICICS).

[11]  Yozo Fujino,et al.  Experimental study of laser Doppler vibrometer and ambient vibration for vibration-based damage detection , 2006 .

[12]  Hani Nassif,et al.  Comparison of laser Doppler vibrometer with contact sensors for monitoring bridge deflection and vibration , 2005 .

[13]  He-yong Zhang,et al.  Acquirement and enhancement of remote speech signals , 2017 .

[14]  Jun Du,et al.  Deep neural network for robust speech recognition with auxiliary features from laser-Doppler vibrometer sensor , 2016, 2016 10th International Symposium on Chinese Spoken Language Processing (ISCSLP).

[15]  Bryan Kok Ann Ngoi,et al.  Two-axis-scanning laser Doppler vibrometer for precision engineering , 2002 .

[16]  K Nakamura,et al.  Laser Doppler vibrometer (LDV)--a new clinical tool for the otologist. , 1996, The American journal of otology.