Dual-Microphone Speech Extraction from Signals with Audio Background

This paper presents a model of a system which is able to support Police investigations. The listening in system acquires audio signals from two microphones located around one meter to each other. System is designed to record suspect conversations under disadvantageous conditions like the loud music or the car machine noise. The presented method uses the differences between records obtained from two microphones. Different delays and different spectra separates the conversation signal from the background. The crosscorelation analysis and the band-pass filters were applied to realise the suggested algorithm.

[1]  Karl-Dirk Kammeyer,et al.  Theoretical noise reduction limits of the generalized sidelobe canceller (GSC) for speech enhancement , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[2]  Parham Aarabi,et al.  Robust digit recognition using phase-dependent time-frequency masking , 2003, 2003 IEEE International Conference on Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03)..

[3]  Parham Aarabi,et al.  Low-Power Dual-Microphone Speech Enhancement Using Field Programmable Gate Arrays , 2007, IEEE Transactions on Signal Processing.

[4]  Yannick Mahieux,et al.  Analysis of noise reduction and dereverberation techniques based on microphone arrays with postfiltering , 1998, IEEE Trans. Speech Audio Process..

[5]  G. DeMuth,et al.  Frequency domain beamforming techniques , 1977 .