A multifeature speech/music discrimination system

This paper presents a new technique for the automatic classification of audio signals into either speech or music signals. The classification is based on the most efficient five features extracted from the input signal. The correct classification ratio is always better than that using previous algorithms.

[1]  Richard F. Lyon,et al.  Auditory model inversion for sound separation , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[2]  B. Kedem,et al.  Spectral analysis and discrimination by zero-crossings , 1986, Proceedings of the IEEE.

[3]  John Saunders,et al.  Real-time discrimination of broadcast speech/music , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[4]  T. Moon.  The expectation-maximization algorithm , 1996, IEEE Signal Process. Mag..