The Hartley Phase Spectrum as an Assistive Feature for Classification

The phase of a signal conveys critical information for feature extraction. In this work is shown that for certain speech and audio classes where their magnitude content underperforms in terms of recognition rate, the combination of magnitude with phase related features increases the classification rate compared to the case where only the magnitude content of the signal is used. However, signal phase extraction is not a straightforward process, mainly due to the discontinuities appearing in the phase spectrum. Hence, in the proposed method, the phase content of the signal is extracted via the Hartley Phase Spectrum where the sources of phase discontinuities are detected and overcome, resulting in a phase spectrum in which the number of discontinuities is significantly reduced.

[1]  John G. Proakis,et al.  Digital signal processing (2nd ed.): principles, algorithms, and applications , 1992 .

[2]  S. Treitel,et al.  Factoring very-high-degree polynomials , 2003, IEEE Signal Process. Mag..

[3]  John G. Proakis,et al.  Digital Signal Processing: Principles, Algorithms, and Applications , 1992 .

[4]  P. Gough.  A particular example of phase unwrapping using noisy experimental data , 1983 .

[5]  Stan Davis,et al.  Comparison of Parametric Representations for Monosyllabic Word Recognition in Continuously Spoken Se , 1980 .

[6]  Kuldip K. Paliwal,et al.  Evaluation of the modified group delay feature for isolatedword recognition , 2005, Proceedings of the Eighth International Symposium on Signal Processing and Its Applications, 2005..

[7]  Thierry Dutoit,et al.  Chirp group delay analysis of speech signals , 2007, Speech Commun..

[8]  José Tribolet,et al.  A new phase unwrapping algorithm , 1977 .

[9]  P. Mahalanobis.  On the generalized distance in statistics , 1936 .

[10]  Ioannis Paraskevas,et al.  The Hartley Phase Cepstrum as a Tool for Improved Phase Estimation , 2009, 2009 16th International Conference on Systems, Signals and Image Processing.

[11]  Douglas Eck,et al.  Finding Meter in Music Using An Autocorrelation Phase Matrix and Shannon Entropy , 2005, ISMIR.

[12]  Kuldip K. Paliwal,et al.  On the usefulness of STFT phase spectrum in human listening tests , 2005, Speech Commun..

[13]  Andrew R. Webb,et al.  Statistical Pattern Recognition , 1999 .

[14]  Ioannis Paraskevas,et al.  The Hartley Phase Cepstrum as a Tool for Signal Analysis , 2007, NOLISP.

[15]  Kuldip K. Paliwal,et al.  Further intelligibility results from human listening tests using the short-time phase spectrum , 2006, Speech Commun..

[16]  Ronald N. Bracewell,et al.  The Fourier Transform and Its Applications , 1966 .

[17]  R. Kuc,et al.  A direct relation between a signal time series and its unwrapped phase , 1982 .

[18]  Hermann Ney,et al.  Using phase spectrum information for improved speech recognition performance , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[19]  Jonathan G. Fiscus,et al.  Darpa Timit Acoustic-Phonetic Continuous Speech Corpus CD-ROM {TIMIT} | NIST , 1993 .

[20]  Hamid Al-Nashi.  Phase unwrapping of digital signals , 1989, IEEE Trans. Acoust. Speech Signal Process..

[21]  Jordi Sole I Casals,et al.  Advances in Nonlinear Speech Processing, International Conference on Nonlinear Speech Processing, NOLISP 2009, Vic, Spain, June 25-27. Revised Selected Papers , 2010, NOLISP.

[22]  S. Furui,et al.  Cepstral analysis technique for automatic speaker verification , 1981 .