Application of Slope Filtering to Robust Spectral Envelope Extraction for Speech/Speaker Recognition

This paper describes a method for speech feature extraction using morphological signal processing based on the so-called "slope transformation". The proposed approach has been used to extract the signal upper spectral envelope. Results of experiments of the automatic speech recognition (ASR) and automatic speaker identification (ASI), which were undertaken to check the performance of the presented method, have shown some evident improvements of the effectiveness of recognition of isolated words, especially for women voices. The benefits of using slope transformation was also observed in speaker identification experiment.

[1]  John H. L. Hansen,et al.  A new perceptually motivated MVDR-based acoustic front-end (PMVDR) for robust automatic speech recognition , 2008, Speech Commun..

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

[3]  R. Bakis.  Continuous speech recognition via centisecond acoustic states , 1976 .

[4]  A. Dabrowski,et al.  Detection Of Endpoints Of Isolated Words Using Slope Transformation , 2006, Proceedings of the International Conference Mixed Design of Integrated Circuits and System, 2006. MIXDES 2006..

[5]  Liang Gu,et al.  Perceptual harmonic cepstral coefficients for speech recognition in noisy environment , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[6]  R. Altes.  Sonar for generalized target description and its similarity to animal echolocation systems. , 1976, The Journal of the Acoustical Society of America.

[7]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

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

[9]  Steve Young,et al.  The HTK book , 1995 .

[10]  Stefan Grocholewski,et al.  Statystyczne podstawy systemu ARM dla języka polskiego , 2001 .

[11]  Petros Maragos.  Slope transforms: theory and application to nonlinear signal processing , 1995, IEEE Trans. Signal Process..