Consistent Modeling of the Static and Time-Derivative Cepstrums for Speech Recognition Using HSPTM
Most speech models represent the static and derivative cepstral features with separate models that can be inconsistent with each other. In our previous work, we proposed the hidden spectral peak trajectory model (HSPTM) in which the static cepstral trajectories are derived from a set of hidden trajectories of the spectral peaks (captured as spectral poles) in the time-frequency domain. In this work, the HSPTM is generalized such that both the static and derivative features are derived from a single set of hidden pole trajectories using the well-known relationship between the spectral poles and cepstral coefficients. As the pole trajectories represent the resonance frequencies across time, they can be interpreted as formant tracks in voiced speech which have been shown to contain important cues for phonemic identification. To preserve the common recognition framework, the likelihood functions are still defined in the cepstral domain with the acoustic models defined by the static and derivative cepstral trajectories. However, these trajectories are no longer separately estimated but jointly derived, and thus are ensured to be consistent with each other. Vowel classification experiments were performed on the TIMIT corpus, using low complexity models (2-mixture). They showed 3% (absolute) classification error reduction compared to the standard HMM of the same complexity.
KeywordsSpeech Recognition Spectral Peak Signal Proc Audio Processing Speak Language Processing
Unable to display preview. Download preview PDF.
- 1.Gish, H., Ng, K.: A segmental speech model with applications to word spotting. In: Proc. of the IEEE Inter. Conf. on Acoust., Speech and Signal Proc., pp. 447–450 (1993)Google Scholar
- 3.Siu, M., Iyer, R., Gish, H., Quillen, C.: Parametric trajectory mixtures for lvcsr. In: Proc. of the Inter. Conf. on Spoken Language Processing (1998)Google Scholar
- 5.Deng, L., Aksmanovic, M., Sun, X., Wu, C.: Speech recognition using hidden markov models with polynomial regression functions as nonstationary states. IEEE Trans. on Speech and Audio Processing, 507–520 (1994)Google Scholar
- 6.Li, C., Siu, M.: An efficient incremental likelihood evaluation for polynomial trajectory model using with application to model training and recognition. In: Proc. of the IEEE Inter. Conf. on Acoust., Speech and Signal Proc., pp. 756–759 (2003)Google Scholar
- 7.Zen, H., Tokuda, K., Kitamura, T.: Trajectory modeling based on HMMs with the explicit relationship between static and dynamic features. In: Proc. of the IEEE Inter. Conf. on Acoust., Speech and Signal Proc., pp. 837–840 (2004)Google Scholar
- 9.Lai, Y., Siu, M.: Hidden spectral peak trajectory model for phone classification. In: Proc. of the IEEE Inter. Conf. on Acoust., Speech and Signal Proc., pp. 17–21 (2004)Google Scholar
- 10.Au Yeung, S., Li, C., Siu, M.: Sub-phonetic polynomial segment model for large vocabulary continuous speech recognition. In: Proc. of the IEEE Inter. Conf. on Acoust., Speech and Signal Proc., pp. 193–196 (2005)Google Scholar
- 11.Huang, X.D., Acero, A., Hon, H.W.: Spoken language processing: A guide to theory, algorithm and system development. Prentice Hall Inc., Upper Saddle River (2000)Google Scholar
- 12.Fukada, T., Sagisaka, Y., Paliwal, K.K.: Model parameter estimation for mixture density polynomial segment models. In: Proc. of the IEEE Inter. Conf. on Acoust., Speech and Signal Proc., pp. 1403–1406 (1997)Google Scholar
- 13.Gish, H., Ng, K.: Parametric trajectory models for speech recognition. In: Proc. of the Inter. Conf. on Spoken Language Processing, pp. 466–469 (1996)Google Scholar
- 14.Young, S., Evermann, G., Kershaw, D., Moore, G., Odell, J., Ollason, D., Valtchev, V., Woodland, P.: The HTK book (2001)Google Scholar