Hybrid combination of knowledge- and cepstral-based features for phoneme recognition
dc.contributor.author | v.d. Merwe Rudolph | |
dc.contributor.author | du Preez Johan A. | |
dc.date.accessioned | 2011-05-15T15:57:29Z | |
dc.date.available | 2011-05-15T15:57:29Z | |
dc.date.issued | 1998 | |
dc.description.abstract | In this paper a new, general, mathematically sound technique is developed to integrate knowledge-based information with standard cepstral features into the formal HMM framework for phoneme recognition. By using these hybrid features, the maximum amount of information contained in the speech signal can be utilized. It is shown that a trivial extension of the statistical models used to model the cepstral features, cannot be used to model the hybrid feature vectors, as this results in a decrease in phoneme recognition accuracy. By using the proposed hybrid technique though, a statistically significant increase in phoneme recognition accuracy is achieved. | |
dc.description.version | Article | |
dc.identifier.citation | Proceedings of the South African Symposium on Communications and Signal Processing, COMSIG | |
dc.identifier.uri | http://hdl.handle.net/10019.1/10432 | |
dc.subject | Feature extraction | |
dc.subject | Knowledge based systems | |
dc.subject | Markov processes | |
dc.subject | Mathematical models | |
dc.subject | Probability density function | |
dc.subject | Speech analysis | |
dc.subject | Statistical methods | |
dc.subject | Vectors | |
dc.subject | Cepstral features | |
dc.subject | Hidden Markov model (HMM) | |
dc.subject | Phoneme recognition | |
dc.subject | Speech recognition | |
dc.title | Hybrid combination of knowledge- and cepstral-based features for phoneme recognition | |
dc.type | Article |