The Influence of a Filter Shape in Telephone-Based Recognition Module Using PLP Parameterization
This paper concerns an influence of a filter shape and a benefit of the Hertz-Bark transformation to the word error rate (WER) obtained in a telephone-based speech recognition application working with the Perceptually-based Linear Predictive (PLP) parameterization. Five various shapes of filters (rectangular, narrow and wide trapezium, triangular and the classical PLP filter shape ) were compared and an effect of a nonlinear frequency transformation between Hertz and generalized Bark axis was explored. Experiments with 100 speakers and with the vocabulary size of 475 words were performed. During all experiments only the zero-gram language model was used to see better an influence of particular variables to changes of the WER.
KeywordsSpeech Recognition Recognition Accuracy Critical Band Word Error Rate Recognition Experiment
Unable to display preview. Download preview PDF.
- 3.Müller, L., Psutka, J., Šmídl, L.: Design of Speech Recognition Engine.-In: Text, Speech and Dialogue. The 3rd International Workshop on TSD’2000. Berlin, Heidelberg, Springer-Verlag 2000. pp.259–264.Google Scholar
- 4.Psutka, J., Müller, L., Psutka, J.V.: Comparison of MFCC and PLP Parameterisations in the Speaker Independent Continuous Speech Recognition Task. (prepared for EUROSPEECH2001).Google Scholar