This paper presents a study on statistical integration of temporal filter banks for robust speech recognition using linear discriminant analysis (LDA). The temporal properties of stationary features were first captured and represented using a bank of well-defined temporal filters. Then these derived temporal features can be integrated and compressed using the LDA technique. Experimental results show that the recognition performance can be significantly improved both in clean and in noisy environments.
Cite as: Shen, J.-L., Hwang, W.-L. (1998) Statistical integration of temporal filter banks for robust speech recognition using linear discriminant analysis (LDA). Proc. 5th International Conference on Spoken Language Processing (ICSLP 1998), paper 0447, doi: 10.21437/ICSLP.1998-528
@inproceedings{shen98c_icslp, author={Jia-Lin Shen and Wen-Liang Hwang}, title={{Statistical integration of temporal filter banks for robust speech recognition using linear discriminant analysis (LDA)}}, year=1998, booktitle={Proc. 5th International Conference on Spoken Language Processing (ICSLP 1998)}, pages={paper 0447}, doi={10.21437/ICSLP.1998-528} }