Yung-Sheng HuangJeih-Weih HungData-driven temporal filters based on maximum mutual information for robust features in speech recognition.105-1082004ISCSLPhttps://doi.org/10.1109/CHINSL.2004.1409597https://www.isca-archive.org/iscslp_2004/huang04b_iscslp.htmlconf/iscslp/2004db/conf/iscslp/iscslp2004.html#HuangH04