Modulation and chaotic acoustic features for speech recognition
Dimitriadis, D, Petros A. Maragos, Vassilis Pitsikalis, Alexandros Potamianos · DSpace - NTUA (National Technical University of Athens) · 2002
Automatic speech recognition systems can benefit from including into their acoustic processing part new features that account for various nonlinear and time-varying phenomena during speech production. In this article, we develop robust methods for extracting from speech signals novel acoustic features, of the modulation- and chaotic-type, based on nonlinear and time-varying models of speech. These new speech features are integrated with the standard linear ones (mel-frequency cesptrum) to develop a generalized hybrid set of acoustic features. The efficacy of this hybrid set is demonstrated by showing significant improvements in HMM-based phoneme recognition over the TIMIT database.