The application of optimization in feature extraction of speech recognition

The speech recognition feature extraction methods used presently are not optimal when they are applied to a specific environment and recognition task. To deal with this problem, the new concepts of regional characteristic and trace characteristic are proposed, accompanied by the definition of the new parameter of regional resolution in the speech feature vector space. Under these concepts, optimization is introduced to turn the previous non-optimal, universal feature extraction method into a new optimal, task-dependent and environment-dependent one, which will improve the speech recognition results without changing the substructure of the original method or increasing the computational complexity