Speech Feature Extraction and Recognition Using Genetic Algorithm
Hitesh Gupta, Deepinder Singh Wadhwa · 2014
Abstract People are so comfortable with speech that they would also like to interact with computers via speech, rather than having to resort to primitive interfaces such as keyboards and pointing devices. However, undesired noises in environment like sound from heavy machines, vehicles are also present that causes undesired effects in speech transmission and acquiring systems reducing the performance of the machine receiving the speech signal. Advanced speech enhancement algorithms can be classified in main three categories, namely; filtering/estimation based noise reduction, beam forming and active noise cancellation (ANC) techniques. Recent work shows improved performance results when GA is applied on speech signals recorded under noisy conditions. The work here applies evolutionary computation in form of genetic algorithm to select the features that are responsible for discriminating the different words. In doing so, the amount of feature elements to be used also gets reduced and hence system can be made to recognize the word-speech with realtime performance.