Pre-processing of the Speech Data

Abstract

In this chapter the necessary pre-processing of the analog speech waveform is briefly described. The description mainly follows [Pau98]and [Fur01]. First,by using a microphone,the speech signal or speech waveform is converted into an electrical signal. Figure 3.1 shows a block diagram of a typical speech analysis procedure, generating Mel-frequency cepstral coefficients (MFCCs).The different steps are explained in the following sections.