Parametric Speech Analysis

Abstract

All the application areas of speech technology, including speech recognition, speech synthesis and speech coding, require some form of preliminary analysis of the speech signal. This chapter describes the basic techniques that are normally used to extract acoustic information directly from the speech signal. Most of these techniques are based on the source-filter model of speech production which was introduced in chapter 1. In this model, the excitation source is assumed to be linearly separable from the transmission characteristics of the vocal tract, which are represented by a quasi-time-invariant filter. The speech waveform itself is then assumed to be the output of this filter in response to the excitation source, which is either a quasi-periodic pulse generator (voiced sounds), a random-noise generator (unvoiced sounds) or, in some cases, a mixture of both (voiced fricatives). ‘Speech analysis’ is mainly the process of estimating the relatively slowly time-varying parameters which specify the filter, from a speech signal that is assumed to be the output of that filter. Other goals might include voiced/unvoiced classification and pitch-period estimation for voiced speech.