A PRELIMINARY STUDY ON THE INF0LUENCE OF SOUND DATA COMPRESSION UPON FORMANT FREQUENCY DISTRIBUTIONS IN VOWELS AND THEIR MEASUREMENT

The aim of this paper is to demonstrate the appearance of spectral differences of formant structures of a chosen vowel that occurred after compressing PCM sound data using a given compression method. Experiments show that sound data compression does not significantly affect stability of formant frequency distributions in terms of statistics, as long as it does not introduce a random, stochastic component into the original speech signal. Since used in experiments the formant frequency measuring technique, based on the Burg’s LPC algorithm, is vulnerable to a noise component in the speech signal, it becomes crucial to estimate the contribution of both, systematic and random errors, introduced by signal compression techniques, to the overall formant frequency measurement error. Bearing in mind that the sound data compression gains more popularity nowadays, and that formant frequency analysis is one of the basic tools for speaker identification in forensic practices, it is essential to know what the chances are for correct speaker identification while dealing with criminal evidence in the form of a compressed sound file