This study focuses on short-term acoustic correlates of voice quality. It assesses the within-speaker stability (across different speaking styles) and between-speaker variability of measurements which compare the amplitudes of various spectral events H1*-H2*, H2*-H4*, H1*-A1*, H1*-A2* and H1*-A3*. Although speakers do differ with regard to the compactness of the parameters in read and spontaneous speaking styles, the parameters H1*-H2*, H1*-A1* and H1*-A2* appear both considerably stable for one speaker in different speaking styles and efficient in between-speaker comparisons. Though not directly applicable in forensic settings, these glottal parameters outperformed vowel formants in classification using LDA.
Cite as: Vaňková, J., Skarnitzl, R. (2014) Within- and Between-Speaker Variability of Parameters Expressing Short-Term Voice Quality. Proc. Speech Prosody 2014, 1081-1085, doi: 10.21437/SpeechProsody.2014-206
@inproceedings{vankova14_speechprosody,
author={Jitka Vaňková and Radek Skarnitzl},
title={{Within- and Between-Speaker Variability of Parameters Expressing Short-Term Voice Quality}},
year=2014,
booktitle={Proc. Speech Prosody 2014},
pages={1081--1085},
doi={10.21437/SpeechProsody.2014-206}
}