This project concerns the communicatively important aspects in spontaneous speech as compared to read speech. Spectro-temporal and intonational aspects of focus words (i.e. words bearing the highest load of semantic information) are studied in spontaneous speech and in the same texts, re-read after orthographic transcription. In order to investigate the acoustic characteristics of both speech styles, we introduced a peak-and-level model with two possible hypotheses, regarding the spectral and durational contrasts in focus and non-focus words in spontaneous speech as compared to read speech. In first instance we used speech material of one professional male speaker, the one whose voice and pronunciation served as a model for the diphone-based component of the Dutch national speech synthesis program. For a number of global parameters it can be concluded that there is a clear difference in the 'level values1 of the two speech styles, but that the 'peak values' display similar contrast to these level values.
Cite as: Koopmans-van Beinum, F.J. (1991) Spectro-temporal reduction and expansion in spontaneous speech and read text: focus words versus non-focus words. Proc. ESCA Workshop on Phonetics and Phonology of Speaking Styles, paper 036
@inproceedings{koopmansvanbeinum91_ppst,
author={Florien J. {Koopmans-van Beinum}},
title={{Spectro-temporal reduction and expansion in spontaneous speech and read text: focus words versus non-focus words}},
year=1991,
booktitle={Proc. ESCA Workshop on Phonetics and Phonology of Speaking Styles},
pages={paper 036}
}