Sixth International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications (MAVEBA 2009)
The presentation concerns the evaluation of a temporal method for tracking cycle lengths in voiced speech. The speech cycles are detected via the saliences of the speech signal samples. The method does not request that the signal is locally periodic and the average period length known a priori. The cycle length extraction is applied to the analysis of dysphonic speakers affected by amyotrophic lateral sclerosis (ALS). Results suggest that salience analysis is able to track reliably glottal cycles in the speech signal. SLA speakers are characterized by higher vocal tremor depths and tremor frequencies than normophonic speakers.
Index Terms. vocal frequency, vocal tremor, speech salience analysis
Full Paper (reprinted with permission from Firenze University Press)
Bibliographic reference. Mertens, C. / Grenez, Francis / Crevier-Buchman, L. / Schoentgen, Jean (2009): "Salience analysis for glottal cycle detection in disordered speech", In MAVEBA-2009, 99-102.