Characterization of Hesitations Using Acoustic Models
Characterization of Hesitations Using Acoustic Models, Proc International Congress of Phonetic Sciences - ICPhS XVII, Hong Kong, China, Vol. -, pp. 2054 - 2057, August, 2011.
Digital Object Identifier:
Spontaneous speech is full of hesitations, such as fillers, word cut-offs, repetitions and segmental extensions. Automatic identification of such hesi-tations has several applications; however, it is a challenging research problem. In this paper acoustic-phonetic properties of hesitation phe-nomena are explored in order to identify and annotate some of these events in a spontaneous speech corpus of Portuguese broadcast television news. Based on pitch, energy, spectral and durational characteristics of the filled pauses and segmental extensions during their production, we intend to characterize the acoustic-phonetic regularity of the phenomena. A speech recognition sys-tem was used to help locating the filled pauses and extensions. The events detected were then manually validated. Our preliminary results suggest that there are regular trends in the pro-duction of these hesitation events, which could distinguish them from other events within the structure of Portuguese. Our purpose with this work is to improve acoustic modeling for spon-taneous speech recognition systems. Some insights into the process of human speech communication for Portuguese are gained as well.