Abstract

This study presents the results of an acoustic and statistical analysis of labiodental and interdental fricative consonants, voiced and voiceless, from a variety of Mapudungun spoken in Alto Biobío. The present study aims to determine whether spectral, durational and amplitude parameters are able to differentiate these sounds by place of articulation and voicing. Results showed that spectral moments, duration, absolute amplitude and normalized amplitude were able to distinguish between voiced and voiceless fricatives, whilst variance, duration and normalized amplitude were able to separate labiodental from interdental consonants. However, the capability that variance and duration absolute had at distinguishing between voiced and voiceless consonants is restricted to labiodental consonants when the interaction between voicing, place of articulation and participant's sex was considered. These results are discussed in the light of literature on the usefulness of these acoustic measurements at capturing contrasts of manner of articulation, place and voicing, and in view of previous studies on the properties of the phonetic and phonemic systems in this variety of Mapudungun.