A Manually Annotated Resource for the Investigation of Nasal Grunts

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 6514–6522 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 6514 A Manually Annotated Resource for the Investigation of Nasal Grunts Aurélie Chlébowski, Nicolas Ballier Université de Paris CLILLAC-ARP, F-75013 Paris, France aurelie.chlebowski@hotmail.fr, nicolas.ballier@u-paris.fr Abstract This paper presents an annotation framework for nasal grunts of the whole French CID corpus (Bertrand et al., 2008). The acoustic components under scrutiny are justified and the annotation guidelines are described. We carefully characterise the acoustic cues and visual cues followed by the annotator, especially for non-modal phonation types. The conventions followed for the annotation of interactional and positional properties of grunts are explained. The resulting datasets after data extraction with Praat scripts (Boersma and Weenink, 2019) are analysed with R (R Core Team, 2017), focusing on duration. We analyse the effect of non-modal phonation (especially ingressive phonation) on duration and discuss a specialisation of grunts observed in the CID for grunts with ingressive phonation. The more general aim of this research is to establish putative core and additive properties of grunts and a tentative typology of grunts in spoken interactions. Keywords: grunts, paralinguistic features, disfluencies, manual annotation 1. Introduction Although they might be found as entries in dictionaries, “non-lexical conversational sounds” (Ward, 2006), also known as interjections, have for a long time been deemed to be negligible and were discarded from linguistics analysis. The idea that those linguistic phenomena might in fact convey some information about the speaker arose later in various domains such as discourse analysis. Nonetheless, for their linguistic status is still debated today (Tottie, 2019), there is no consensus as to the way those sounds should be transcribed or analysed. Non-lexical conversational sounds being sounds in the first place, Chlébowski and Ballier (2015) hypothesised that part of the information conveyed by non-lexical conversational sounds must be related to their acoustic form. Their semasiological approach contrast with previous work considered as onomasiological (e.g. Delomier, 1999; Clark and Tree, 2002; Tottie, 2019), in the sense that they wished to characterise the various acoustic signs under scrutiny to possibly capture and document putative meanings for some types of grunts. Chlébowski and Ballier (2015) proposed to follow the “compositional model” established by Ward (2006) and analysed a sub- category of non-lexical conversational sounds that they called “nasal grunts”. Despite being preliminary, their study revealed the need for a homogenisation of the transcription of nasal grunts and indicated that a bottom- up analysis of such sounds was required in order to 1) elaborate a sound-meaning mapping of their acoustic components and 2) study their behaviour in interactional situation. In that respect, this paper proposes to focus on the acoustic analysis of nasal grunts in the Corpus of Interactional Data (hereafter CID; Bertrand et al., 2008). We propose a procedure for the annotation of the acoustic categorisation of nasal grunts based on observable features, in order to study their distribution in interactional context. The remainder of the paper is …

Content truncated to 500 words