Annotating Interruption in Dyadic Human Interaction

15797

Authors

nan

Abstract

AbstractIntegrating the existing interruption and turn switch classification methods, we propose a new annotation schema to annotate different types of interruptions through timeliness, switch accomplishment and speech content level. The proposed method is able to distinguish smooth turn exchange, backchannel and interruption (including interruption types) and to annotate dyadic conversation. We annotated the French part of NoXi corpus with the proposed structure and use these annotations to study the probability distribution and duration of each turn switch type.

Keywords

nan

Full Content

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 2292–2297 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 2292 Annotating Interruption in Dyadic Human Interaction Liu YANG, Catherine ACHARD, Catherine PELACHAUD Institut des Syst`emes Intelligents et de Robotique Sorbonne University, 75005, Paris, France yangl@isir.upmc.fr, catherine.pelachaud@isir.upmc.fr, catherine.achard@isir.upmc.fr Abstract Integrating the existing interruption and turn switch classification methods, we propose a new annotation schema to annotate different types of interruptions through timeliness, accomplishment and speech content level. We annotated the French part of NoXi corpus with the proposed structure and use these annotations to study the probability distribution and duration of each turn switch type. Keywords: Interruption Annotation, Turn-taking, Dyadic Conversation 1. Introduction Interruption is a phenomenon that often occurs in daily conversations and cannot be ignored when studying turn-taking during human-human interaction. Interrup- tion is viewed as a special phenomenon in turn-taking, which plays a role in regulating the rhythm of dialogue. Many researchers interpret interruptions as evidence of power and dominance because they violate the princi- ple of one person speaking at a time (Ferguson, 1977), (Tannen and others, 1991). However, depending on the context, some interruptions are not regarded as signs of power and discomfort, but as cooperative behaviors, which provide help to the speaker (Yang, 2001). Inter- ruptions are therefore important during a dialogue and can have very diverse functions. Our aim is to study interruptions in human-human in- teraction to understand the impact of interruption types on the on-going interaction and how the interrupter is perceived by the interruptee. To this aim, we propose a schema to annotate interruption based on theoretical models and use it to analyze a human-human dialogue corpus. 2. Background Conversation analysis (CA) has been carried out since the 1960s with the work of Harvey Sacks. He proposed a study on the orderliness of conversation by describ- ing its most basic structure: turn-taking (Sacks et al., 1978). Turn-taking is a fundamental aspect of the con- versation since it is not possible to speak and listen at the same time. Interlocutors need to coordinate to ex- change speaking turns to maintain the conversation. Several conversation coding structures have been used for previous studies (Nakazato, 2000; Truan and Ro- mary, 2021; Ten Bosch et al., 2004; Christodoulides and Avanzi, 2015; Jokinen et al., 2013; Enomoto et al., 2020; Heeman et al., 2006), however they didn’t give details about different interruption types. As a special embodiment of turn-taking principles, in- terruption is a natural phenomenon in conversations and also an important topic in the conversational struc- ture analysis. It reflects the communicator’s attempt to take the floor without letting the speaker finish his/her utterance. Allwood and colleagues have integrated interruption in their coding schema according its function (Allwood, 2001). However they considered interruption as only part of speech overlap and excluded interruption initi- ated within a pause. Schegloff and Sacks (Schegloff and Sacks, 1973) de- fines three types of simultaneous speech: interruption, overlap and parenthetical comments such …

Content truncated to 500 words

Labeled Text

AbstractIntegrating the existing interruption and turn switch classification methods, we propose a new annotation schema to annotate different types of interruptions through timeliness, switch accomplishment and speech content level. The proposed method is able to distinguish smooth turn exchange, backchannel and interruption (including interruption types) and to annotate dyadic conversation. We annotated the French part of NoXi corpus with the proposed structure and use these annotations to study the probability distribution and duration of each turn switch type.

Publication Details

Paper ID:
15797
Year:
2022
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan