Annotation of Valence Unfolding in Spoken Personal Narratives

14258

Authors

nan

Abstract

AbstractPersonal Narrative (PN) is the recollection of individuals’ life experiences, events, and thoughts along with the associated emotions in the form of a story. Compared to other genres such as social media texts or microblogs, where people write about experienced events or products, the spoken PNs are complex to analyze and understand. They are usually long and unstructured, involving multiple and related events, characters as well as thoughts and emotions associated with events, objects, and persons. In spoken PNs, emotions are conveyed by changing the speech signal characteristics as well as the lexical content of the narrative. In this work, we annotate a corpus of spoken personal narratives, with the emotion valence using discrete values. The PNs are segmented into speech segments, and the annotators annotate them in the discourse context, with values on a 5-point bipolar scale ranging from -2 to +2 (0 for neutral). In this way, we capture the unfolding of the PNs events and changes in the emotional state of the narrator. We perform an in-depth analysis of the inter-annotator agreement, the relation between the label distribution w.r.t. the stimulus (positive/negative) used for the elicitation of the narrative, and compare the segment-level annotations to a baseline continuous annotation. We find that the neutral score plays an important role in the agreement. We observe that it is easy to differentiate the positive from the negative valence while the confusion with the neutral label is high. Keywords: Personal Narratives, Emotion Annotation, Segment Level Annotation

Keywords

nan

Full Content

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 7004–7013 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 7004 Annotation of Valence Unfolding in Spoken Personal Narratives A. Tammewar2, F. Braun1, G. Roccabruna2, S. P. Bayerl1, K. Riedhammer1, G. Riccardi2 1 Technische Hochschule N¨urnberg Georg Simon Ohm, Keßlerplatz 12, 90489 N¨urnberg, GERMANY 2 Signals and Interactive Systems Lab, DISI, University of Trento, ITALY {franziska.braun, sebastian.bayerl, korbinian.riedhammer}@th-nuernberg.de {aniruddha.tammewar, gabriel.roccabruna, giuseppe.riccardi}@unitn.it Abstract Personal Narrative (PN) is the recollection of individuals’ life experiences, events, and thoughts along with the associated emotions in the form of a story. Compared to other genres such as social media texts or microblogs, where people write about experienced events or products, the spoken PNs are complex to analyze and understand. They are usually long and unstructured, involving multiple and related events, characters as well as thoughts and emotions associated with events, objects, and persons. In spoken PNs, emotions are conveyed by changing the speech signal characteristics as well as the lexical content of the narrative. In this work, we annotate a corpus of spoken personal narratives, with the emotion valence using discrete values. The PNs are segmented into speech segments, and the annotators annotate them in the discourse context, with values on a 5 point bipolar scale ranging from -2 to +2 (0 for neutral). In this way, we capture the unfolding of the PNs events and changes in the emotional state of the narrator. We perform an in-depth analysis of the inter-annotator agreement, the relation between the label distribution w.r.t. the stimulus (positive/negative) used for the elicitation of the narrative, and compare the segment-level annotations to a baseline continuous annotation. We find that the neutral score plays an important role in the agreement. We observe that it is easy to differentiate the positive from the negative valence while the confusion with the neutral label is high. Keywords: Personal Narratives, Emotion Annotation, Segment Level Annotation 1. Introduction The Personal Narrative (PN) is the recollection of in- dividuals’ life experiences, events, and thoughts along with the associated emotions in the form of a story. PNs could be shared with others in different ways such as by meeting and telling them the story directly or by calling them, by posting them on social media, or by writing a blog. Different genres or domains have different structures of PNs and may involve the use of different modalities such as speech, writings, facial expressions, or gestures for conveying emotions. So- cial media posts tend to be concise and more specific about the events and emotions. Spoken personal nar- ratives, on the other hand, have a more complex struc- ture. They are long and contain descriptions of multiple sub-events, characters involved and the emotions felt (Tammewar et al., 2020). Rich information provided through PNs can help better understand the emotional state of the narrator, thus PNs are frequently used in psychotherapy and mental well-being applications. In psychotherapy sessions, therapists often ask clients to narrate …

Content truncated to 500 words

Labeled Text

AbstractPersonal Narrative (PN) is the recollection of individuals’ life experiences, events, and thoughts along with the associated emotions in the form of a story. Compared to other genres such as social media texts or microblogs, where people write about experienced events or products, the spoken PNs are complex to analyze and understand. They are usually long and unstructured, involving multiple and related events, characters as well as thoughts and emotions associated with events, objects, and persons. In spoken PNs, emotions are conveyed by changing the speech signal characteristics as well as the lexical content of the narrative. In this work, we annotate a corpus of spoken personal narratives, with the emotion valence using discrete values. The PNs are segmented into speech segments, and the annotators annotate them in the discourse context, with values on a 5-point bipolar scale ranging from -2 to +2 (0 for neutral). In this way, we capture the unfolding of the PNs events and changes in the emotional state of the narrator. We perform an in-depth analysis of the inter-annotator agreement, the relation between the label distribution w.r.t. the stimulus (positive/negative) used for the elicitation of the narrative, and compare the segment-level annotations to a baseline continuous annotation. We find that the neutral score plays an important role in the agreement. We observe that it is easy to differentiate the positive from the negative valence while the confusion with the neutral label is high. Keywords: Personal Narratives, Emotion Annotation, Segment Level Annotation

Publication Details

Paper ID:
14258
Year:
2022
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan