Decoding sarcasm and sincerity: a comparative analysis of prosodic features in English-Chinese bilingual speech using SVM
This study explores the nuanced expression of sarcasm among English-Chinese bilinguals using Support Vector Machine (SVM) classification and analysis of prosodic features in spoken language. Addressing a gap in current research, this paper investigates the manifestation of sarcasm and sincerity in a...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
Nanyang Technological University
2024
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/174808 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | This study explores the nuanced expression of sarcasm among English-Chinese bilinguals using Support Vector Machine (SVM) classification and analysis of prosodic features in spoken language. Addressing a gap in current research, this paper investigates the manifestation of sarcasm and sincerity in a bilingual context, examining cross-linguistic differences and similarities. With a participant pool of 10 bilingual undergraduate students, the study analyses pitch, intensity, speech rate, and harmonics-to-noise ratio to identify patterns that classify sarcastic and sincere utterances. The results indicated subtler sarcasm cues in English and more pronounced cues in Mandarin, with the radial kernel SVM model achieving high accuracy in the Chinese dataset. These findings highlight the complex interplay of linguistic and cultural factors in sarcasm perception and expression among bilingual individuals, offering valuable insights for enhancing communication technology. Despite limitations like a small sample size and controlled elicitation scenarios, the study contributes to a better understanding of sarcasm in bilingual communication, suggesting a need for further research in more naturalistic settings and with more detailed language proficiency assessments. |
---|