Human Computer Interaction, Artificial Intelligence and Intelligent Augmentation
Disentangling the Effects of Paralinguistic Cues in Bolstering Listeners’ Engagement with Podcasters
Paper Type
short
Paper Number
2498
Description
Despite fierce competition among podcasters, there is a dearth of research that has sought to elucidate the role of paralinguistic elements in fostering listeners’ engagement with podcasters on audio platforms. Building on voice persuasion theory, we not only posit prosodic cues (i.e., speech rate, speech pause, and speech stress) as factors affecting listeners’ engagement, but we further postulate pitch, a habitual property of podcasters’ voice, as having a moderating influence on the effects of the abovementioned prosodic cues. To empirically validate our hypotheses, we employ deep learning to extract the four paralinguistic cues from 10,000 audio files in 221 educational albums obtained from a leading audio platform in China. While analytical results confirm speech pause and speech stress as antecedents of speech rate and that lower speech rate is effective in bolstering listeners’ engagement, pitch is revealed to exert no moderating influence on relationship between speech rate and engagement.
Recommended Citation
Liu, Xiaohui; Liu, Fei; Li, Yijing; and Lim, Eric, "Disentangling the Effects of Paralinguistic Cues in Bolstering Listeners’ Engagement with Podcasters" (2020). ICIS 2020 Proceedings. 22.
https://aisel.aisnet.org/icis2020/hci_artintel/hci_artintel/22
Disentangling the Effects of Paralinguistic Cues in Bolstering Listeners’ Engagement with Podcasters
Despite fierce competition among podcasters, there is a dearth of research that has sought to elucidate the role of paralinguistic elements in fostering listeners’ engagement with podcasters on audio platforms. Building on voice persuasion theory, we not only posit prosodic cues (i.e., speech rate, speech pause, and speech stress) as factors affecting listeners’ engagement, but we further postulate pitch, a habitual property of podcasters’ voice, as having a moderating influence on the effects of the abovementioned prosodic cues. To empirically validate our hypotheses, we employ deep learning to extract the four paralinguistic cues from 10,000 audio files in 221 educational albums obtained from a leading audio platform in China. While analytical results confirm speech pause and speech stress as antecedents of speech rate and that lower speech rate is effective in bolstering listeners’ engagement, pitch is revealed to exert no moderating influence on relationship between speech rate and engagement.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.