Interspeech 2019 2019
DOI: 10.21437/interspeech.2019-2131
|View full text |Cite
|
Sign up to set email alerts
|

Conversational and Social Laughter Synthesis with WaveNet

Abstract: The studies of laughter synthesis are relatively few, and they are still in a preliminary stage. We explored the possibility of applying WaveNet to laughter synthesis. WaveNet is potentially more suitable to model laughter waveforms that do not have a well-established theory of production like speech signals. Conversational laughter was modelled with a spontaneous dialogue speech corpus based on WaveNet. To obtain more stable laughter generation, conditioning WaveNet by power contour was proposed. Experimental… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
5
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
5
1

Relationship

1
5

Authors

Journals

citations
Cited by 7 publications
(5 citation statements)
references
References 22 publications
0
5
0
Order By: Relevance
“…In order for conversational agents to be able to produce fillers or laughs, at least two research questions should be considered: how to generate them, and when to generate them. Regarding how, we are working on a research project to synthesize laughter from corpora of natural conversation [63], [64]. There are also attempts to synthesize fillers [37], [38], [65].…”
Section: Discussionmentioning
confidence: 99%
“…In order for conversational agents to be able to produce fillers or laughs, at least two research questions should be considered: how to generate them, and when to generate them. Regarding how, we are working on a research project to synthesize laughter from corpora of natural conversation [63], [64]. There are also attempts to synthesize fillers [37], [38], [65].…”
Section: Discussionmentioning
confidence: 99%
“…To analyze naturally occurred speech-laugh, online gaming voice chat corpus with emotional label (OGVC) [5] was used as the speech material. The dialog speech of seven out of thirteen speakers was used for this analysis because these dialogs were already annotated with laugh label and used for laughter synthesis [4]. The duration of the dialog was about 60 minutes long for each speaker and a total of seven hours of dialog were used for this analysis.…”
Section: Speech-laugh Materials 21 Corpusmentioning
confidence: 99%
“…Laughter generation in robots and agents has also been addressed but primarily in terms of animation or movement to produce realistic laughs ( Niewiadomski and Pelachaud, 2012 ; Ishi et al, 2016a ; Ishi et al, 2019 ). For the majority of agents, the range of laughter utterances is restricted by the text-to-speech system and is unable to generate speech laughs, although recent research has addressed this issue by producing a large variety of natural-sounding laughs ( Mori et al, 2019 ; Tits et al, 2020 ; Luong and Yamagishi, 2021 ).…”
Section: Related Workmentioning
confidence: 99%