Mono-syllabic interjections are often used to express a reaction in conversational speech. It is known that there is relationship between the speaking style, given by intonation and voice quality-related prosodic features, and the paralinguistic information carried by an interjection. However, it is also known that this relationship is dependent on the interjection type. In the present work, we...

The paralinguistic features, however this conference classifies the “Paralinguistic analysis” as Speaker identification, Keyword/topic spotting, and Language identification, include emotional aspects of voice, which is focused recently on interpersonal communications. Study starts from description and statistics of those features. Among many acoustic characteristics, we investigated features re...

Speaker and paralinguistic properties of dialogue speech that affect the timing at turn changes are investigated by analyzing the UU Database for Paralinguistic Information Studies. The results showed a large variation among speakers and a strong interaction with partner in pause/overlap duration. In addition, perceived emotional states of utterances had significant effects on the pause/overlap...

Interjections are often used in dialogue communication for expressing a reaction (such as agreement, surprise and disgust) to the interlocutor. Thus, a correct interpretation of the paralinguistic information (intention, attitude or emotion) carried by interjections is important for achieving a smooth dialogue interaction between humans and robots. In the present work, analyses are conducted on...

Corpus based text-to-speech systems currently produce very natural synthetic sentences, though limited to a neutral inexpressive speaking style. Paralinguistic elements are some of the expressive features one would most like to introduce. In this paper, we describe a new method for introducing laughter and hesitation in synthetic speech. Thanks to a small dedicated acoustic database, this metho...

Many paralinguistic tasks are closely related and thus representations learned in one domain can be leveraged for another. In this paper, we investigate how knowledge can be transferred between three paralinguistic tasks: speaker, emotion, and gender recognition. Further, we extend this problem to cross-dataset tasks, asking how knowledge captured in one emotion dataset can be transferred to an...

The local speech rate is one of the important prosodic parameters for representing both linguistic and paralinguistic information. The present paper presents a command-response model for the process of speech rate control, and applies it to the analysis of the effects of emphasis on the local speech rate in English utterances. It shows that the local speech rate as a function of time can be app...

“Rikimi” is a “pressed-type” voice quality that appears in Japanese conversational speech for expressing paralinguistic information related to emotional or attitudinal behaviors of the speaker. We conducted acoustic, electroglottographic (EGG) and paralinguistic analyses on speech segments including “rikimi”, extracted from spontaneous dialogue speech data. “Rikimi” may be accompanied by severa...