This paper describes work that shows how the acoustic features of laughter in Japanese speech vary according to conversational partner, reflecting the social status of laughter, and confirming that even such a simple sound is affected by non-linguistic factors such as social or intercultural relationships. Neural networks were successfully trained to identify the nature of the interlocutor from...

In this work, a spontaneous speech corpus of broadcasted television material in European Portuguese (EP) is presented. We decided to name it ProGmatica as it is meant to combine prosody information under a pragmatic framework. Our purpose is to analyse, describe and predict the prosodic patterns that are involved in speech acts and discourse events. It is also our goal to relate both prosody an...

The aim of this paper is to present a systematical analysis of the acoustic correlates of hierarchical prosodic boundaries. This analysis is based on a large labeled corpus of Mandarin Chinese. The acoustic correlates include the lowest value of pitch and the duration of the silence. The prosodic structure is defined thanks to a perception experiment. The main results are : a) the declination o...

This is an experimental study of tonal correlates of prosodic phrasing and focus production in Greek. The results indicate: (1) the tonal correlates of phrasing are a rising tonal command at phrase boundaries and a deaccentuation of the preboundary lexical stress; (2) the tonal correlates of focus are a local tonal range expansion aligned with the stressed syllable of the last lexical unit in f...

In this paper we investigate the performance of different classification paradigms, testing each with a range of acoustic features, to find a system that is well suited to speaker likeability classification. We introduce a Sparse Representation Classifier for paralinguistic classification and explore the role of training data selection for a GMM classifier. Results demonstrate that (1) Single d...

Prosody is an important cue for identifying dialog acts. In this paper, we show that modeling the sequence of acoustic– prosodic values as n-gram features with a maximum entropy model for dialog act (DA) tagging can perform better than conventional approaches that use coarse representation of the prosodic contour through summative statistics of the prosodic contour. The proposed scheme for expl...

It has been widely documented that child-directed speech (CDS) may differ notably from adult-directed speech in terms of prosody (e.g. [1] [2] [3] [4]). Common features include higher and greater range of pitch, especially in stressed syllables; longer duration of individual words, more prominent final lengthening, slower speech rate, higher amplitude, shorter utterances and longer pauses, more...

Research on verbal irony has found that prosodic features such as pitch range expansion, syllable lengthening, and specific intonational contours are common prosodic resources that languages use to mark irony in speech. Yet little is known about the relative weight of these prosodic features in the detection of irony in languages that use these three prosodic correlates. In this paper we presen...

Variation in pitch, amplitude and rhythm adds crucial paralinguistic information to human speech. Such prosodic cues can reveal information about the meaning or emphasis of a sentence or the emotional state of the speaker. To examine the hypothesis that sensitivity to prosodic cues is language independent and not human specific, we tested prosody perception in a controlled experiment with zebra...

Higher-level features based on linguistic or long-range information have attracted significant attention in automatic speaker recognition. This article briefly summarizes approaches to using higher-level features for text-independent speaker verification over the last decade. To clarify how each approach uses higher-level information, features are described in terms of their type, temporal span...