Interpreting fttUy natural speech is an important goal for spoken language understanding systems. However, while corpus studies have shown that about 10% of spontaneous utterances contain selfcorrections, or REPAIRS, little is known about the extent to which cues in the speech signal may facilitate repair processing. We identify several cues based on acoustic and prosodic analysis of repairs in...

The authors present a study of prosodic turn-taking indicators. The aim was to investigate whether some of the prosodic cues increase in quality or quantity if the optical feedback channel in the verbal conversation is missing. For the study we built up an experimental setup in which conversational partners held a conversation once with and once without an optical feedback channel. A detailed t...

This study investigates the effect of sentence-level prosody on production of English lexical stress, comparing L1 English and L1 Taiwan Mandarin speaker groups. 4 L1 North American English speakers and 9 L1 Taiwan Mandarin speakers were asked to produce a set of 20 disyllabic and multisyllabic words embedded in three different prosodic contexts: neutral broad focus, at a phrase/sentence bounda...

This study investigates the role of prosodic unit boundaries in on-line lexical access by Brazilian Portuguese adult speakers. Two types of prosodic constituents are considered: prosodic words (ω) and phonological phrases (ɸ). Motivated by French experimental results, we proposed two experiments in order to examine on-line lexical access in auditory sentences, considering that prosodic unit bou...

Prosody simultaneously encodes different kinds of information, including the type of speech act of an utterance (e.g., falling declarative vs. rising interrogative intonational tunes), the location of semantic focus (via prosodic prominence), and syntactic constituent structure (via prosodic phrasing). The syntactic/semantic functional dimensions (speech act, focus, constituency) are orthogonal...

This study explores the potential of utilising certain prosodic qualities of function-specific vocal expressions in order to design effective non-speech user interface sounds. In an empirical setting, utterances with four context-situated communicative functions were produced by 20 participants. Time series of fundamental frequency (F0) and intensity were extracted from the utterances and analy...

This paper reports an investigation of features relevant for classifying two speaking styles, namely, conversational speaking style and clear (e.g. hyper-articulated) speaking style. Spectral and prosodic features were automatically extracted from speech and classified using decision tree classifiers and multilayer perceptrons to achieve accuracies of about 71% and 77% respectively. More intere...

Previous work has shown that Dutch listeners use prosodic information in the speech signal to optimise morphological processing: Listeners are sensitive to prosodic differences between a noun stem realised in isolation and a noun stem realised as part of a plural form (in which the stem is followed by an unstressed syllable). The present study, employing a lexical decision task, provides an add...