The intonational phrase is hypothesized to represent a meaningful unit of analysis in spoken language interpretation. We present results on the identification of intonational phrase boundaries from acoustic features using classification and regression trees (CART). Our training and test data are taken from the Boston Directions Corpus (task-oriented monologue) and the HUB-IV Broadcast News data...

This paper introduces a new research project Simulating Intonational Varieties of Swedish (SIMULEKT). The basic goal of the project is to produce more precise and thorough knowledge about some major intonational varieties of Swedish. In this research effort the Swedish prosody model plays a prominent role. A fundamental idea is to take advantage of speech synthesis in different forms. In our an...

Automatic transcription of prosody is necessary for spoken language understanding. Prominence and intonational boundaries are routinely used to convey meaning beyond that expressed in the lexical content of speech. Using a classiÞcation rule learning algorithm and computationally light acoustic and syntactic features, detection of pitch accent at 87% on spontaneous elicited speech were attained...

A method is introduced for using hidden Markov models (HMMs) to model intonational structure. HMMs are probabilistic and can capture the variability in structure which previous finite state network models lack. We show how intonational tunes can be modelled by separate HMMs and how HMMs can be used in a recognition system to automatically determine the tune type of an utterance.

BACKGROUND The right shift (RS) theory of handedness suggests that poor phonology may occur in the general population as a risk associated with absence of an agent of left cerebral speech, the hypothesised RS + gene. The theory predicts that poor phonology is associated with reduced bias to right-handedness. METHODS A representative cohort of primary school children was assessed on tests of p...

This paper examines the tonal structure of yes-no questions in Chaha, using the autosegmental-metrical model of intonational phonology. Yes-no questions can be formed in two ways: lengthening the final syllable or attach the question suffix to the final word. They exhibit different F0 patterns at the end of the utterance. We propose that the same underlying tonal structure is realized different...

A universal characteristic of speech is that utterances are generally broken down phonologically into smaller phrases which are marked by suprasegmental features such as intonational events and/or final lengthening. Moreover, phrases can be further divided into smaller-sized constituents. These constituents of varying size, or 'prosodic units', are typically characterised as performing the dual...