Prosodic boundary prediction is the key to improving the intelligibility and naturalness of synthetic speech for a TTS system. This paper investigated the problem of automatic segmentation of prosodic word and prosodic phrase, which are two fundamental layers in the hierarchical prosodic structure of Mandarin Chinese. Maximum Entropy (ME) Model was used at the front end for both prosodic word a...

One major feature of the prosody of Mandarin Chinese speech flow is prosodic phrase grouping [1, 2, and 3]. Phrasal and sentential intonations are governed by a prosody framework that structurally group phrases into a larger/longer and identifiable unit. An overall prosody pattern of such phrase grouping with prosodic specifications is superimposed on phrase group. In turn, individual phrasal i...

Prosodic information is crucial for spoken language comprehension and especially for syntactic parsing, because prosodic cues guide the hearer's syntactic analysis. The time course and mechanisms of this interplay of prosody and syntax are not yet well-understood. In particular, there is an ongoing debate whether local prosodic cues are taken into account automatically or whether they are proce...

Speech is structured into parts by syntactic and prosodic breaks. In locally syntactic ambiguous sentences, the detection of a syntactic break necessarily follows detection of a corresponding prosodic break, making an investigation of the immediate interplay of syntactic and prosodic information impossible when studying sentences in isolation. This problem can be solved, however, by embedding s...

We compare the F0 output and chunking size of speech units of L1 English, TW L2 English and L1 Mandarin to see what some of the major intrinsic prosodic difference could be, what prosodic features could account for TW L2 English and in what way prosodic transfer occurs. Results show that the fundamental prosodic difference of the two languages is how in the pitch domain English requires sharper...

This paper presents a study of prosodic phrasing in a non-tonal dialect of Kammu, a MonKhmer language spoken in Northern Laos. Prosodic phrasing is seen as correlated with syntactic and informational structures, and the description is made referring to these two levels. The material investigated comprises sentences of different lengths and syntactic structures, uttered by seven male speakers. I...

This paper presents a new framework for improved large vocabulary Mandarin speech recognition using prosodic features. The prosodic information is formulated in a probabilistic model well compatible to the conventional maximum a posteriori (MAP) framework for large vocabulary speech recognition. A set of prosodic features considering the special characteristics of Mandarin Chinese is developed,...

The study explored the perception and acoustic features of prosodic hierarchy boundaries in Uygur and Mandarin Chinese. The results of the perception study showed that native speakers of Mandarin Chinese were more sensitive to prosodic hierarchy boundaries than native speakers of Uygur. The results of a detailed acoustic analysis indicated that the acoustic features of Uygur prosodic hierarchy ...

In Prosodic Phonology Theory (Selkirk 1981, 1984, Nespor & Vogel 1986, Hayes 1989), the particular hierarchical organization of a set of universally available prosodic constituents (σ, ω, φ, etc.) captures the rhythmic patterns of languages and the domain of application of a wide variety of phonological processes. Rather than targeting isolated segments or arbitrary groups of segments within th...

In the present paper an analysis by synthesis study of mandarin speech prosody is carried out. The mandarin prosodic features are discussed from two salient perspectives, specifically: the function of prosody and the form of prosody. The symbolic representation of prosodic form with the INTSINT (INternational Transcription System for INTonation) system [1] reduces the surface complexity of a pr...