This paper presents another approach of Thai word segmentation, which is composed of two processes : syllable segmentation and syllable merging. Syllable segmentation is done on the basis of trigram statistics. Syllable merging is done on the basis of collocation between syllables. We argue that many of word segmentation ambiguities can be resolved at the level of syllable segmentation. Since a...

In our previous research, we demonstrated the validity of segmental unit input hidden Markov model (HMM), which regards successive four frame MEL-cepstrum coefficients as a feature vector. The vector is reduced to lower dimensions using the KL transform. However, the model considers only the correlation between frames in a short section, but not the correlation between the frames over a long se...

Modeling phonological units of speech is a critical issue in speech recognition. In this paper, we report our recent development of an overlapping-feature based phonological model which represents long-span contextual dependency in speech acoustics. In this model, high-level linguistic constraints are incorporated in automatic construction of the feature overlapping patterns and the associated ...

Tone is an essential language phenomenon for Mandarin Chinese language. Until now, we still do not know exactly how context affects tone pattern variation in continuous Mandarin speech. In this paper, we proposed a decision tree based approach to obtain the quantitative result of tone pattern variation in continuous Mandarin speech. Many possible factors other than tone of neighboring syllables...

The perception of duration-based syllabic rhythm was examined within a metrical framework. Participants assessed the duration patterns of four-syllable phrases set within the stress structure XxxX (an Abercrombian trisyllabic foot). Using on-screen sliders, participants created percussive sequences that imitated speech rhythms and analogous non-speech monotone rhythms. There was a tendency to e...

Recordings of careful readings and unscripted monologues in Swedish and Argentinean Spanish were auditorily and acoustically analyzed to test the hypothesis that temporal equalization of syllable-sized contoid-vocoid (CV) sequences typically occurs in casual as opposed to elaborated speaking styles. Results indicated 1) that CV durations and standard deviations appearing in unscripted Swedish t...

Languages vary in the types of contexts that affect prosodic prominence. This paper reports on a production study investigating how different types of foci influence prosody in Polish. The results show that focus and givenness in Polish are both marked prosodically, with pitch and intensity as the main acoustic correlates. Polish patterns like English in showing prosodic focus marking in a broa...

Previous work on the rhythm of Malaysian English has been based on impressionistic observations. This paper utilizes acoustic analysis to measure the rhythmic patterns of Malaysian English. Recordings of the read speech and spontaneous speech of 10 Malaysian English speakers were analyzed and compared with recordings of an equivalent sample of Singaporean English speakers. Analysis was done usi...

Duration is a primary factor both to achieve more naturalsounding synthesis and as an indicator of phrasal organization in speech recognition. In this study, we investigate pauses and durational patterns in spontaneous conversation, as well as how reliably such elements can serve as boundary-marking predictors across different types of speech corpora. Our results show that pause duration is sig...