The paper presents the module for automatic generation of prosodic features of synthesized speech, namely, f0 targets and phonetic segment durations, within the speech synthesizer AlfaNumTTS, the most sophisticated speech synthesis system for Serbo-Croatian language to date. The module is based on regression trees trained on a studio recorded single speaker database of Serbo-Croatian. The datab...

The Dutch dialect of Venlo has a lexical tone opposition comparable to the distinction between Accent I and Accent II in Scandinavian. The two word tone patterns are realised in a variety of different ways, depending on the intonation contour, on whether the word has a focus tone, and on whether it occurs finally or nonfinally in the intonational phrase (IP). Twelve such contexts are identified...

In intonational phonology and speech synthesis research, it has been suggested that the relative informativeness of a word can be used to predict pitch prominence. The more information conveyed by a word, the more likely it will be accented. But there are others who express doubts about such a correlation. In this paper, we provide some empirical evidence to support the existence of such a corr...

This paper contributes phonetic evidence to ongoing debate regarding the position of Moroccan Arabic in prosodic typology, with the aim of determining how phrase-edge tonal events should be represented in the intonational phonology of the language. A salient phrase-final rise-fall tonal event, found in MA polar questions, is used as a case study. We examined the alignment of the f0 peak of this...

The goal of this paper is to investigate the nature of the high tones realized on finally stressed words in Turkish. Following Ipek & Jun’s [1] AM model of intonational phonology of Turkish, it was hypothesized that the high tone realized on the last syllable of a phrase (i.e., Intermediate Phrase (ip)) is realized differently from that of a phrase-medial prosodic word (PW), reflecting the pros...

Our long-term goal is to test the feasibility of basing a recognition system on a model of human speech processing. Key hypotheses are: (1) Computational models for reasoning from incomplete knowledge provide a useful metaphor for many aspects of human speech understanding even at levels where highly automatic perceptual processes are at work; (2) A computational model of human cochlear process...

Previous work on the intonational phonology of Miami Cuban (MC) Spanish [1, 2, 3] has revealed that broad focus declaratives and absolute interrogatives have the same nuclear tones. In laboratory speech, [2] found that these two sentence types, when compared paradigmatically, show significant differences in peak scaling and pitch range [2]. The present study examined peak scaling with syntagmat...

Although the pitch of the human voice is continuously variable, some linguists contend that intonation in speech is restricted to a small, limited set of patterns. This claim is tested by asking subjects to mimic a block of 100 randomly generated intonation contours and then to imitate themselves in several successive sessions. The produced f0 contours gradually converge towards a limited set o...

A central issue in speech intonation research concerns how fundamental frequency (f0) variation relates to phonological categories. The hypothesis was tested that pitch range variation which affects whether one syllable is higher or lower than another would elicit categorical shifts in f0 extremum timing in an imitation task. Participants heard synthetic versions of the phrase Some lemonade wit...