van Santen, J.P.H., "Prosodic modeling in text-to-speech synthesis", Eurospeech 97, KN -- 19-27. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Positional Effects On Stressed Vowel Duration In Standard.. - van Santen, Imperio (Correct)
.... tend to keep durations of larger phonological entities (e.g. phrases, words, feet, syllables) relatively constant; this implies that as a given entity contains more constituents (e.g. syllables in words and feet; words in phrases) the durations of these individual constituents are reduced [13]. Examples of constituency effect hypotheses are Lehiste s work on isochrony [5] and Campbell s on syllable duration [1] The hypothesis in [6] correctly predicts that the [e] of setola would be shorter than that of seta, since in the first case the word in which the vowel occurs has an extra ....
van Santen, J. 1997. Prosodic modeling in text-to-speech synthesis. In Proceedings of Eurospeech-97 (Rhodes).
Reducing Audible Spectral Discontinuities - Klabbers, Veldhuis (2001) (9 citations) (Correct)
....synthesis the task is to distinguish these instances when their spectra are perceptually different. Therefore, it should be investigated whether some distance measures can be found that correspond to human perception in that they are able to distinguish perceptually relevant differences in spectra [26]. An investigation that ran parallel to ours [20] 31] also aimed at performing a perceptual evaluation of distance measures in the context of speech synthesis. In their study, listeners had to judge the difference between a pair of stimuli on a scale from zero to five. One stimulus was the ....
....to the conclusion that besides coarticulation there is always random variation in the pronunciation of the stimuli. This was also observed by [24] who found variations in excess of 50 Hz for a vowel in repetitions of the exact same phrase as uttered by a highly professional speaker. Reference [26] reports even larger and variations (up to 250 Hz) in the repeated pronunciation of I in six and million by a professional speaker. This indicates the need to record several instances of a nonsense word and choose the one that is optimal for the database. ....
J. Van Santen, "Prosodic modeling in text-to-speech synthesis," in Proc. 5th Eur. Conf. Speech Communication Technology (EUROSPEECH'97), Rhodes, Greece, 1997, pp. KN19--28.
Description Of The Bell Labs Intonation System - Jan Van Santen (1998) (1 citation) Self-citation (Van santen) (Correct)
No context found.
van Santen, J. Prosodic modeling in text-to-speech synthesis. In Proceedings of Eurospeech-97 (Rhodes, September 1997).
Description Of The Bell Labs Intonation System - van Santen, Möbius, Venditti.. (1998) (1 citation) Self-citation (Van santen) (Correct)
....position (in the minor phrase, the minor phrase in major phrase, etc. factors predictive of prominence, and intrinsic pitch. The multiplicative model is often used in segmental duration modeling. It makes the important and not necessarily accurate assumption of directional invariance [9]: holding all factors but one constant, the effects of the varying factor always have the same direction. This may often be true in segmental duration; e.g. when two occurrences of the same vowel involve identical contexts, except for syllabic stress, the stressed occurrence is likely to be ....
van Santen, J. Prosodic modeling in text-to-speech synthesis. In Proceedings of Eurospeech-97 (Rhodes, September 1997).
Duration Modeling in a Restricted-Domain.. - Cordoba, Montero, .. (2001) (Correct)
No context found.
van Santen, J.P.H., "Prosodic modeling in text-to-speech synthesis", Eurospeech 97, KN -- 19-27.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC