Seminar


       Department of Systems Engineering and Engineering Management,

                    The Chinese University of Hong Kong

------------------------------------------------------------------------------

Title:
From Ripples and Waves Onward-Framework and Modeling of Fluent
Speech Prosody

Speaker:
Professor Chiu-yu Tseng,
Research Fellow, Phonetics Laboratory,
Institute of Linguistics,
Academia Sinica, Taipei, Taiwan

Date : November 9, 2005 (Wednesday)

Time : 4:30 p.m. - 5:30 p.m.

Venue : Room 513, William M.W. Mong Engineering Building

(Engineering Building Complex Phase 2), CUHK

Abstract:
The prosody of fluent connected speech is much more complicated than
concatenating individual sentence intonations into strings.  We analyzed
speech corpora of read Mandarin Chinese discourses from a top-down perspective
on perceived units and boundaries, and consistently identified speech
paragraphs of multiple phrases that reflected discourse rather than sentence
effects in fluent speech.  Subsequent cross-speaker and cross-speaking-rate
acoustic analyses of identified speech paragraphs revealed systematic cross-
phrase prosodic patterns in every acoustic parameter, namely F0 contours,
duration adjustment, intensity patterns and in addition, boundary breaks. We
argue for a higher prosodic node that governs, constrains and group phrases to
derive speech paragraphs.  A hierarchical multi-phrase framework is
constructed to account for the governing effect, with complimentary production
and perceptual evidences.  We show how cross-phrase F0 and syllable duration
patterns templates are derived to account for the tune and rhythm
characteristic to fluent speech prosody, and argue for a prosody framework
that specifies phrasal intonations as subjacent sister constituent subject to
higher terms.  Output fluent speech prosody is thus cumulative results of
contributions from every prosodic layer.  To test our framework, we further
construct a modular prosody model of multiple phrase grouping with four
corresponding acoustic modules and begin testing a model with speech
synthesis.  To conclude, we argue that any prosody framework of fluent speech
should include prosodic contributions above individual sentences in
production, with considerations of its perceptual effects to online processing
and development of unlimited TTS could benefit most appreciably by capturing
and including cross-phrase relationships in prosody modeling.

Bio:
Professor Chiu-yu Tseng is Research Fellow of the Institute of Linguistics,
Academia Sinica. She received her Ph.D. in Linguistics from Brown University
in 1981. Her recent research focuses on speech prosody of Chinese as well as
related speech technologies. Details may be found at:
http://phslab.ihp.sinica.edu.tw/teacher/index.html