Abstract

This paper introduces the tilt intonational model and describes how this model can be used to automatically analyse and synthesize intonation. In the model, intonation is represented as a linear sequence of events, which can be pitch accents or boundary tones. Each event is characterised by continuous parameters representing amplitude, duration and tilt (a measure of the shape of the event). The paper describes a event detector, in effect an intonational recognition system, which produces a transcription of an utterance's intonation. The features and parameters of the event detector are discussed and performance figures are shown on a variety of read and spontaneous speaker independent conversational speech databases. Given the event locations, algorithms are described which produce an automatic analysis of each event in terms of the Tilt parameters. Synthesis algorithms are also presented which generate F0 contours from Tilt representations. The accuracy of these is shown by comparing...

Citations

422	An Inequality and Associated Maximization Technique in Statistical Estimation of a Markov Process - Baum - 1972
411	SWITCHBOARD: Telephone speech corpus for research and development - Godfrey, Holliman, et al. - 1992
240	The Phonology and Phonetics of English Intonation - Pierrehumbert - 1980
218	ToBI: A standard for labelling English prosody - Silverman, Beckman, et al. - 1992
83	Pitch Determination of Speech Signals - Hess - 1983
78	Intonational invariance under changes in pitch range and length - Liberman, Pierrehumbert - 1984
56	The 1994 HTK large vocabulary speech recognition system - Woodland, Leggetter, et al. - 1995
52	The Intonational System of English - Liberman - 1975
43	Prosody and Speech Recognition - Waibel - 1988
40	Super Resolution Pitch Determination of Speech Signals - Medan, Yair, et al. - 1991
36	The boston university radio news corpus - Ostendorf, Price, et al. - 1995
31	Prosodic Aids to Speech Recognition - Lea - 1980
30	The rise/fall/connection model of intonation - Taylor - 1994
25	Enhanced pitch tracking and the processing of f0 contours for computer aided intonation teaching - Bagshaw, Hiller, et al. - 1993
23	Fundamental frequency and perceived prominence of accented syllables - Terken - 1991
21	Synthesizing conversational intonation from a linguistically rich input - Taylor, Black - 1994
15	The DICEM map task corpus: Spontaneous dialogue under sleep deprivation and drug treatment - Bard, Sotillo, et al. - 1996
15	Guidelines for ToBI labelling - Beckman, Elam - 1993
11	Declination: Construct or intrinsic feature of speech pitch - Cohen, Collier, et al. - 1982
7	A dynamical system model for recognising intonation patterns - Ross, Ostendorf - 1995
6	Fundamental frequency declination in Dutch: testing three hypotheses - Gussenhoven, Rietveld - 1998
6	Prediction of prosody: An overview - Hirst - 1992
5	A model of intonational phonology for use with speech synthesis by rule - Ladd - 1987
5	Constraints on the gradient variability of pitch range - Ladd - 1994
5	Intonational Phonology. Cambridge Studies in Linguistics - LADD - 1996
5	Integrating different levels of intonation analysis - t’Hart, Collier - 1975
4	A dynamical system model for generating f0 for synthesis - Ross, Ostendorf - 1994
4	Modelling intonational structure using hidden Markov models - Wright, Taylor - 1997
2	Generating intonation contours for speech synthesis using the tilt intonation theory - Dusterhoff, Black - 1997
2	Comparison and assessment of models in the study of fundamental frequency contours of speech - Fujisaki, Ohno - 1997
2	Developments in continuous speech dictation using the 1995 ARPA NAB news task - Gauvain, Lamel, et al. - 1996
2	Declination: A review and some hypotheses. Phonlogy Yearbook 1 - Ladd - 1984
1	Predicting the intonation of discource segments from examples in dialogue speech - Black - 1997
1	Applications of intonation - an overview - Granstrom - 1997
1	An integrated pitch tracking algroithm for speech systems - Secrest, Doddington - 1993
1	Janus-II - advances in spontaneous speech recognition - Waibel, Finke, et al. - 1996