ProSynth: An Integrated Prosodic Approach to Device-Independent, Natural-Sounding Speech Synthesis

Richard Ogden; Sarah Hawkins; Jill House; Mark Huckvale; John Local; Paul Carter; Jana Dankovicová; Sebastian Heid

doi:10.1006/csla.2000.0141

ProSynth: An Integrated Prosodic Approach to Device-Independent, Natural-Sounding Speech Synthesis

Richard Ogden, Sarah Hawkins, Jill House, Mark Huckvale, John Local, Paul Carter, Jana Dankovicová, Sebastian Heid

Research output: Contribution to journal › Article › peer-review

Abstract

This paper outlines ProSynth, an approach to speech synthesis which takes a rich linguistic structure as central to the generation of natural-sounding speech. We start from the assumption that the acoustic richness of the speech signal reflects linguistic structural richness and underlies the percept of naturalness. Naturalness achieved by paying attention to systematic phonetic detail in the spectral, temporal and intonational domains produces a perceptually robust signal that is intelligible in adverse listening conditions. ProSynth uses syntactic and phonological parses to model the fine acoustic–phonetic detail of real speech. We present examples of our approach to modelling systematic segmental, temporal and intonational detail and show how all are integrated in the prosodic structure. Preliminary tests to evaluate the effects of modelling systematic fine spectral detail, timing, and intonation suggest that the approach increases intelligibility and naturalness.

Original language	English
Pages (from-to)	177-210
Number of pages	34
Journal	Computer Speech and Language
Volume	14
DOIs	https://doi.org/10.1006/csla.2000.0141
Publication status	Published - 2003

Access to Document

10.1006/csla.2000.0141

Cite this

@article{31ba16eefc964aeabfb8e1882284f429,

title = "ProSynth: An Integrated Prosodic Approach to Device-Independent, Natural-Sounding Speech Synthesis",

abstract = "This paper outlines ProSynth, an approach to speech synthesis which takes a rich linguistic structure as central to the generation of natural-sounding speech. We start from the assumption that the acoustic richness of the speech signal reflects linguistic structural richness and underlies the percept of naturalness. Naturalness achieved by paying attention to systematic phonetic detail in the spectral, temporal and intonational domains produces a perceptually robust signal that is intelligible in adverse listening conditions. ProSynth uses syntactic and phonological parses to model the fine acoustic–phonetic detail of real speech. We present examples of our approach to modelling systematic segmental, temporal and intonational detail and show how all are integrated in the prosodic structure. Preliminary tests to evaluate the effects of modelling systematic fine spectral detail, timing, and intonation suggest that the approach increases intelligibility and naturalness. ",

author = "Richard Ogden and Sarah Hawkins and Jill House and Mark Huckvale and John Local and Paul Carter and Jana Dankovicov{\'a} and Sebastian Heid",

year = "2003",

doi = "10.1006/csla.2000.0141",

language = "English",

volume = "14",

pages = "177--210",

journal = "Computer Speech and Language",

}

TY - JOUR

T1 - ProSynth: An Integrated Prosodic Approach to Device-Independent, Natural-Sounding Speech Synthesis

AU - Ogden, Richard

AU - Hawkins, Sarah

AU - House, Jill

AU - Huckvale, Mark

AU - Local, John

AU - Carter, Paul

AU - Dankovicová, Jana

AU - Heid, Sebastian

PY - 2003

Y1 - 2003

N2 - This paper outlines ProSynth, an approach to speech synthesis which takes a rich linguistic structure as central to the generation of natural-sounding speech. We start from the assumption that the acoustic richness of the speech signal reflects linguistic structural richness and underlies the percept of naturalness. Naturalness achieved by paying attention to systematic phonetic detail in the spectral, temporal and intonational domains produces a perceptually robust signal that is intelligible in adverse listening conditions. ProSynth uses syntactic and phonological parses to model the fine acoustic–phonetic detail of real speech. We present examples of our approach to modelling systematic segmental, temporal and intonational detail and show how all are integrated in the prosodic structure. Preliminary tests to evaluate the effects of modelling systematic fine spectral detail, timing, and intonation suggest that the approach increases intelligibility and naturalness.

AB - This paper outlines ProSynth, an approach to speech synthesis which takes a rich linguistic structure as central to the generation of natural-sounding speech. We start from the assumption that the acoustic richness of the speech signal reflects linguistic structural richness and underlies the percept of naturalness. Naturalness achieved by paying attention to systematic phonetic detail in the spectral, temporal and intonational domains produces a perceptually robust signal that is intelligible in adverse listening conditions. ProSynth uses syntactic and phonological parses to model the fine acoustic–phonetic detail of real speech. We present examples of our approach to modelling systematic segmental, temporal and intonational detail and show how all are integrated in the prosodic structure. Preliminary tests to evaluate the effects of modelling systematic fine spectral detail, timing, and intonation suggest that the approach increases intelligibility and naturalness.

U2 - 10.1006/csla.2000.0141

DO - 10.1006/csla.2000.0141

M3 - Article

VL - 14

SP - 177

EP - 210

JO - Computer Speech and Language

JF - Computer Speech and Language

ER -