Издательство Wiley, 2005, -357 pp.Методы синтеза речи: от простых до unit selection и моделирования просодии. Использование XML-языков. Способы достижения натурального звучания.I. Current Work
High-Level and Low-Level Synthesis
Low-Level Synthesisers: Current Status
Text-To-Speech
Different Low-Level Synthesisers: What Can Be Expected?
Low-Level Synthesis Potential
II. A New Direction for Speech Synthesis
A View of Naturalness
Physical Parameters and Abstract Information Channels
Variability and System Integrity
Automatic Speech Recognition
III. High-Level Control
The Need for High-Level Control
The Input to High-Level Control
Problems for Automatic Text Markup
IV. Areas for Improvement
Filling Gaps
Waveform Concatenation Systems: Naturalness and Large Databases
Unit Selection Systems
V. Markup
VoiceXML
Speech Synthesis Markup Language (SSML)
SABLE
The Need for Prosodic Markup
VI. Strengthening the High-Level Model
Speech
Basic Concepts
Underlying Basic Disciplines: Expression Studies
Labelling Expressive/Emotive Content
The Proposed Model
Types of Model
VII. Expanded Static and Dynamic Modelling
The Underlying Linguistics System
Planes for Synthesis
VIII. The Prosodic Framework, Coding and Intonation
The Phonological Prosodic Framework
Sample Code
XML Coding
Prosody: General
Phonological and Phonetic Models of Intonation
IX. Approaches to Natural-Sounding Synthesis
The General Approach
The Expression Wrapper in XML
Advantages of XML in Wrapping
Considerations in Characterising Expression/Emotion
Summary
X. Concluding Overview
