This book gives a concise treatment of modern speech synthesis: it comprises four parts which conduct from the fundamentals via a state-of-art display to an outlook of modern and future applications. The first introductory part contains a general discussion of the role of speech synthesis in the field of speech technology. In the second part, fundamentals are presented using the classical subdivision of text-to-speech systems into a linguistic and an acoustic part. The algorithms which are used in these parts build the fundament of speech synthesis. The third part presents advanced topics: * Embedded TTS, inventory compression with standard codecs * Databases and tools * HMM synthesis * Adaptive personal features * Prosody, models of emotions * Multilinguality, special problems with tone languages In the last part the authors describe key applications (automotive, telecommunication, etc.) and describe methods for evaluating the quality of TTS systems.
Publisher: Springer-Verlag Berlin and Heidelberg GmbH & Co. KG