CHULA TTS: A Modularized Text-To-Speech Framework

CHULA TTS: A Modularized Text-To-Speech Framework Natthawut Kertkeidkachorn Supadaech Chanjaradwichai 1Department of Computer Engineering Faculty of Engineering, Chulalongkorn University Bangkok, Thailand 2Department of Informatics The Graduate University for Advanced Studies, Tokyo, Japan Natthawut@nii.ac.jp Proadpran Punyabukkana Department of Computer Engineering Faculty of Engineering, Chulalongkorn University Bangkok, Thailand Proadpran.p@chula.ac.th Department of Computer Engineering Faculty of Engineering, Chulalongkorn University Bangkok, Thailand Supadaech.C@student.chula.ac.th Atiwong Suchato Department of Computer Engineering Faculty of Engineering, Chulalongkorn University Bangkok, Thailand Atiwong.s@chula.ac.th Abstract Spoken and written languages evolve constantly through their everyday usages. Combining with practical expectation for automatically generating synthetic speech suitable for various domains of context, such a reason makes Text-to-Speech (TTS) systems of living languages require characteristics that allow extensible handlers for new language phenomena or customized to the nature of the domains in which TTS systems are deployed. ChulaTTS was designed and implemented with a modularized concept. Its framework lets components of typical TTS systems work together and their combinations are customized using simple human-readable configurations. Under .NET development framework, new text processing and signal synthesis components can be built while existing components can simply be wrapped in .NET dynamic-link libraries exposing expected methods governed by a predefined programming interface. A case of ChulaTTS implementation and sample applications were also discussed in this paper. 1 Introduction A Text-to-Speech (TTS) system is a system which artificially produces human speech by converting a target text into its corresponding acoustic signal. TTS systems are crucial components to many kinds of computer applications, particularly applications in assistive technology, E.g. applications for assisting the visually-impaired to access information on the Internet (Chirathivat et al. 2007), applications for automatically producing digital talking books (DTB) (Punyabukkana et al. 2012), and etc., Over the past decades, several TTS systems had been developed to fulfill applications on various computing platforms including mobile devices (Chinathimatmongkhon et al. 2008). Given specific domains, some applications of TTS systems require the systems to produce word pronunciations or generating speech signals that sound more natural to the listeners than ones generated with systems designed for texts of more general domains. For example, an application to read text from a social media web site might need a TTS system that performs a normalization of wordplays rather than attempting to pronounce them straightforwardly according to their exact spellings. While such a TTS system produced more PACLIC 28 414 Copyright 2014 by Natthawut Kertkeidkachorn, Supadaech Chanjaradwichai, Proadpran Punyabukkana, and Atiwong Suchato 28th Pacific Asia Conference on Language, Information and Computation pages 414–421 naturally-sounded speech utterances (Hirankan et al. 2014), the normalization process might degrade a TTS’s performance on a domain involving more formal texts where wordplays are scarce. For a TTS system aiming for expressive speech utterances, with multiple handlers, each of which is responsible for handling a different expression, the system could produce better results as well as easier handler development. A TTS system that allows interoperation of components, such as Grapheme-To-Phoneme (G2P) or signal generation components, deploying different speech and text processing algorithms without re-compiling of the system is obviously …

Content truncated to 500 words