COMPUTER SCIENCE TTS DEVELOPMENT ENVIRONMENT RESULTS

Arpad Zsolt Bodo, Gavril Toderean, Ovidiu Buza
unpublished
The development of a text-to-speech synthesis system is a complex process that has to be elaborated in several fields like: speech signal processing, prosody prediction-generation, synthesis techniques/engines, database construction, etc. During research and development phase, each of these domains requires a set of tools. The need for these tools is multiple: sometimes from the complete TTS process flow, intermediate results are required to be visualised, sometimes external (manual)
more » ... to the system helps the researcher's work. The current paper shows a collection of these self-developed tools, which finally have formed a development environment. However this framework is built around a general voice signal editor freeware software, all the TTS specific tools and the graphical user interfaces, are the result of an own development. The main focuses of this paper is to present the development environment of PSOLA and other synthesis techniques, the diphones database processing environment, the speech signal processing toolset, prosody modification related issues and intonation curve computation algorithm.
fatcat:wdnoinsrujbe5anya5y5bgpu5y