5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

The New Version of the ROMVOX Text-to-Speech Synthesis System Based on a Hybrid Time Domain-LPC Synthesis Technique

Attila Ferencz, Istvan Nagy, Tunde-Csilla Kovacs, Maria Ferencz, Teodora Ratiu

Software ITC, Romania

Through the years we developed several TTS systems for the Romanian language, each of them presenting some advantages and disadvantages [2]. Taking into account that waveform coding (time domain) methods assures a maximum level of intelligibility and naturalness of the synthesized speech, and that prosodic effects superimposing requires the alteration of pitch (frequency domain), we developed a hybrid time domain-LPC method, obtaining a better quality of the synthesized voice than those obtained with our former systems. This paper presents some theoretical considerations, signal processing and implementation aspects of this new synthesis method developed for the ROMVOX TTS system.

Full Paper

Bibliographic reference.  Ferencz, Attila / Nagy, Istvan / Kovacs, Tunde-Csilla / Ferencz, Maria / Ratiu, Teodora (1998): "The new version of the ROMVOX text-to-speech synthesis system based on a hybrid time domain-LPC synthesis technique", In ICSLP-1998, paper 0144.