Ectaco Development Center Releases New English Speech Synthesis System
NEW YORK, NY - Ectaco Development Center (EDC), an arm of Ectaco, Inc., a developer and manufacturer of linguistic handheld devices, has created an English-language text-to-speech (TTS) speech synthesis system. Due to its low resource requirements, it may be integrated into virtually any handheld device. Contemporary systems that use the concatenate model require storage space for a database of sounds, the consecutive reproduction of which produces the effect of speech synthesis. This approach reduces the flexibility of the system as a whole with regard to changes in intonation and timbre. The technology developed at EDC is based on parametric (formant) models, analogous to the way in which the human vocal tract modulates sound. This principle gives the system a small footprint, which enables it to be used in handheld devices. The system also has the capability to produce male, female, child and other types of voices by modulating timbres, speeds, tonalities of pronunciation, etc. "Many people say that the parametric model makes the voice synthesized by the system robot-like," commented EDC Strategic Research Manager Vyacheslav Baryshnikov, "However, this opinion was formulated due to the imprecision of the separate speech sounds in the actualization of the earliest systems, as well as the labor-intensiveness of actualizing mathematical algorithms that provide all possible permutations of speech sound modulation. We have managed to avoid these pitfalls." EDC General Manager Anton Epifanov added, "Taking into account that the human vocal apparatus is virtually identical in everyone, we have managed to emulate it using an unique mathematical apparatus. To make a speech synthesis system for other world languages like Russian or Spanish will now take considerably less effort than when using concatenate synthesis."