Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Embedded

Text-to-speech software reproduces real voice

Posted: 06 Aug 2008 ?? ?Print Version ?Bookmark and Share

Keywords:text-to-speech software? VoIP? voice reproduction?

Oki Electric Industry Co. Ltd has launched the Polluxstar text-to-speech (TTS) software that enables reproduction of a user's real voice complete with that individual's unique intonations and tones. By using Polluxstar on a computer, users can communicate using their own voice, instead of a mechanical non-human voice.

"OKI has many years of experience in voice communication technology and has achieved a high market share in the VoIP and CTI markets in Japan. By expanding its technology capabilities, OKI has succeeded in offering TTS software that not only is easy to listen to, but also can express each user's personality by reproducing his or her real voice," said Yuichiro Hiranuma, president of Ubiquitous Service Platform Co. at Oki. "As a company, we aim to contribute to the realization of an e-society, full of ubiquitous services, and where users can access information in the format they desire. Thus, instead of just offering methods of communication, we are offering a rich communication environment that can also convey emotions."

A person's voice is a part of their identity and emotions and personality are important when conveying meaning. However, to reproduce natural-sounding voice using TTS software, long hours and hard work must be invested. This is because manufacturers build a database with a vast amount of voice data with properly pronounced words, and other voice data drawn from users reading predetermined sentences. In contrast, Oki has been researching and developing ways to reproduce voice by collecting a reasonable amount of voice data without the requirement that words necessarily be pronounced accurately. As a result, Oki succeeded in establishing a stable and effective voice database and was able to develop "real voice" technology at a practical level.

Real-life application
Izumi Maki, a computer science professor at Osaka University of Arts, cooperated in the process of verifying the value of this technology in a real-life situation. Prior to his surgery to remove his vocal chords from an illness, Oki recorded his voice data and installed it in Polluxstar for him to use. After six months after his surgery, Maki returned to the University to give lectures again, using his "real voice" on the software. The professor completed all lectures for the first half of the term in July.

Maki has also been actively using Polluxstar in his everyday life. He cooperated with Oki in verifying its usefulness as well as the challenges involved in practical use. As a result, it was proven that the technology was at a level viable enough to be offered to the general public for practical use, and the software was further enhanced for general use.

Hiranuma added, "We decided to offer Polluxstar to the Japan market, aiming to help and support the people and families who lose their voice due to illness or accidents. We will continue to develop TTS technology and apply it to new applications in which one's unique voice is necessary for richer voice communication."

Article Comments - Text-to-speech software reproduces r...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top