Animate Yourself

CyberExtruder has integrated Acapela's text-to-speech engine technology into its 3-D animated face model, allowing individuals to use their own face and words to communicate in Web, mobile and chat applications.

The innovation is a spin-off of CyberExtruder's technology, in which a person's head is photographed and then turned into a 3-D model that can be animated.

CyberExtruder's facial animation program works by uploading a digital picture, after which the face is outlined and given a unique geometric shape. From this shape, a 3-D character is created and sent back to its original source through a Web push, whether it's a cell phone or a Web application. Any cell phone or laptop equipped with mobile messaging service and that is video capable can use the software.

With the addition of Acapela's text-to-speech engine, the 3-D animation can now speak in 22 different languages.

To reproduce natural-sounding voices, an acoustic database was created from 10 hours of recorded feed, which was then reduced into one-hour segments, according to Antoine Kaufferisen, vice marketing director for Acapela.

To convert the words from a text, text-to-speech transposes the message into phonetic writing, where a grammatical and syntax analysis of the sentences enables the system to define a word's pronunciation in order to give it the correct meaning, according to the company. Then, the system associates phonetic writing to the tone and duration of the pronunciation required. Finally, the speech is accomplished by selecting the best units stored in the acoustic database.

In the telecom industry Acapela's text-to-speech engine is used for a variety of products, from call units, office and information systems, carrier and operator centers and now virtual animation.

The program also has the capability to take lines from movies or select photos ranging from political figures to movie and music stars that are ready for face and vocal animation.

“You can take any wave file [from a movie] or put your face on a series of canned animations that have voice and deliver those to other cell phones,” Larry Gardner, CEO of CyberExtruder said.

While the software can be used on the Web, it is primarily aimed at cell phone users. The program, however, will not launch until Acapela and CyberExtruder are able to bring carriers on board, which Gardner expects to happen over the next couple of years.

Gardner and Kaufferisen are optimistic about the impact this software can have on the wireless community.

“People like to have their own personal touch,” Gardner added. “Now to be able to take this to a whole new level, where you can actually watch your friends articulating their words, will be a great way to communicate.”