TEXT TO VOICE’
So you think you know who you are speaking to?
Virtual World users will soon be able to have their avatars speak in the voice of their choosing, by typing what they want the avatar to say.
This will result from of a “strategic” partnership announced Wednesday between leading Text-to-Speech (TTS) vendo Cepstral LLC, of Pittsburg, PA, USA, and virtual world entrpreneurial pioneerAnshe Chung Studios Ltd. (ACS), today a leading developer and distributor of virtual goods in 3D online environments.
The deal will enable Cepstral to distribute the audio into the required virtual world from its hosted (SaaS) platform at VoiceForge.com.
VoiceForge currently offers the world’s largest catalog of more than 50 unique English voices, allowing users to choose the personality their avatar speaks with. Voices range from novelty accents like Evil-Genius, Dallas, Top- Hat, Belle, and Vixen to classic young adults, American and British (pictures at right).
Initially, ACS will use the technology in the 3D virtual chat environment IMVU, and probably later into such virtual worlds as Second Life, Twinity, SMeet and Entropia Universe.
ACS has sold more than 1.2 million pieces of flash-enabled furniture in use in IMVU. Users of that furniture will be able access stre aming TTS immediately. Users who have not purchased furniture from ACS can still demo the VoiceForge TTS service provided they are in a room with flash-enabled ACS furniture. This opens up the TTS service to all 20 million current IMVU users.
“Cepstral’s VoiceForge service provides real-time surrogate voices for users,” said Guntram Graef, current CEO of ACS, founded by first virtual world, real-dollar “millionaire,” Anshe Chung (RL: Ailin Graef), and based in Wuhan, China. “Their expansive voice variety is consistent with our mission to provide users with quality digital goods and online personality. Their elegant hosted solution made it a snap to integrate with our products. In looking for a long-term partner, it was also important to us that Cepstral owns all their speech technology (demo) and drives innovation in th e speech technology field,” he said .
“ACS approached us to provide this capability to their users,” Cepstral CEO Craig Campbell said. “They recognise that in a Virtual World, many users don’t want to use their real world voice when chatting. VoiceForge allows the user to match a voice to their avatar. A proper lady avatar can use the Southern Belle voice, while a more whimsical avatar might use the French-Fry voice.”
I find that the voices, although realistic, leave something to be desired when it comes to emphasis, intonation and timing and unable to fully get emotional content across. However, with the right spelling – no typonese – and punctuation they are surprisingly good.