SeeVoice Real-Time 3G

for Avatar Video Call and other real-time Mobile Avatar Services

SeeVoice Real-Time server software product visualizes voice conferencing via animated 3D avatars: voice stream is automatically converted into video stream in real time. Synthetic video is generated by speech-driven avatar animation.

Next-generation synthetic video enables live visual communication, just like plain video does, but goes beyond it by leting user choose how to be seen online.

Avatars for a synthetic video service can be prepared by service provider, or created by users from their own photos; this feature is enabled by SeeStorMe add-on software.

SeeVoice Real-Time processes voice stream in real time, automatically extracts speech phonemes and converts them into avatar animation commands. The product core is LipSync, speech-to-motion animation technology, based on SeeStorm's proprietary Phoneme Recognition technology.

The input is voice stream in RAW PCM or G.711 format, and IDs of avatar, background picture and emotional state. The output is video stream in H.261/H.263/H.264/MPEG4 format. Other output formats are available upon request.