Core Technology

Based on the research accumulation of 15 Ph.Ds in artificial intelligence technology for nearly 20 years, Mofa Technology has made major breakthroughs in computer graphics and AI core technologies. The full-stack self-developed core technology system has overcome all aspects of intelligentization. Platform virtual human and virtual content technical problems.

智能建模智能绑定

核心美术能力

智能特效

实时引擎渲染

机器人
流程自动化

云端协作流程管理

通过演员表演实时驱动
三维虚拟角色的高质量动画

基于单个RGB摄像头的三维
身体运动捕捉和合成技术

高质量实时解算技术
(毛发和布料解算)

基于单个RGB摄像头的三维人脸
表情、眼神实时捕捉和迁移技术

虚拟形象的高质量建模、
绑定、材质灯光以及渲染

基于单个RGB摄像头的
三维 AI 人脸建模技术

真人和虚拟人直播互动

Real-time generation of virtual human voice and 3D animation technology based on text (TTSA)

Using the industry’s advanced deep learning technology to automatically generate high-quality virtual human voice and animation data based on text in real time, it has realized the transformation from a simple voice assistant to an interactive virtual digital assistant, allowing customers to use the most natural Ways to have a conversation with virtual people.

Real-time generation of virtual human 3D animation technology based on real human voice input (STA)

With Mofa's Speech To Animation technology, it can drive virtual humans to synthesize mouth animations, facial expressions, body movements, etc. in real time, and the synthesized content and voice information are highly matched, so that virtual digital humans can behave like real people Vivid and lifelike, natural and smooth performance.

Emotional speech synthesis (ETTS)

ETTS (emotional text to speech) is emotional tts (speech synthesis). The general tts is more mechanical, while etts retains the tone and rhythm similar to the real person during the tts process, and it sounds more emotional.

Dialogue System (NLP)

Deal with the user's question, understand the user's semantics and intentions, and then match or generate the corresponding reply.

Smart action generation

Use in TTSA or STA, understand semantics, and then generate corresponding actions and expressions based on semantic matching or real-time

Voice Conversion Technology (VC)

The voice-based tone conversion technology can convert the sound of the initial tone into the audio of the target tone in real time through the VC algorithm developed by Mofa for output.