keep on record Console
<View all products
Virtual digital human
With AI driven real-time rendering as its core capability, virtual digital human provides a fast and lightweight path for building digital human. Based on the technical advantages of Dharma Institute in the fields of voice, image, 3D art, natural language processing, etc., it provides "digital streaming media" products aimed at real-time interaction, "digital human video synthesis" products aimed at efficient content production, and rich expressive digital human asset construction customization services. Buy now and enjoy the infinite possibilities brought by digital people!
rails c Product Dynamics Product documentation Contact for consultation

Core functions

Relying on the powerful AI technology capabilities of Alibaba Dharma Institute, Alibaba Cloud Virtual Digital Human provides Alibaba Cloud customers with low threshold, lightweight, and easy to integrate 3D digital human, 2D digital human real-time and offline driving capabilities, enabling Alibaba Cloud customers to easily connect their digital human capabilities to their own businesses, thus experiencing the business improvement brought by digital human technology; At the same time, it also provides a rich image library of digital human assets, and a complete digital human video creation SAAS product for the convenience of non developers to use and experience.

Video synthesis

Users can select style, digital person, voice color, background and driving style to generate digital person broadcast video, which is applicable to education, e-commerce, news, etc.

Streaming Media

Self developed cloud rendering technology enables real-time intelligent interaction of digital people and supports API and SDK calls. Applicable to video customer service, assistant, live broadcast and other scenes.

Digital star making

It includes character planning, design, 3D modeling, action expression production, drive configuration, adaptive intelligence and human drive, and multi scene and multi terminal reuse.

Functional experience

Change image
Change background
Replace sound
Female voice
Male voice
Child voice
english
Composite Text
3D digital streaming media

Ability demonstration

AI drive
Alibaba Cloud "Virtual Digital Human" uses AI real-time drive and real-time rendering to provide customers with technology development capabilities with a PaaS technology base based on AI driven digital human real-time rendering and streaming. The product includes functions such as text real-time driven streaming media, end cloud collaborative drive, digital human image management and intelligent interactive cards, and will introduce more driving modes in the future. Customers can use this set of real-time streaming media products to develop applications suitable for their own business, and derive more digital human related products or services.
Multimodal interaction
Centering on the multimodal technologies such as CV, 3D, voice, NLP, and the big dialogue model of the quaternion (knowledge, emotion, memory, personality), and based on the technical architecture of end cloud collaboration, the emotional digital human is built to support duplex interaction.

Application scenarios

Content creation
Digital employees
Video customer service
Virtual live broadcast
It can be applied to news broadcast, policy interpretation and courseware explanation
Translate text, voice and other content into video content based on digital human image, so that it can meet the needs of launching or spreading in audio-visual media such as TV, websites, video platforms or social platforms. Provide video content creation tools with lower threshold and stronger expression for content creators, improve creation efficiency and enlarge content value.
Typical adaptation scenarios
Media industry
In response to the planning of the State Administration of Radio, Film and Television, promote the application of virtual presenters in news broadcasting, weather forecasting, variety arts, science and education and other program production, innovate program forms, and improve the production and broadcasting efficiency and intelligent level.
Education industry
A large number of term explanations, courseware explanations, and wrong topic explanations make teachers overwhelmed. Through this product, teachers' workload can be reduced.
New media operations in various industries
It needs to produce videos quickly and put them on short video platforms, such as interpretation of party building policies, interpretation of company policies, analysis of the financial industry, etc.
Your exclusive digital employees
Support the combination of digital human image and intelligent assistant scene, upgrade the traditional text interaction to digital human interaction, and create a digital employee image with both cordiality and professionalism. The application of digital staff can replace real people to complete basic services such as business consultation and question answering, information broadcast, data query, business process handling, etc.
Implementation path
Through the text driven capability of the open platform for digital people, technology service providers can integrate digital people into all terminals, help enterprises to upgrade digitally, reduce costs and improve efficiency.
Our functional advantages
By giving the customer service robot the image of a digital person, the traditional online/hotline customer service is upgraded to support WebRTC or VoLTE video customer service. Support personalized configuration of customer service image and voice to make the AI service process more friendly and warm; It supports two-way interaction of text+voice+video, and can customize the service response strategy according to the user's expression, action and environmental information.
Implementation path
Through the text driven capability of the digital human open platform, technical service providers can integrate digital human into each terminal to complete the upgrade of ordinary voice customer service.
Our functional advantages
Drive digital people to play the role of "virtual anchor", intelligently generate broadcast video stream content, realize 7 * 24 non-stop live broadcast in the live studio, help brands create unique live scenes, and provide professional, intelligent, and interesting live viewing experience for the audience.
More scene adaptation
Product intelligent recommendation
Multi modal display of goods
User interaction operation
Content continues to accompany
Brand mental communication

Product purchase and opening

Digital Human Video Synthesis
The official version was released heavily
Drive 2D/3D digital human based on text or voice to synthesize corresponding video. Each new AliCloud user will be given 10 minutes of free video generation time.
Buy Now
3D digital streaming media
First choice for digital human real-time interaction
Text driven generation of 3D digital human broadcast video stream can be used to build interactive digital human with intelligent dialogue capability.
Buy Now
2D live image customization
New capabilities are hot online
Upload 3-8 min live video materials, and after training, you can customize your own 2D live image, which can be used for video synthesis products to produce broadcast videos.
Open now
3D asset customization "digital star making"
Customized service personnel docking
Asset services that include character design, 3D modeling&binding, action expression production, and drive configuration, adapt to intelligent/human driven, and multi scene and multi end reuse.
Programme consultation

Documentation and Tools

Product Introduction
View the product introduction of the virtual digital human open platform
View the product introduction of the brand intelligent live broadcast room
quick get start
View the Virtual Digital Human Open Platform Quick Start
View the quick introduction of brand intelligent live broadcast room products
User Guide
View the development guide of virtual digital human open platform
View the user guide of the brand intelligent live broadcast room
common problem
FAQ of Virtual Digital Human Open Platform
FAQs of Brand Smart Live Room