This topic answers frequently asked questions (FAQ) about ZhiChuang Platform usage scenarios.
Digital Human
-
Do digital humans integrate with Large Language Models (LLMs)? Is separate development required?
Digital human script generation and real-time interaction are powered by the Qwen large language model. The system also supports knowledge bases and agents, and integrates with Alibaba Cloud Model Studio, so you do not need to write additional code.
-
Is free testing supported?
Yes, it is. If you have a business need, you can contact an operations specialist to activate a free trial for 2D digital human (public model) video synthesis. Customizing real human images is not currently supported for trials.
-
Do digital humans support live streaming?
Live interaction is not supported. Only offline videos are supported.
-
Is API integration supported?
OpenAPI integration is currently available.
For more information about digital human offline video synthesis, see Digital Human Offline Synthesis OpenAPI.
Real-time digital human interaction. For more information, see the documentation: Digital Human Real-time Interaction OpenAPI.
-
How long does it take to synthesize a video?
The video synthesis speed is approximately 1:2. This means that it takes about 2 minutes to synthesize a 1-minute video.
-
Which languages do digital humans support?
2D digital humans currently support Chinese and English, and more languages are continuously being added. 3D digital humans support Chinese, English, Thai, Japanese, Vietnamese, Indonesian, Portuguese, Spanish, Korean, and Cantonese.
-
Do 2D/3D interactive digital humans support monthly subscriptions or pay-as-you-go billing?
No, they do not. Only annual subscription packages per ingest endpoint are currently available. These packages provide dedicated resources to ensure a high-quality user experience.
-
What are the hardware requirements for 3D interactive digital humans?
There are no specific hardware requirements because the solution is cloud-based. A stable internet connection is sufficient. If you need interactive voice response, we recommend using a microphone with denoising capabilities.
-
What is the approximate latency for 3D interactive digital humans?
The latency is typically within 3 seconds.
Marketing Graphics
-
Is free testing supported?
Business customers can contact an operations representative to obtain a test quota for marketing image-and-text content.
-
Is API integration supported?
You can integrate with OpenAPI. For more information, see Text-to-Image Generation OpenAPI.