Qimiaoyuan – AI Digital Human Video & Livestream Platform by Mobvoi 195

奇妙元 (Qimiao Yuan) is a one-stop digital human video creation and livestreaming platform launched by Nanjing Xiaowen Intelligence under Mobvoi. It streamlines the entire workflow from AI writing, AI painting, and AI voiceover to digital human video generation, offering multiple digital avatar solutions such as "photo-driven," "2.5D real-person clone," and "3D customization & IP activation." Users can complete digital human video and livestream production in a one-stop editor without appearing on camera.

Core Features

Digital Human Video Creation

• Layered Video: Enter a script or upload audio, and generate a digital human talking-head video in minutes

• Editing Video: Professional-grade track editing with support for transitions, filters, and effects

• Photo Video: Make a photo speak with just one image, supporting multi-person dialogue

• PPT Video: Upload a PPT and the system will automatically analyze the document and generate narration scripts

• One-click Video Translation: Replace the voice while imitating the character's lip movements

Digital Human Livestreaming

• 7×24 unattended livestream room

• Multilingual digital human avatars, compatible with overseas platforms such as TikTok

• 3D virtual IPs with support for face sculpting, costume styling, and interactive gameplay

Interactive Digital Human

• Powered by large models, customizable into roles such as digital customer service agents, receptionists, and narrators

Cloning & Customization

• Avatar Cloning: Record a 5-minute video to achieve 1:1 reproduction of a real person, with lip movements close to those of a real person

• Voice Cloning: Multilingual cloning and emotional synthesis, supporting cross-language and cross-age transfer

• 3D Digital Human Customization: Cartoon and realistic styles both available

AI-Assisted Creation

• AI Copywriting Assistant: 10 writing styles, supporting Chinese and English

• AI Image Generation: 8 styles to choose from

• Video Script Extraction: Automatically extract text content from videos

Why Choose 奇妙元

The biggest advantage is its full closed-loop workflow—from writing, painting, and voiceover to video production, all in one place, with no need to switch back and forth between multiple tools. It has a low operational threshold: just enter a script or upload a PPT to automatically generate a video, making it ideal for users who are pressed for time and have no editing experience. The digital human avatar library has 200+ options, including both real-person and 3D avatars, and also supports online face sculpting customization. The cloning solutions cover everything from photo-driven and real-person cloning to 3D customization, meeting the needs of both individuals and enterprises. Multilingual support is strong, making it suitable for overseas marketing. The pricing model is flexible, with a free version available for trial and paid versions supporting long videos and unlimited use.

Application Scenarios

• Education & Training: Digital humans produce courseware, training videos, and course explanations

• Corporate Promotion: Customized brand IP images and product introduction videos

• Self-Media Creation: Faceless creators batch-generate commentary videos

• E-commerce Livestreaming: Round-the-clock product demonstrations and interactive livestreams

• Government Services: AI digital receptionists provide 24/7 consultations

• Finance Industry: VIP notifications, product promotions, and knowledge base explanations

• Tourism Industry: Virtual tour guides and scenic area introductions

• Entertainment Livestreaming: 3D virtual anchor livestreams

• Phone Customer Service: Interactive digital humans boost service efficiency

How to Use 奇妙元

Four Steps

1. Create a project, choose a template or start from scratch

2. Select a digital human and enter the text

3. Edit the background and add background music

4. Export the video

Detailed Workflow

1. Register and log in to weta365.com, then click "Create Video"

2. Pick a digital human avatar (the platform offers globalized models to choose from); for content import, you can choose text-to-video or PPT/image-to-video

3. Add background music, adjust subtitles (auto-generated subtitles can be manually edited), preview, and export as MP4, with support for custom resolution and bitrate

Frequently Asked Questions

Q: How long of a video can each plan produce?

The free version supports 30 seconds; the monthly plan at 298 yuan/month supports 30 minutes; the annual plan at 2,499 yuan/year supports 360 minutes. Livestream memberships range from 1,500 to 5,000 minutes/year.

Q: How much material is needed to clone a digital counterpart?

Record about 10 minutes of real-person video for training, and the avatar can be used for a lifetime after completion.

Q: Which languages and voices are supported?

Chinese, English, Japanese, Korean and other languages; voice cloning offers 600+ voices in 16 languages, with support for cross-language transfer.

Q: Can I make a talking video from a photo?

Yes, the "Photo Video" feature lets you generate one by uploading just one image.

Q: What's included in the free version?

30-second video synthesis, some free digital humans, voices, and templates; the 奇妙 Assistant is available for a limited 3-day trial, with 200 characters and 10 images.

Q: Is there a livestream plan?

Yes. The Basic version is 9,999 yuan/year, the Advanced version is 4,999 yuan/month, and the Real-Person-Driven version is 49,999 yuan/year, supporting 60+ digital humans, 500+ voices, and 1080P resolution.

Q: Do I need editing skills?

No. Simply enter a script or upload a PPT to automatically generate a video, while advanced users can use professional editing tools.

奇妙元 vs. Competitors