- VIDEO
- AVATAR
- STATUS
- HUMAN
Short answer
How can I make an AI avatar talk from a photo?
You upload a face image and provide a script to your agent. The AI then uses D-ID to animate the face and generate a video of that person speaking. You'll get a finished video file.
The results
What you get from this task
The AI handles the heavy lifting of animation and video rendering.
VIDEO
Talking video files
The AI generates a video of the uploaded face speaking your text. You get a finished file you can download or share.
AVATAR
Custom digital personas
By using a specific photo, the AI creates a persona that looks like the subject. This allows for consistent representation.
STATUS
Processing updates
The AI tracks the video generation progress. You can check if the animation is finished or still rendering.
HUMAN
Script and photo selection
You decide which face to use and what the avatar says. The AI cannot choose the subject or the message for you.
The workflow
What your AI does when the request arrives.
Your agent acts as the bridge between your instructions and the D-ID animation engine.
Upload the face
The AI takes the image you provide and sends it to the service to establish the visual base for the avatar.
upload_imageAnimate the subject
The agent instructs the engine to combine the uploaded face with your specific text or audio input.
create_talkCheck video status
Since video rendering takes time, the AI monitors the job to see when the animation is complete.
get_talkDeliver the file
Once the video is ready, the AI retrieves the final result so you can access the talking avatar.
get_talk
Try it
Copy these to start.
Use these prompts to kick off the video creation process.
Starting points
These are opening lines. You should swap the bracketed details with your own image files and specific scripts.
D-ID Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_EsdDkqmzbnqPkxJveIE4hetjKifO1MDkdPsmTMlt/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — D-ID capabilities are ready to use.
{
"mcpServers": {
"d-id-mcp": {
"url": "https://edge.vinkius.com/vk_preview_EsdDkqmzbnqPkxJveIE4hetjKifO1MDkdPsmTMlt/mcp"
}
}
}Upload this photo [image_path] and create a talking avatar saying 'Welcome to our new product launch!'
Use the face in [image_path] to make a video where the person says 'Hello, I am your new digital assistant.'
I've uploaded [image_path]. Please generate a talking video using this script: 'The meeting has been moved to 3 PM.'
Check the status of my last video request to see if the animation is finished.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect D-ID once, then ask.
Just link your account once. Your credentials stay encrypted, and you can immediately start making videos through your AI client.
Connect D-ID to your AIFAQ
How this task behaves
- 01
Can the AI choose which photo to use?
No. You must provide the specific image you want to animate.
- 02
Can the AI edit the video after it is generated?
No. The tools only allow for creating new videos or checking the status of existing ones.
- 03
Does the AI handle the video rendering?
The AI sends the instructions to D-ID, which performs the actual rendering work.
- 04
Can I use audio files instead of text?
Yes, the AI can use audio to drive the animation of the uploaded face.
- 05
Will the AI automatically post the video to social media?
No. The AI creates the video file, but you are responsible for uploading it to any platforms.
More questions about D-ID? The Connector page answers them. See everything the D-ID Connector can do
Connect D-ID to Claude, Cursor, ChatGPT & more
