The Artlist AI Toolkit enables you to create AI-powered avatar videos. You can generate realistic talking avatars from images, synchronize speech with facial movement, and localize videos into multiple languages while preserving natural timing and expression.
This guide explains how to use the three main workflows:
- Avatar video generation
- Lip-sync animation
- Video dubbing and translation
How to create an avatar video
An avatar is a digitally generated on-screen presenter or character that can speak, move, and interact like a real person in a video. Our Avatar models allow you to generate a realistic talking avatar from a single image and audio file.
Avatars are ideal for:
- Presenter videos
- Marketing content
- Explainer videos
- Training materials
- Social media content
- Product introductions
Select an Avatar AI model
- From the top panel on the Homepage click AI Toolkit.
- Click Video from the dropdown menu on the bottom of the prompt field.
- Select the desired Avatar model.
Note: Avatar models are currently not available with the AI Agent.
Upload an image
Upload a clear image of the person or character you want to animate.
For best results:
- Use a front-facing image
- Make sure the face is clearly visible
- Use high-resolution images when possible
- Avoid blurry or heavily cropped photos
- Keep lighting natural and even
Supported image types include JPG and PNG formats.
Add audio
Upload your audio file. You can choose from one of your voiceovers or upload a file from your computer.
Note: An audio file must be uploaded. Scripts are not supported in the prompt field.
💡Pro Tip: Make sure to perfect your voice before using it for an avatar to ensure it matches your character.
Customize video settings
Available settings include:
- Aspect ratio
- Resolution
- Video duration
Add a prompt
Describe your character’s expressions and gestures.
Note: This is optional and model dependent.
Prompt box without prompt field
Prompt box with optional prompt field
Generate the avatar video
- Review your settings.
- Click Generate. The number of credits required to generate the video will be shown when you hover over the button.
Once completed, your avatar video will appear inside the session.
How to create a lip-sync video
The lip-sync AI models let you animate a face or existing visual so the mouth movements match a provided audio track or voiceover.
Lip-sync is useful for:
- Character animation
- Social media edits
- AI presenters
- Meme content
Select a lip-sync AI model
- From the top panel on the Homepage click AI Toolkit.
- Click Video from the dropdown menu on the bottom of the prompt field.
- Select the desired lip-sync model.
Note: Lip-sync models are currently not available with the AI Agent.
Upload your media
Upload:
- A a video containing a visible face
- An audio file or generated voiceover
Note: An audio file or generated voiceover must be uploaded. Scripts are not supported in the prompt field.
For best results:
- Keep the face clearly visible
- Use high-quality audio
- Avoid overlapping voices and loud background noise
Customize video settings
Available settings include:
- Aspect ratio
- Resolution
- Video duration
Generate the lip-sync video
- Review your uploaded media and settings.
- Click Generate. The number of credits required to generate the video will be shown when you hover over the button.
- Wait for processing to complete.
The generated lip-sync video will appear in your session.
How to create a dubbed video
Translate existing videos into multiple languages while preserving natural lip sync and voice timing.
Dubbed videos are ideal for:
- International marketing
- Educational content
- Creator localization
- Business presentations
- Training materials
Select a dubbing AI model
- From the top panel on the Homepage click AI Toolkit.
- Click Video from the dropdown menu on the bottom of the prompt field.
3. Select the desired dubbing model.
Note: Dubbing models are currently not available with the AI Agent.
Upload your source video
Upload the original video you want to dub or translate.
For best results:
- Use videos with clear speech
- Minimize background noise
- Keep the speaker visible on screen
- Use high-quality audio recordings
Note: The original video reference can be in any language. It is not limited to English.
Choose the target language
Select:
- The target language for translation. There are 50+ languages to choose from.
Customize video settings
Available settings include:
- Aspect ratio
- Resolution
- Video duration
Generate the dubbed video
- Review your settings
- Click Generate. The number of credits required to generate the video will be shown when you hover over the button.
When finished, the dubbed video will appear in your session workspace.
Tips for better avatar, lip-sync, and dubbing videos
To improve output quality:
- Use clear, high-resolution media
- Keep scripts concise and natural
- Avoid fast speech and complicated phrasing
- Use good lighting in uploaded videos
- Reduce background noise for translation workflows
- Match aspect ratio to your publishing platform
Need some more help? Head back to our Help Center.