Avatar Videos

Use the POST /v1/avatar-videos to generate AI avatar videos using AI generated profiles or direct media URLs. To generate an avatar video, structure your JSON payload based on your desired input combination. The endpoint accepts flexible set of configurations for the avatar source and audio generation.

1. Base video model

Specify the AI model to use for avatar video generation.

• model: The ID of the AI model to use. (e.g., "zoice/avatar-x")

{
	"model": "zoice/avatar-x"
}

2. Avatar source (choose one)

• Option A: Pre-configured avatar

Provide the specific avatar_id and one of its associated image_id from your collection.

{
  "avatar_id": "avatar_12345",
  "image_id": "img_67890"
}

• Option B: Custom Image URL

Provide a direct link to a custom image in the form of input_image_url.

{
  "input_image_url": "https://example.com/images/avatar.png"
}

3. Audio Source (Choose One)

• Option A: Text-to-Speech (Voice Profile & Model)

Provide a script with script_text, a voice profile ID with voice_id, and an optional voice synthesis model with voice_model.

{
  "voice_id": "voice_prof_001",
  "voice_model": "en-US-neural-v2",
  "script_text": "Welcome to our platform! Today, we are excited to launch our new features."
}

• Option B: Direct Audio File

Provide a direct link to a pre-recorded audio file in the form of input_voice_url. Add optional enhance_audio as true for cleaning audio.

{
  "input_voice_url": "https://example.com/audio/speech.mp3",
  "enhance_audio": true
}

4. Action and motion

Direct the movement, facial expressions, or gestures of the avatar with action_prompt.

{
  "action_prompt": "Nod gently while speaking and maintain a friendly smile."
}

5. Video formatting

Define the target aspect ratio and output quality in the form of aspect_ratio and resolution respectively.

{
  "aspect_ratio": "16:9",
  "resolution": "1080p"
}

For detailed information regarding the body parameters, request and response format and schemas, please visit the Avatar Videos API documentation page.