Skip to main content
POST
Create Scenes Video
Try it out! Use the API playground on the right to test the Scenes endpoint directly.

Overview

Scenes allows you to create complex, multi-scene videos by combining different content types in a storyboard format. Each scene can be one of four types:
  • Avatar: A talking avatar scene with script and voice
  • Video: A video or image clip with optional voiceover
  • Picture in Picture: An avatar overlaid on a video/image background
  • Start/End Frame: AI-generated video transition between two images
This format is ideal for:
  • Long-form content with multiple segments
  • Tutorials with avatar explanations and demo clips
  • Marketing videos with varied visual styles
  • Storytelling with scene transitions
  • Educational content mixing talking heads and visuals
Scenes automatically handles transitions between different scene types and maintains consistent captions throughout the video.

Endpoint


Required Fields

array
required
Array of scene objects (1-50 scenes). Each scene defines a segment of your video.

Optional Fields

string
Video name (max 100 characters). If not provided, a name will be auto-generated from the scene scripts.
string
Background music ID from /v1/music/list.
object
Caption settings for the video
string
default:"ratio_9_16"
Video aspect ratio:
  • ratio_9_16: Vertical (TikTok, Reels, Shorts) - Recommended
  • ratio_16_9: Horizontal (YouTube)
  • ratio_1_1: Square (Instagram)
string
default:"en"
Language code for the video (2 characters). Example: en, es
string
HTTPS URL to receive completion notification (max 500 characters). Highly recommended for production use.
object
Custom metadata object (max 5KB). Store any additional data you need to associate with this video.

Scene Types

Avatar Scene

A talking avatar presents content with synchronized lip movements.
Required fields: script, avatarId, voiceId

Video Scene

A video or image clip plays, optionally with voiceover.
Required fields: mediaId

Picture in Picture Scene

An avatar appears overlaid on a video/image background.
Required fields: script, avatarId, voiceId, mediaId

Start/End Frame Scene

AI generates a video transition between two images.
Required fields: startFrameMediaId, endFrameMediaId, framePrompt

Request Examples

Basic Multi-Scene Video

Tutorial with Picture in Picture

AI Transition Video

Response


Webhook Notification

When your video is ready, we’ll POST to your webhook URL:
Webhook Payload
Your webhook endpoint must return a 200 status code. We’ll retry up to 3 times if the request fails.

Scene Type Requirements


AI Models for Start/End Frame


Caption Presets

Available caption presets for the caption.preset field:

Best Practices

Plan Your Storyboard

Outline your scenes before creating. Each scene should serve a clear purpose in the narrative.

Consistent Avatar

Use the same avatar across scenes for continuity, unless intentionally switching perspectives.

Vary Scene Types

Mix avatar, video, and PiP scenes to keep viewers engaged and add visual variety.

Smooth Transitions

Write scripts that flow naturally between scenes. End one scene leading into the next.

Optimize Duration

Keep individual scenes focused. Split long content into multiple shorter scenes.

Use Start/End Frames

Leverage AI transitions for impressive visual effects between static images.

Error Handling


Credits

Scenes videos use a credit model based on:
  • Base credits: Fixed cost per video
  • Avatar scene credits: Charged based on script duration (similar to UGC Video)
  • Video scenes: Included in base credits (no additional charge)
  • Start/End Frame: Additional credits for AI generation

Next Steps

List Avatars

Browse available avatars for your scenes

List Voices

Find the perfect voice for your avatars

List Music

Add background music to your video

Webhooks Guide

Learn how to handle webhook notifications

Authorizations

x-api-key
string
header
required

Body

application/json
scenes
object[]
required

Array of scene objects (1-50 scenes)

Required array length: 1 - 50 elements
name
string

Video name (max 100 characters)

Maximum string length: 100
musicId
string

Background music ID from /v1/music/list

Maximum string length: 30
aspectRatio
enum<string>
default:ratio_9_16

Video aspect ratio

Available options:
ratio_9_16,
ratio_16_9,
ratio_1_1
caption
object
addStickers
boolean
default:false

Add AI-generated stickers to enhance video engagement

language
string
default:en

Language code (2 characters)

Required string length: 2
webhook
string

HTTPS URL to receive completion notification

Maximum string length: 500
metadata
object

Custom metadata object (max 5KB)

Response

Scenes project created successfully

success
boolean
projectId
string
message
string