Dynamic parameters
Only supported controls are shown for each selected video model.
AI VIDEO STUDIO
Create AI Videos with Seedance from Text, Images, and References
Choose a supported Seedance or other AI video model and the workspace will adapt its controls, estimated credits, and supported media automatically.
Video result
Your generated video will appear here
Model parameters and estimated credits update automatically. Final validation and charging happen securely on the server.
SEEDANCE WORKFLOWS
One Seedance video generator for three starting points
Use text-to-video when you want to explore a scene from a written direction. Describe the subject, action, environment, camera movement, lighting, and mood, then choose an available Seedance model and supported output settings.
Use image-to-video when appearance and composition should start from a specific frame. Upload the reference images requested by the selected model, explain what should move or remain stable, and review the credit estimate before submitting the task.
Use multi-reference generation when a supported workflow should consider several media inputs. SongLee keeps each model's accepted inputs, duration, resolution, and credit rules visible so you can choose a suitable path without calling a provider API directly.
SongLee is an independent multi-model studio. Seedance is a supported third-party model family, and model availability may change as providers update their services.
CONFIGURATION-DRIVEN
Every control follows the selected model
Server-side credits
The server recalculates every request cost before reserving credits.
Safe task lifecycle
Credits are settled on success and released on failure or cancellation.