Video
SadTalker
Turn a portrait and an audio recording into a talking-face video.
- Image to video
Sign in to create. Review the credit cost in the studio before running.
Catalog preview for SadTalker. Review your own result before publishing.
A starting point.
Use clear speech and a face with unobstructed mouth and eyes.
Before you publish.
Check lip timing, facial expression and head movement against the recording.
Review the quote before running. A failed generation releases its reserved credits. Download the results you want to keep; stored media is retained for 30 days.
Settings and model questionsCheck the available settings and guidance when preparing your input.
Inputs and settings
Default mode requires source_image_url and driven_audio_url. The reference operation additionally requires reference_pose_video_url to supply pose motion.
expression_scale adjusts expression; still_mode reduces head movement with full preprocessing. Change preprocess and face_model_resolution separately so framing and facial detail remain easy to judge.
Inputs, all modes and APIModel IDs, current input limits and the API generation workflow.
Supported inputs
These fields come from the current studio form for sadtalker. Other modes may require different inputs.
| Input | Type | Requirement | Options and limits |
|---|---|---|---|
driven_audio_url | Text | Required | — |
source_image_url | Text | Required | — |
expression_scale | Number | Optional | ≥ 0 · ≤ 3 |
face_enhancer | Text | Optional | — |
face_model_resolution | Text | Optional | — |
pose_style | Whole number | Optional | ≥ 0 · ≤ 45 |
preprocess | Text | Optional | — |
still_mode | On/off | Optional | — |
Check the full model schema for all options and cross-field rules. Request a quote for your selected settings before generating.
Other modes and their inputs
Open the operation you need and check its inputs. Settings from one mode do not automatically apply to another.
image-to-video (reference)sadtalker/reference
Supported inputs
These fields come from the current studio form for sadtalker/reference. Other modes may require different inputs.
| Input | Type | Requirement | Options and limits |
|---|---|---|---|
driven_audio_url | Text | Required | — |
reference_pose_video_url | Text | Required | — |
source_image_url | Text | Required | — |
expression_scale | Number | Optional | ≥ 0 · ≤ 3 |
face_enhancer | Text | Optional | — |
face_model_resolution | Text | Optional | — |
pose_style | Whole number | Optional | ≥ 0 · ≤ 45 |
preprocess | Text | Optional | — |
still_mode | On/off | Optional | — |
Check the full model schema for all options and cross-field rules. Request a quote for your selected settings before generating.
Use SadTalker through the API.
Start with this model ID for the endpoint featured on this page:
sadtalker
Input requirements belong to this model ID. Check a different mode’s schema separately when switching modes.
Available modes
- image-to-video ·
sadtalker - image-to-video (reference) ·
sadtalker/reference
Check the current model schema for supported inputs and options. Other tasks may use different model IDs.
From quote to result.
Request a credit quote for your input and settings before submitting a generation. A sample output or the price of another model is not a quote for your request.
Follow the generation workflow, then poll the job for its status and result. The API and studio use the same balance.