Create Talking Avatar Using WAN 2.2 Sound To Video Model
5.0
0 reviews
Description
this workflow allows you how to use Wan2.2 sound to video model to create talking avatar video based on an image and audio file. The workflow is dedicated for Low Vram Graphic card pc
all you need is
- Input image of person
- Sound file in mp3 or wav format
- Selecting the resolution
- Add the necessary model & click run
Video tutorial link
Discussion
(No comments yet)
Loading...
Reviews
No reviews yet
Versions (1)
- latest (11 days ago)
Node Details
Primitive Nodes (85)
866cdb67-cb97-4a54-8738-e9ece4f5f25f (1)
92cfa20b-adb6-4f00-8596-f14b534cf926 (1)
AudioEncoderEncode (1)
AudioEncoderLoader (1)
CreateVideo (2)
Fast Groups Bypasser (rgthree) (1)
GetNode (47)
LatentConcat (1)
LoaderGGUF (1)
MarkdownNote (3)
ModelSamplingSD3 (1)
Note (2)
PrimitiveFloat (1)
PrimitiveInt (3)
SaveVideo (2)
SetNode (15)
WanSoundImageToVideo (1)
ad3f1d6b-f6de-4f22-abcf-caeed4bf8ce4 (1)
Custom Nodes (12)
ComfyUI
- CLIPLoader (1)
- CLIPTextEncode (2)
- ImageFromBatch (1)
- VAEDecode (2)
- ImageScale (1)
- VAELoader (1)
- LoraLoaderModelOnly (1)
- LoadImage (1)
- KSampler (1)
- LoadAudio (1)
Model Details
Checkpoints (0)
LoRAs (1)
WAN2.2 T2V high_noise_model.safetensors