Wan 2.2 speech-driven video
Generate a speech-driven video from a reference image, a speech recording and a scene prompt. Separated vocals condition generation; the trimmed original recording supplies the soundtrack. Additional outputs preview the prepared reference, separated vocals and instruments.
Loading this version’s controls and setup…
Workflow details
Workflow details
Laufzeitprofil
- Verleger
- Sweet Tea Automation
- Status
- Aufgelistet
- Version
- v0.1.1
- Aktualisiert
- 2. Nov. 2026
Hover nodes for Pipe controls
Core
Primary Pipe surface
Reference image
LoadImage
Speech audio
LoadAudio
Positive prompt
CLIPTextEncode
Output size
ImageResize+
Extended
Additional surfaced controls
Negative prompt
CLIPTextEncode
Generation length
PrimitiveInt
Playback
PrimitiveFloat
Sampling
PrimitiveInt
Prompt guidance
PrimitiveFloat
Sampler
Sampler Selector
Scheduler
Scheduler Selector
Initial segment
KSampler
Continuation
KSampler
Generation batch
WanSoundImageToVideo
Lightning HIGH LoRA
LoraLoaderModelOnly
Lightning LOW LoRA
LoraLoaderModelOnly
Lightx2v LoRA
LoraLoaderModelOnly
Sampling shift
ModelSamplingSD3
Negative-attention guidance
WanVideoNAG
Starting frame offset
PrimitiveInt
What you need
- Model files
- 1 GiB known subtotal
Remaining hardware or download-size estimates are not available yet.
Outputs
- Speech-driven video · image
- Prepared reference image · image
- Separated vocals · audio
- Separated instruments · audio
Anforderungen
Modelle
6MelBandRoformer_fp16.safetensors
GefundenCheckpoint · 435 MB · SAFETENSORS · Unknown
- Nicht aufgelöst
Text encoder · Unknown
WAN_2_1_VAE_bf16.safetensors
Nicht aufgelöstVAE · Unknown
wan2.2_t2v_lightx2v_4steps_lora_v1.1_high_noise.safetensors
Nicht aufgelöstLoRA · Hugging Face · Comfy-Org/Wan_2.2_ComfyUI_Repackaged
wan2.2-s2v-14b-q5_0.gguf
Nicht aufgelöstDiffusion model · Unknown
Checkpoint · 602 MB · SAFETENSORS · FP16 (unverified) · Unknown
Knotenpakete
9comfyui_essentials
GefundenKnotenpaket · Registry
comfyui-kjnodes
GefundenKnotenpaket · Registry
- Gefunden
Knotenpaket · GitHub · kijai/ComfyUI-MelBandRoFormer
comfyui-multigpu
GefundenKnotenpaket · Registry
image-saver
MöglichKnotenpaket · Registry
soundflow
Nicht aufgelöstKnotenpaket · Registry
SweetTeaPreviewImage
GefundenKnotenpaket · Unknown
SweetTeaPreviewVideoFromImages
GefundenKnotenpaket · Unknown
- Gefunden
Knotenpaket · GitHub · tea-time-labs/sweet-tea-nodes
Quellen
1 Quelle
Quellen
1 QuelleModellartefakte
1 Artefakt
Modellartefakte
1 Artefakt6ca3cb27-48cc-496d-92fd-6df66b87f0f5-catalog-workflow-3c166172f6667466ff83c9103b34a138948c0913055090b8d14127f2f9af32d-0.1.1.tea
tea · 53,5 KB · SHA-256 b05952427183…a147
Download file