logo
RunComfy
ComfyUIPlaygroundPricing
discord logo
ComfyUI>Workflows>Sonic | Lip-Sync Portrait Animation

Sonic | Lip-Sync Portrait Animation

Workflow Name: RunComfy/Sonic
Workflow ID: 0000...1191
Updated 6/16/2025: ComfyUI version updated to v0.3.39 for improved stability and compatibility. Sonic revolutionizes portrait animation by leveraging global audio perception for smoother, more expressive facial movements. By capturing the full audio context, Sonic ensures lifelike, emotionally resonant animations that go beyond phoneme-based methods. Experience the next generation of portrait animation with Sonic.

ComfyUI Sonic redefines portrait animation by harnessing global audio perception for ultra-realistic facial movements and expressions. Unlike traditional methods, it captures the full context of speech—beyond phonemes—to generate fluid, emotionally rich animations. With cutting-edge AI technology, Sonic ensures seamless sync between voice and visuals, bringing characters to life with unmatched realism. Elevate your animations with Sonic and make every expression feel truly alive.

The ComfyUI Sonic nodes and related workflow were developed by smthemex. For more information, please visit smthemex's GitHub.

1.1 How to Use Sonic Workflow?

Sonic

Left nodes are your inputs for Audio and Avatar Image. Middle one is the Sonic Processing Node. Right side is the video combine node for outputting video.

Follow these Steps:

  1. Input your Avatar Image which will be used to visualize the dialogues from the audio.
  2. Input your Audio for generating an audio-driven voice-over of the inserted image.
  3. Click Queue Prompt!!

Done! Your rendered video will be stored in the Outputs folder.

Strengths and Weaknesses of Sonic:

Strengths:

  • Sonic generates highly realistic and expressive portrait animations driven by audio.
  • Sonic uses SVD, so there is no flickering between frames.
  • Consistency is better than previously released audio2video models.

Weaknesses:

  • As Sonic uses SVD, far or full body shots may struggle with projecting vocals on the face properly.
  • Side view faces, or faces at complex angles might give distorted results.

1.2 Sonic Audio and Video Input

Sonic

  • Upload your Audio in the load audio node (Dialogues or Vocals)
  • Upload your image in the Load image node (A close-up or medium shot of a person)

1.3 Sonic Processing Node

Sonic

ComfyUI Sonic uses SVD Model under the hood for processing, so the results and settings are according to the SVD model. These settings are set to optimum; there's no necessity to change them.

  • Keep min resolution near 768 or under if there are artifacts like morphing or distorted hands.

Sonic transforms portrait animation by focusing on global audio perception for seamless, lifelike expressions. By capturing the full depth of speech, it creates animations that feel natural, emotive, and engaging. Whether for storytelling, virtual avatars, or content creation, Sonic delivers unmatched realism. Step into the future of animation with Sonic—where every word comes to life.

Want More ComfyUI Workflows?

Vid2Vid Part 1 | Composition and Masking

The ComfyUI Vid2Vid offers two distinct workflows to creating high-quality, professional animations: Vid2Vid Part 1, which enhances your creativity by focusing on the composition and masking of your original video, and Vid2Vid Part 2, which utilizes SDXL Style Transfer to transform the style of your video to match your desired aesthetic. This page specifically covers Vid2Vid Part 1

BRIA AI RMBG 1.4 vs Segment Anything | Background Removal

BRIA AI RMBG 1.4 vs Segment Anything | Background Removal

Efficiently removes backgrounds by comparing BRIA AI's RMBG 1.4 with Segment Anything.

LivePortrait | Animate Portraits | Vid2Vid

Updated 6/16/2025: ComfyUI version updated to v0.3.39 for improved stability and compatibility. Transfer facial expressions and movements from a driving video onto a source video

Face to Many | 3D, Emoji, Pixel, Clay, Toy, Video game

utilizes LoRA models, ControlNet, and InstantID for advanced face-to-many transformations

InstantID | Portraits to Art

InstantID | Portraits to Art

InstantID accurately enhances and transforms portraits with style and aesthetic appeal.

FLUX Kontext OmniConsistency LoRA

FLUX Kontext OmniConsistency LoRA

22 unique styles, perfect consistency, clean results, all done faster.

Era3D | ComfyUI 3D Pack

Era3D | ComfyUI 3D Pack

Generate 3D content, from multi-view images to detailed meshes.

Wan2.2 Fun Camera | Cinematic Motion from Images

Turn still images into lively cinematic shots with smooth camera moves.

Follow us
  • LinkedIn
  • Facebook
  • Instagram
  • Twitter
Support
  • Discord
  • Email
  • System Status
  • Affiliate
Resources
  • Free ComfyUI Online
  • ComfyUI Guides
  • RunComfy API
  • ComfyUI Tutorials
  • ComfyUI Nodes
  • Learn More
Legal
  • Terms of Service
  • Privacy Policy
  • Cookie Policy
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Playground, enabling artists to harness the latest AI tools to create incredible art.