Generate lip-synced cinematic videos from image, video, and audio references.
npx skills add https://github.com/doany-ai/skills --skill seedance-v2-doany-ai
Please help me install this Agent Skill. Skill: seedance-v2 Source: https://github.com/doany-ai/skills/tree/main/seedance-v2 Command: npx skills add https://github.com/doany-ai/skills --skill seedance-v2-doany-ai
Create cinematic short-form videos that preserve identity across multiple media references and deliver native lip-synced audio without manual editing.
Provide a prompt and optional media URLs to generate a lip-synced cinematic video using Seedance 2.0 Pro.
High-intent search queries and answers about installing and using this skill.
To generate a lip-synced cinematic video, provide a prompt along with up to 9 image references, 3 video references, and 3 audio clips. The tool outputs a 4–15 second 720p video with synchronized native speech.
Yes, you can use up to 3 audio references to drive natural-sounding speech synchronization. The generation process applies in-pass lip-sync to align the visual output with the provided audio inputs.
Multi-modal video generation is built for marketing, branding, and multi-language narratives. It preserves identity across image, video, and audio references while delivering native lip-synced audio without manual editing.
Short-form cinematic videos generated from multi-modal references are output at 720p resolution with a duration of 4 to 15 seconds. This concise schema ensures repeatable, deterministic results for short scenes.
No manual video editing is needed to achieve lip-sync with multi-modal references. The tool applies in-pass lip-sync during generation, automatically aligning natural-sounding speech to the visual output.
You should switch to other RunComfy models like HappyHorse, Wan, or Kling when you need routing flexibility for different video outputs. Seedance 2.0 Pro handles 4–15s lip-synced scenes, but alternative models may suit other specific generation requirements.
jongwonyGenerate high-fidelity videos via Veo 3.1 API with text, image, or video input.
rediumvexGenerate Seedance 2.0 prompts for SaaS product videos with camera, lighting, and sound instructions.
doany-aiRoute inputs to avatar-generation models via the RunComfy CLI.
dennisonbertramOrchestrate Nano Banana base images and Kling AI to generate realistic talking-head video clips.
