Nodes/VRGameDevGirl Video Enhancement Nodes/VRGDG Music Video Builder UI
ComfyUI Node

VRGDG Music Video Builder UI

The assembly line for a music video, audio in and scene plan out

By vrgamegirl19·Created about a year ago·Updated about 7 hours ago· 718
VRGDG Music Video Builder UI
    • project_folder
    • session_path
    • srt_path
    audio_path
    project_folder
    session_path
    srt_path

    The flagship of the VRGameDevGirl pack is the AI Video Builder - the scene-by-scene production workspace where a song becomes storyboards, and storyboards become rendered video. VRGDG Music Video Builder UI is that idea in its prototype form: a node that takes an audio file plus your project's working paths, and gives you back the folder structure and timing files the rest of the builder workflow runs on. It's the door in - most of what it hands you is "here's where everything lives."

    How it works

    Four inputs, all paths, all strings. audio_path is the song. project_folder is where the project lives, session_path a session state file within it, and srt_path the timing/lyric file (the kind your lyric extractor produces) that tells the builder when scenes start and end. The node validates and hands back project_folder, session_path, and srt_path as outputs - the same three paths, structured so downstream nodes can consume them without re-deriving where things are.

    Under the hood, this is the plumbing for the builder's project model: a project holds audio + timing segments + prompts, and the flow the README describes is audio or SRT in → wizard/storyboard planning → generate and approve scene images → render scene videos → stitch. The "approved ZImage previews" in the description are the storyboard step - you generate scene stills, review them on the visual timeline, approve or reject, and only the approved ones get rendered. That review loop is the entire point of the builder, because rendering LTX 2.3 scene videos is expensive (a first-timer's full music video took ~17 hours of generation on a 4070 Ti) and nobody wants to animate an unapproved still.

    A note on the two names: this node is the prototype UI described as building a music video from audio, timing segments, prompts, and approved previews. The README's current flagship is the newer VRGDG AI Video Builder UI, which is where the full wizard, storyboard builder, timeline, and project-save/branch/export features live. If you're on a recent pack and looking for the big workspace, that's the node - this one is the earlier, simpler iteration that still shows up in published workflows and does the same core job: get the project standing, then get out of the way.

    The inputs that matter

    • audio_path - the music file that defines the project.
    • srt_path - the timing file; without it the builder has no scene grid.
    • project_folder / session_path - where the project and its session live; leave them to the defaults in a workflow that's already set them up.

    Install

    Pack-wide: ComfyUI Manager → search vrgamedev (or add https://github.com/vrgamegirl19/comfyui-vrgamedevgirl), restart, hard-refresh. Manual: git clone https://github.com/vrgamegirl19/comfyui-vrgamedevgirl into custom_nodes, pip install -r requirements.txt; on Windows portable, pip install Cython scikit-build-core first and prefer Python 3.12. The README's Video Builder guide is the real setup manual - it covers model paths and the full workflow.

    Troubleshooting

    Empty or mismatched paths are the whole failure surface - if the node hands back nothing or the downstream stages can't find files, the four inputs don't agree on where the project actually is. The community's other recurring pain: an intro or instrumental stretch with no lyrics. That's a timing problem, not a builder problem - fix it upstream (the SRT advanced extractor's preserve_nonvocal_segments exists for exactly this) before you start approving scene previews. And save often: the README's own advice is to branch or export before major experiments.

    CategoryVRGDG/UI

    Inputs (4)

    NameTypeDefaultDescription
    audio_pathSTRING
    project_folderSTRING
    session_pathSTRING
    srt_pathSTRING

    Outputs (3)

    NameTypeDescription
    project_folderSTRING
    session_pathSTRING
    srt_pathSTRING