ComfyUI-SloppyAudio

SoX and BS-RoFormer nodes for ComfyUI. SoX for sound editing, BS-RoFormer for audio stem separation.

Nodes

SloppyAudio Stem Separate

Splits audio into vocals, drums, bass, and other stems using Mini-BS-RoFormer-V2-46.8M. Model auto-downloads from HuggingFace on first run (~94 MB, stored in ComfyUI/models/sloppyaudio/). Connect only the stem outputs you need.

SloppyAudio Stem Merge

Mix up to 4 audio inputs back together with per-input gain control (dB). Auto-normalizes to prevent clipping. Handles mono/stereo and mismatched lengths.

SloppyAudio Fade

Fade-in and fade-out using SoX. Supports linear, quarter-sine, half-sine, logarithmic, and parabolic curves.

SloppyAudio Pitch

Pitch-shift in semitones using SoX. Changes pitch without altering tempo.

Install

Clone into ComfyUI/custom_nodes/:

cd ComfyUI/custom_nodes
git clone https://github.com/forcepusher/ComfyUI-SloppyAudio.git

Dependencies install automatically via ComfyUI-Manager, or manually:

pip install -r requirements.txt

SoX binaries are embedded in bin/ for Windows, macOS, and Linux — no separate SoX install required.