Audio Combine ROTI ⚡🅡🅞🅣🅘

This is the plainest node in the whole pack, and that's a compliment - sometimes you just need to mix two audio clips together with control over how much of each you're getting, without reaching for a full DAW. The "ROTI" in the display name is just RyanOnTheInside's own initials tacked on, presumably to avoid colliding with the dozen other packs that also ship a generic "Audio Combine" node.

How it works

Two audio inputs, two weights, one summed output. It's a weighted mix, not a true crossfade with timing offsets - both clips are combined across their full length at the ratio you set.

The inputs and outputs that matter

• audio1 and audio2 (AUDIO, required) - the two clips to combine.
• weight1 and weight2 (default 0.5 each, range 0–1) - how much of each clip contributes to the mix. These aren't automatically normalized to sum to 1, so setting both to 1.0 gives you a louder, fully-summed mix rather than an even 50/50 blend - useful to know if your output is clipping or unexpectedly loud.
• Output - a single AUDIO.

How to install it

Via ComfyUI Manager: search "RyanOnTheInside," install, restart. Manually:

cd ComfyUI/custom_nodes
git clone https://github.com/ryanontheinside/ComfyUI_RyanOnTheInside.git
cd ComfyUI_RyanOnTheInside
pip install -r requirements.txt

then restart ComfyUI. Nothing to download for this one - pure audio math.

Common issues & troubleshooting

Output clips or sounds harsh. Since weight1/weight2 aren't normalized, pushing both up near 1.0 sums two full-volume signals into one, which is a classic recipe for clipping. If your mix sounds distorted, that's almost certainly the cause - pull the weights down (0.5/0.5 is a safe, literal average) rather than looking for a bug.

Two clips of different lengths. The node doesn't document any special handling for mismatched durations, so the sane assumption - and the one worth testing before you build a bigger graph around this - is that it combines up to the shorter clip's length or pads with silence. Trim or pad your clips to match upstream if you need predictable behavior.

You actually wanted a real crossfade (fade one out while the other fades in), not a static blend. This node applies one fixed ratio for the entire clip length - it's not doing anything time-varying. If you want the mix to shift over the duration of the clips, that's a job for one of the pack's Flex audio nodes, which can modulate a parameter over time via a FEATURE input, rather than this straightforward combine node.

When this is actually the right node. Don't overthink it - if you just want to layer a music bed under a vocal track, blend two takes of the same performance, or mix in a texture/ambience layer at a fixed level, this is the whole job in one node. It's a utility, not an effect, and it's worth keeping in your back pocket precisely because it doesn't try to be more than that. Most audio graphs in this pack end up needing a plain combine step somewhere, and this is the one that does it without dragging in filters, dithering, or reactivity you didn't ask for.