IAMCCS Dialogue AudioBoard Bridge

In the IAMCCS cine ecosystem, dialogue starts life as a JSON description - speakers, lines, timings - and only becomes actual audio once it reaches the AudioBoard. The Dialogue AudioBoard Bridge is the connector between those two worlds. You hand it the cine_linx that came out of IAMCCS_DialogueTagEditor (or its script planner cousin), and it translates that metadata into AudioBoard-compatible audio lanes, ready for whatever comes next.

It's a small node with a narrow job, but it's the difference between your dialogue data being a pretty plan and being something the audio board can actually edit, stem, and mix.

How it works

It reads the dialogue payload stored inside the cine_linx (the dialogue tag editor writes an audio_board_template into the linx's resources), then re-derives a set of audio lanes according to lane_mode:

• speaker_stems - one lane per speaker, the default and the most useful for mixing/balancing.
• per_line_clips - a separate lane/clip per line, best when you want fine per-line editing.
• single_dialogue_track - everything flattened onto one track, the simplest possible output.

write_policy decides what happens to the result: audio_lanes_to_cine_linx writes the lanes back into the linx so downstream AudioBoard nodes can consume them, while metadata_only just validates and records the plan without materializing lanes - handy when you want to check the structure before committing.

There's also a frame_rate input to keep the lane timings aligned with your video's fps. The only output is an updated cine_linx, so the bridge is a pure transform in the pack's data flow: dialogue metadata in, audio lanes out, everything else untouched.

Inputs that matter

• cine_linx - the required input from DialogueTagEditor (or DialogueScriptPlanner).
• lane_mode - the one you'll actually switch between. Default speaker_stems is right for most dialogue scenes.
• write_policy - audio_lanes_to_cine_linx unless you're just checking your work.

Install

Ships with IAMCCS-nodes:

cd ComfyUI/custom_nodes
git clone https://github.com/IAMCCS/IAMCCS-nodes.git

Or ComfyUI Manager → "IAMCCS" → install → restart. No extra pip packages; it's pure metadata plumbing on top of the pack's own linx type.

Gotchas

The bridge is picky about what it accepts upstream. If the cine_linx you feed it doesn't carry dialogue data (an audio_board_template or the raw dialogue JSON in its resources), it falls back to whatever it can find and may produce an empty plan - so always chain it directly after the dialogue editor. And metadata_only doesn't write lanes, so don't set it and then wonder why AudioBoard is empty. This is orchestration glue for the pack's bigger workflows; on its own it does nothing visible.