Audio Batch Queue (ASA)

Audio Batch Queue (ASA) is the bridge between "a pile of audio clips" and "a list of audio clips," which sounds redundant until you realize that Combine Audio From List - the node that actually merges audio into a timeline - only speaks LIST. It takes two AUDIO inputs, audio1 and audio2, and hands you a single audio_batch list containing them in order. Feed that list to the combiner and you're assembling a finished audio track.

This is the node that makes the pack's signature workflow work: you generate a line of TTS, generate another, queue them both here, and the combiner stitches them - with or without subtitle-timestamp alignment - into one continuous narration. It also plays nice with the batch nodes in other utility packs if you need to merge several of these lists later.

How it works

The mechanism is refreshingly honest: it builds a Python list and appends each non-empty input in order. The "skip empty inputs" behavior is the one bit of real logic - if audio2 isn't connected (or is None), you get a one-item list rather than a crash. Order is preserved: audio1 lands at index 0, audio2 at index 1.

The inputs

• audio1 - first clip, goes into the queue first.
• audio2 - second clip.

Both required, though the node tolerates one being missing. The single output, audio_batch, is a LIST of AUDIO objects - exactly what Combine Audio From List wants on its audio_batch socket.

Installing it

Part of whmc76/ComfyUI-AudioSuiteAdvanced (display "AudioSuiteAdvanced"). ComfyUI Manager → search "AudioSuiteAdvanced", or manually:

cd ComfyUI/custom_nodes
git clone https://github.com/whmc76/ComfyUI-AudioSuiteAdvanced
cd ComfyUI-AudioSuiteAdvanced
pip install -r requirements.txt

Restart ComfyUI after installing. No models to download for this one.

Where people get burned

The honest limitation is right in the signature: exactly two audio inputs. There's no "+" button to add a third. If your workflow generates four TTS lines per scene, you can't just keep stacking these - you'll chain batches (batch A + batch B, then merge) or build your list elsewhere. The pack's own example workflow runs a for loop per chunk and queues two at a time into the combiner, so the intended pattern is per-scene pairs, not one giant queue.

Second, and this is the pack-wide rule that bites everyone once: the two clips should share a sample rate. The combiner downstream hard-fails on mixed rates, and MakeAudioBatch won't catch it for you - it's just making a list, not checking rates. Resample before you queue.

It's not a glamorous node, but it's the boring half of the pipeline that actually gets your TTS clips onto a timeline. Queue two, combine, repeat.