Output Waveform

ComfyUI generates a lot of audio these days, and "trust me, it sounds fine" is a terrible way to work. Output Waveform is the node you hang on the end of any AUDIO branch to actually check the result: it plays the audio, shows you its waveform with adaptive time ticks, and displays duration, sample rate, channel count, and peak level. Then it passes the audio through unchanged, so it slots in as a pure inspection step anywhere in the graph.

It's one of the 11 nodes in ComfyUI ALICE Lab Audio Tools, the fresh alpha pack (Apache 2.0, no pip deps, bilingual EN/JA) from ALICE Lab. The pack's philosophy is "preview everything, decide before you save," and this is the audio half of it - the video half being Preview Video.

How it works

The node uses ComfyUI's built-in audio preview (a temporary FLAC behind the scenes) for actual playback, then layers its own waveform rendering on top with a richer envelope so the preview stays readable even when you widen the node. The stats panel is the useful part: duration, sample rate, channels, and peak level let you sanity-check a clip before it goes anywhere. A clipped render shows up as a peak at 1.0; an unexpectedly short duration catches a failed trim.

Two inputs:

• audio - the AUDIO to preview.
• waveform_color - default auto. In auto mode it picks up color metadata attached to the AUDIO value, including the per-track colors the Audio Mixer stamps on its outputs. Type any CSS color to override. That's a small touch that matters in a busy graph: mix eight tracks and each keeps its own color through the preview, so you can tell them apart at a glance.

Single output: audio, exactly what went in. Unchanged, unprocessed - this is observation, not transformation.

Where you'll use it

Anywhere you want to listen before committing: after a TTS node, after Compare Audio's 1−2 difference output (hearing the residual tells you far more than staring at a number), or at the end of a Media Range trim. If you're comparing two takes, run each through its own Output Waveform and A/B them by eye before trusting any automated score.

Install

# ComfyUI Manager → search "ComfyUI ALICE Lab Audio Tools" → Install → restart
# or:
cd ComfyUI/custom_nodes
git clone https://github.com/alice-lab-dev/ComfyUI-ALICE-Lab-Audio-Tools
# restart, then Add Node → ALICE_Lab → Audio

Pack-wide requirements: ffmpeg + ffprobe on the ComfyUI process PATH (macOS Homebrew paths auto-checked) and a current ComfyUI with the AUDIO type. No model downloads, no Python dependencies.