Nodes/ComfyUI_StarNodes/⭐ Star Video Sound Enricher Option
ComfyUI Node

⭐ Star Video Sound Enricher Option

The same sound fix, as a settings bundle — for when you don't want extra wires

By Starnodes2024·Created 2 years ago·Updated 2 days ago· 106
⭐ Star Video Sound Enricher Option
    • sound_settings
    presetCinematic Warm
    harsh_freq3500
    harsh_cut5.5
    high_cut_freq9000
    high_cut_db9.0
    bass_freq110
    bass_boost6.0
    warmth2.5
    drive0.25
    intensity1.00
    normalizetrue

    ⭐ Star Video Sound Enricher Option is the answer to a very specific question: "I want the sound-enricher fix, but I'm already using the Star LTXV All-in-One node and I don't want a whole extra audio subgraph cluttering my workflow."

    The standalone ⭐ Star Video Sound Enricher processes audio directly - audio in, audio out. This one doesn't touch audio at all. It exposes the exact same preset dropdown and the exact same custom knobs, then packages them into a SOUND_SETTINGS bundle and hands that bundle to another node. Specifically the Star LTXV 2.5 All-in-One, which accepts it and applies the same filter chain to its own audio output internally. Zero extra wires, zero double-processing.

    When to use which

    The decision is genuinely binary:

    • Your audio comes out of the LTXV 2.5 All-in-One node → use the Option node. Plug its sound_settings output into the all-in-one's settings input and the soundtrack comes out cleaned and enriched as part of the same run.
    • Your audio comes from anywhere else (a loaded video, another model, a file) → use the standalone Enricher, which takes an AUDIO input and returns an AUDIO output.

    The inputs

    Same as the standalone version, minus the audio input. preset is the main decision: Cinematic Warm (default), Smooth & Soft, Voice Clarity, Deep Bass Boost, or Custom. In Custom mode the knobs come alive - harsh_freq/harsh_cut, high_cut_freq/high_cut_db, bass_freq/bass_boost, warmth, drive - plus intensity (dry/wet, applies to every preset) and normalize (peak-normalize to -1 dBFS). Everything is documented in the tooltips, and they're identical to the standalone node, so whatever you learned there transfers.

    The single output is sound_settings, a small bundle holding the preset name, the resolved parameters, intensity, and the normalize flag.

    Installing it

    ComfyUI Manager, search Starnodes, install, restart. Manual:

    cd ComfyUI/custom_nodes
    git clone https://github.com/Starnodes2024/ComfyUI_StarNodes
    cd ComfyUI_StarNodes
    pip install -r requirements.txt
    

    It's under ⭐StarNodes/Video, right next to its sibling.

    The honest take

    This is a niche node with a narrow job: it exists so one particular all-in-one node can sound better without adding nodes. If you're not using the Star LTXV All-in-One, you probably don't want this - you want the standalone Enricher, which does the same filtering with an actual audio input and output. But if you've already committed to the all-in-one pipeline, this is the cleanest way to get a better soundtrack, and "clean" is the whole point of the all-in-one philosophy.

    Category⭐StarNodes/Video

    Inputs (11)

    NameTypeDefaultDescription
    presetCOMBOCinematic WarmTuned filter chains. 'Custom' uses the knobs below.
    harsh_freqoptINT35001000–10000Custom: center of the de-harsh cut (the scratchy region, typically 2.5-5 kHz).
    harsh_cutoptFLOAT5.50–12Custom: how much to cut at harsh_freq, in dB.
    high_cut_freqoptINT90004000–20000Custom: high-shelf corner - everything above gets tamed. Clamped below Nyquist of the input automatically.
    high_cut_dboptFLOAT9.00–18Custom: high-shelf cut amount, in dB.
    bass_freqoptINT11040–300Custom: low-shelf corner for the deep-bass boost.
    bass_boostoptFLOAT6.00–12Custom: bass boost amount, in dB.
    warmthoptFLOAT2.5-6–6Custom: low-mid bell at 300 Hz, in dB. Positive = warmer, negative = thinner.
    driveoptFLOAT0.250–1Custom: gentle tube-style saturation - adds harmonics for a richer sound. 0 = off.
    intensityoptFLOAT1.000–1Dry/wet mix of the whole effect. 100% = fully processed, 0% = original sound. Applies to every preset too.
    normalizeoptBOOLEANtruePeak-normalize the result to -1 dBFS for a consistent, clipping-free output level.

    Outputs (1)

    NameTypeDescription
    sound_settingsSOUND_SETTINGS