Pith. sign in

Source separation by steering pretrained music models

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.SD 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

Unleashing the Power of Natural Audio Featuring Multiple Sound Sources

cs.SD · 2025-04-24 · conditional · novelty 5.0

ClearSep iteratively decomposes naturally mixed audio into label-specific tracks, filters them by remix consistency, and uses them to train a query-based sound separation model, reporting modest SDR improvements on AudioCaps, AudioSet, and ESC-50.

citing papers explorer

Showing 1 of 1 citing paper.

  • Unleashing the Power of Natural Audio Featuring Multiple Sound Sources cs.SD · 2025-04-24 · conditional · none · ref 14

    ClearSep iteratively decomposes naturally mixed audio into label-specific tracks, filters them by remix consistency, and uses them to train a query-based sound separation model, reporting modest SDR improvements on AudioCaps, AudioSet, and ESC-50.