AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

USE: A Unified Model for Universal Sound Separation and Extraction

Hongyu Wang, Chenda Li, Xin Zhou, Shuai Wang, Yanmin Qian

AAAI 2026 · Vision-language and multimodal › Cross-modal learning · approach: Audio-visual speech separation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train on 2Mix and 3Mix with EDA (stage 1) instead of fixed 2Mix baseline (ablation)Sepformer / BSRNN · AudioSet 2Mix SeenSNRi: 7.4 -> 8.7 (Sepformer); 8.6 -> 9.2 (BSRNN)Improved
add joint TSE training (stage 2) to separation training (ablation)USE-S · AudioSet 3Mix UnseenSNRi: 5.2 -> 6.3Improved
add joint TSE training (stage 2) to separation training (ablation)USE-B · AudioSet 3Mix UnseenSNRi: 4.8 -> 4.8No clear effect
add joint TSE training (stage 2) to separation training (ablation)USE-B · AudioSet 2Mix SeenSNRi: 9.2 -> 9.4Improved
joint SS+TSE training instead of TSE-only training (ablation)USE-B · AudioSet 2Mix Seen TSESNRi: 8.4 -> 8.9Improved
joint SS+TSE training instead of TSE-only training (ablation)USE-S · AudioSet 2Mix Seen TSESNRi: 8.5 -> 8.5No clear effect
joint SS+TSE training instead of TSE-only training (ablation)USE-B · AudioSet 3Mix Unseen TSESNRi: 4.8 -> 5.0No clear effect
use all three clue modalities (text+tag+video) instead of a single clue (ablation)USE-S · AudioSet 2Mix Seen TSESNRi: 6.8 -> 8.5Improved
predict source count with EDA instead of fixed known count (ablation)USE-B · Multi-dataset separationSNRi: AudioSet Seen 3Mix 6.7 -> 6.3; 2Mix 8.9 -> 9.2Mixed
use USE-B instead of TDCN++FUSS dry 4Mix · FUSS drySI-SNRi: 7.4 -> 11.9Improved
MSE plus InfoNCE alignment loss versus MSE only or InfoNCE only (ablation)USE · AudioSetalignment accuracy / SI-SNRiImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy