2026
SPATIAL-CLAP: LEARNING SPATIALLY-AWARE AUDIO–TEXT EMBEDDINGS FOR MULTI-SOURCE CONDITIONS
ICASSP 2026poster
Contrastive language--audio pretraining (CLAP) has achieved remarkable success as an audio--text embedding framework, but existing approaches are limited to monaural or single-source conditions and cannot fully capture spatial information. The central challenge in modeling spatial information lies i…