ReZero: Region-Customizable Sound Extraction

We introduce region-customizable sound extraction (ReZero), a general and flexible framework for the multi-channel region-wise sound extraction (R-SE) task. R-SE task aims at extracting all active target sounds (e.g., human speech) within a specific, user-defined spatial region, which is different f...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE/ACM transactions on audio, speech, and language processing speech, and language processing, 2024, Vol.32, p.2576-2589
Hauptverfasser:	Gu, Rongzhi, Luo, Yi
Format:	Artikel
Sprache:	eng
Schlagworte:	Ablation Array signal processing Data mining Feature extraction multi-channel band-split RNN Neural networks Region-customizable sound extraction region-wise sound extraction ReZero Sound Speech enhancement Task analysis Vectors
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!