id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_k3tmdyj3vzekrjwk53daausqkm	Jose A. Belloch	On the performance of multi-GPU-based expert systems for acoustic localization involving massive microphone arrays	2015	37	.pdf	application/pdf	11031	1732	73	by using high-resolution spatial grids and a high number of microphones, accurate acoustic localization systems require high computational power. The use of GPUs for implementing sound source localization algorithms has issues and do not discuss how localization performance is affected by using different numbers of microphones or a finer spatial grid. a 2-D SRP-PHAT implementation with different Nvidia GPU architectures. is aimed at demonstrating how localization systems using a high number of microphones distributed within a room can perform sound source localization in proposed acoustic environments for real-time sound source localization are presented in Section 4, describing the experiments conducted for studying the performance of the method in a real application context. Figure 3: Computational cost when for different number of microphones M and spatial resolutions r. The real-time implementation of the SRP-PHAT algorithm uses 50% timewindow overlap, with audio sample buffers of size L. GPU-based approaches for real-time sound source localization using the SRP-590	./cache/work_k3tmdyj3vzekrjwk53daausqkm.pdf	./txt/work_k3tmdyj3vzekrjwk53daausqkm.txt