"In our home environment the current system can locate sound events fromtalking, footsteps, putting glasses on the table, chewing food, and clashes of silver-ware with dishes. Our location updating rate for continuous talking that faces thequads are 1-5 seconds per reading. Generally, it is especially responsive to crisp soundevents such as eating, sniffing, or putting a backpack on a table."
Better link to original PDF: <a href="https://smartech.gatech.edu/bitstream/handle/1853/3729/04-20.pdf" rel="nofollow">https://smartech.gatech.edu/bitstream/handle/1853/3729/04-20...</a>