TY - GEN
T1 - GPU-based fast signal processing for large amounts of snore sound data
AU - Guo, Jian
AU - Qian, Kun
AU - Xu, Huijie
AU - Janott, Christoph
AU - Schuller, Bjorn
AU - Matsuoka, Satoshi
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/12/27
Y1 - 2016/12/27
N2 - Snore sound (SnS) data has been demonstrated to carry very important information for diagnosis and evaluation of sleep related breathing disorders with high prevalence, such as Primary Snoring and Obstructive Sleep Apnea (OSA) - a serious chronic sleep disorder with a big community. With the increasing number of collected SnS data from subjects, how to handle such large amount of data is a big challenge, and a huge opportunity for further study on optimally combining signal processing techniques with machine learning algorithms. In this study, we utilize the Graphics Processing Unit (GPU) to process a large amount of SnS data collected from hospitals in Germany (37 subjects, 38.34 hours, 15.10 GB). Experimental results prove that, our GPU-based platform significantly speeds up the audio processing for features extraction of SnS data, compared with the traditional Central Processing Unit (CPU) system.
AB - Snore sound (SnS) data has been demonstrated to carry very important information for diagnosis and evaluation of sleep related breathing disorders with high prevalence, such as Primary Snoring and Obstructive Sleep Apnea (OSA) - a serious chronic sleep disorder with a big community. With the increasing number of collected SnS data from subjects, how to handle such large amount of data is a big challenge, and a huge opportunity for further study on optimally combining signal processing techniques with machine learning algorithms. In this study, we utilize the Graphics Processing Unit (GPU) to process a large amount of SnS data collected from hospitals in Germany (37 subjects, 38.34 hours, 15.10 GB). Experimental results prove that, our GPU-based platform significantly speeds up the audio processing for features extraction of SnS data, compared with the traditional Central Processing Unit (CPU) system.
UR - http://www.scopus.com/inward/record.url?scp=85010424318&partnerID=8YFLogxK
U2 - 10.1109/GCCE.2016.7800498
DO - 10.1109/GCCE.2016.7800498
M3 - Conference contribution
AN - SCOPUS:85010424318
T3 - 2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016
BT - 2016 IEEE 5th Global Conference on Consumer Electronics, GCCE 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 5th IEEE Global Conference on Consumer Electronics, GCCE 2016
Y2 - 11 October 2016 through 14 October 2016
ER -