Abstract
With continuing growth of the world's population and rapid economic development, our need to preserve the natural environment, especially our oceans, is becoming an increasing concern. For the past several decades scientists have been monitoring the oceans using a variety of sensors and tools. Passive acoustic monitoring is one of the primary methods used to investigate the behavior patterns of soniferous marine animals. Analyzing the vast amount of collected data poses an enormous challenge. This paper presents a new system designed for high speed acoustic processing called the High Performance Computer Acoustic Data Accelerator (HPC-ADA). Together with an appropriate software suite, the HPC-ADA is a powerful tool currently being used by the Bioacoustics Research Program (BRP) at the Cornell Lab of Ornithology, Cornell University. This paper provides a high level technical overview of the HPC-ADA system’s architecture, software suite, and operation of the HPC-ADA. We also summarize the projects that have successfully used the HPC-ADA system; totaling over one million hours of processed sound to date.
Original language | English |
---|---|
Title of host publication | Proc. ICML Unsupervised learning for Bioacoustics |
Pages | 1-8 |
Number of pages | 8 |
Volume | 1 |
Publication status | Published - 2014 |
Keywords
- - ocean acoustics
- big data
- data science
- high performance computing
- passive acoustic monitoring