An important part of the study of whaling killings in the southern part is finding them and quickly alert experts to send boats out to collect stomach samples or victim scraps to better understand what whales they eat.
Hydrophones, underwater microphones used to detect whales, are particularly useful at night or in bad weather when locating networks are ineffective. Computer algorithms play a growing role in hydrophone audio data analysis, but human listeners can complement and enhance these algorithms.
A research program known as Orcasound has created an online application that will allow scientists to hear hydrophone headphones near the San Juan Island to identify killer whales and other new sounds.
Scott Veirs, a Seattle biochemist and lead investigator of the Orcasound project, will describe the new web application and the value of civil science at the 176th American Audiovisual Society Meeting, held in partnership with Canada's Canadian Consumers Audiovisual Week 2018, 5-9 November at the Victoria Conference Center in Victoria, Canada.
Civilian scientists were useful in detecting whales and observing unusual activity, such as the presence of other animals or shipping noise. The goal of Orcasound is to offer a cheap and user-friendly way for people interested in studying and preserving marine life to participate in the research, Veirs said. The question at the heart of the project, he added, is how to organize and train people who listen to the streaming sound to be better whale detectors. Orcasound also stores audio data on cloud online servers for later analysis – from people and algorithms.
Each node on the network uses a cheap Raspberry Pi computer with additional audio material. Computers run the Linux operating system and open source software to encode and send audio using standard data formats that are popular with online streaming video services such as YouTube. This minimizes costs while maximizing browser compatibility and ease of use. "We want to make it really easy for scientists to hear signals," Veirs said.
Future versions of the app will have a button that users can click when they hear something interesting, which will help comment on the data for algorithms that will be analyzed later. While there may be somewhat friendly rivalry between machines and people in this arena, the Orcasound application aims to bring synergy between civilian scientists and sophisticated algorithms.
How to reduce the impact of shipping noise on fish? Release them
Presentation # 2pAO1, "The Orcasound Application: An Open Source Solution for Live Ocean Sound Flow to Scientist Citizens and Cloud-Based Algorithms" by Scott Veirs will take place on Tuesday, November 6, 1:00 pm. in the Esquimalt Hall of the Victoria Congress Center in Victoria, British Columbia, Canada. More information about the project can be found at www.orcasound.net