UHURA: Dolphin Discovery Toolkit

August 2014 - May 2016


Advisors/Collaborators: Denise Herzing, Thad Starner

Notes:
  1. The PBS Newshour piece on our research work. The tool I had developed is featured in 1:13-1:30 in the video.
  2. Denise Herzing's TED talk on the Wild Dolphin Project.

Unsupervised Harvesting and Utilization of Recognizable Acoustics (or UHURA): This is a software toolkit that allows researchers to analyze dolphin audio files using various machine learning techniques. I worked with the Wild Dolphin Project since 2014 on developing different kinds of technologies to aid marine biologists and divers. The toolkit consists of an Ocean Noise Detector (which runs the raw dolphin acoustic data through an FFT, to filter out the unwanted segments of data), a Discovery Tool (that analyzes animal (here, dolphin) vocalizations and discovers patterns in the data), and a front-end display interface to view the signals. One instance of the front-end interface was developed using the Noldus Observer XT event logging software (shown below).

Tasks Performed:
  1. Spectrogram Analyses and Signal Processing
  2. K-Means Clustering and other Machine Learning techniques
  3. Design of a visualization tool using the Noldus Observer XT platform
  4. Design of an ocean noise detector which is currently being used in the Wild Dolphin Project