PhD Studentship, AI Tools for The Identification and Characterisation of Underwater Sounds

Institute of Sound & Vibration Research

Location:  Highfield Campus
Closing Date:  Friday 09 April 2021
Reference:  1344021DA
Supervisory Team:  Paul White

Project description

This PhD will develop tools based on Machine Learning (a form of Artificial Intelligence) to automate the process of detecting and classifying sounds heard in the ocean.  Sound is the most effective method of communication underwater, so animals, like whales and dolphins, have evolved to rely upon it to explore their environment and to communicate with each other. Physical processes, such as rain and wind at the sea surface, create a persistent level of background noise, the character of which evolves over time and varies from location to location. In addition to these naturally occurring sounds, there are many man-made acoustic sources in the ocean, including instruments like sonars and depth sounders and incidental noise sources like the noise from ships as the move across the ocean (dosits). Together this creates a complex acoustic environment in which detecting and identifying the source of a sound is an extremely challenging problem.-

The potential impact on marine ecosystems from noise pollution provides a strong motivator for monitoring of sounds in the ocean (https://tinyurl.com/1et89y7n). Recent technological advances have enabled the collection of acoustic data from the aquatic world at an ever increasing rate.  Historically, it was possible to analyse recordings “by-hand”, studying sound in detail to determine its source.  Such a time-consuming process is no longer feasible and automated methods of performing this classification are becoming vital (https://tinyurl.com/y8e9cto5). Artificial intelligence (AI) and Machine Learning (ML) offer one possible solution to this problem.  However, whilst there is a plethora of data now available, very little of it comes from a validated source.  In many cases the most interesting sounds are those for which there is very little confirmed data, e.g. the call of a rare whale. This project will focus on the challenge of building classification systems when there a few examples of the sounds of interest. We will seek to exploit our knowledge of sound propagation and of underwater noise in general to maximise the utility of AI/ML systems and to build effective classification systems.

This PhD is part sponsored by BAE Systems and will be conducted with the Signal Processing and Hearing group (https://tinyurl.com/ahoupzkv) in the ISVR (https://tinyurl.com/auy7bqho). The sponsorship allows us to offer an enhanced stipend for the successful candidate, who will also be given the opportunity to spend 3 months with BAE during the project. The work will employ datasets from a range of sources, many of which already exist. This makes the PhD robust to conditions imposed by the pandemic, should they persist beyond the start of the project.

Entry Requirements

A very good undergraduate degree (at least a UK 2:1 honours degree, or its international equivalent).

Closing date: 9 April 2021, but later applications may be considered depending on the funds remaining in place.

Funding: For UK students, Tuition Fees and a stipend of £20,285 tax-free per annum for up to 3.5 years. 

How To Apply

Apply online, programme (Research), 2021/22, Faculty of Physical Sciences and Engineering, next page “PhD  Engineering & Environment (Full time)”. In Section 2 you should insert the name of the supervisor Paul White 

Applications should include: 

Curriculum Vitae
Two reference letters
Degree Transcripts to date

Apply online: https://www.southampton.ac.uk/courses/how-to-apply/postgraduate-applications.page

For further information please contact: feps-pgr-apply@soton.ac.uk