Computer auditory systems are a very important application of AI techniques. This course aims to give students knowledge and hand-on experience in computer auditory systems. It will also introduce major audio signal processing methods based on machine learning techniques including audio detection, audio recognition, audio description algorithms and audio synthesis. It aims to introduce most recent advances in the field of machine listening such as acoustic event detection, music information retrieval, and deep fake detection.
A. Implement major time-frequency analysis methods for audio signal processing. B. Analyse the principles of human auditory perception and relate them to the design constraints and evaluation criteria for computer auditory systems. C. Design and develop components of computer auditory systems (such as for detection or recognition), evaluating their performance and effectiveness. D. Construct and validate a complete audio processing solution, demonstrating professional practice through individual technical work and collaborative project management.
This module will be delivered through a combination of lectures, tutorials and labs. Lectures will be designed to provide key theoretical foundations and introduce to students the essential algorithms and techniques in computer auditory systems. Tutorials will introduce the fundamental tools and design procedures for constructing such systems, and provide students with the opportunity to communicate with each other. Labs will be designed to introduce students to statistical software used in the evaluation of computer auditory systems.