Abstract
In this talk we describe Bayesian computational models and methods for inference about noisy audio signals, with the aim of performing tasks such as musical trnscription, source separation, automated annotation with content descriptors, noise reduction and object‐based coding. The models are structured models of non‐stationary sparsity in audio sources, usually expressed in the time‐frequeny plane. The computational inference methods are based around Monte Carlo techniques including the particle filter for rapid on‐line inference and Markov chain Monte Carlo for batch problems of higher compleixity. Examples will be given from the spheres of source separation, multiresolution noise reduction, denoising and intetrpolation of missing batches from audio. We will also describe their use for acoustical analysis of the properties of a musical instrument, focusing on the parameters of inharmonicity in struck or plucked strings.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
