{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/unifying-probabilistic-models-for-time","title":"Unifying Probabilistic Models for Time-Frequency Analysis","arxiv_id":"1811.02489","date":"2018-11-06","proceeding":null,"authors":["William J. Wilkinson","Michael Riis Andersen","Joshua D. Reiss","Dan Stowell","Arno Solin"],"abstract":"In audio signal processing, probabilistic time-frequency models have many\nbenefits over their non-probabilistic counterparts. They adapt to the incoming\nsignal, quantify uncertainty, and measure correlation between the signal's\namplitude and phase information, making time domain resynthesis\nstraightforward. However, these models are still not widely used since they\ncome at a high computational cost, and because they are formulated in such a\nway that it can be difficult to interpret all the modelling assumptions. By\nshowing their equivalence to Spectral Mixture Gaussian processes, we illuminate\nthe underlying model assumptions and provide a general framework for\nconstructing more complex models that better approximate real-world signals.\nOur interpretation makes it intuitive to inspect, compare, and alter the models\nsince all prior knowledge is encoded in the Gaussian process kernel functions.\nWe utilise a state space representation to perform efficient inference via\nKalman smoothing, and we demonstrate how our interpretation allows for\nefficient parameter learning in the frequency domain.","url_abs":"http://arxiv.org/abs/1811.02489v6","url_pdf":"http://arxiv.org/pdf/1811.02489v6.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"unifying-probabilistic-models-for-time","repo_url":"https://github.com/wil-j-wil/unifying-prob-time-freq","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"audio-signal-processing","task_name":"Audio Signal Processing"},{"task_slug":"gaussian-processes","task_name":"Gaussian Processes"},{"task_slug":"resynthesis","task_name":"Resynthesis"}],"methods":[{"method_slug":"gaussian-process","method_name":"Gaussian Process"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}