Audio Signal Modelling Using Bayesian Atomic Decompositions

Wolfe PJ, Godsill SJ, Ng WJ, Doerfler M. Audio Signal Modelling Using Bayesian Atomic Decompositions. In: AES Convention 112. Audio Engineering Society; 2002. Paper 5624. Available from: https://aes.org/publications/elibrary-page/?id=11317

@inproceedings{Wolfe2002_11317,
  author = {Wolfe, Patrick J. and Godsill, Simon J. and Ng, Wee Jing and Doerfler, Monika},
  title = {{Audio Signal Modelling Using Bayesian Atomic Decompositions}},
  booktitle = {AES Convention 112},
  note = {Paper 5624},
  year = {2002},
  month = apr,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=11317}
}

TY  - CPAPER
TI  - Audio Signal Modelling Using Bayesian Atomic Decompositions
AU  - Wolfe, Patrick J.
AU  - Godsill, Simon J.
AU  - Ng, Wee Jing
AU  - Doerfler, Monika
T2  - AES Convention 112
M1  - Paper 5624
PY  - 2002
DA  - 2002/04/06
UR  - https://aes.org/publications/elibrary-page/?id=11317
PB  - Audio Engineering Society
LA  - en
AB  - We present an investigation into signal processing models appropriate for audio, and especially high quality musical signals, by means of Bayesian atomic decompositions. At present, many models rely on short-term stationarity of the audio, or highly limiting forms of non-stationarity. Moreover, they are well-suited only to low-level inference tasks. We seek to formulate a new generation of audio models that will address the main limitations of the existing ones and permit high-level inference. As we show, such models result from the marriage of an overcomplete dictionary of time-frequency atoms with structured hierarchical prior probability distributions developed specifically for audio signals, in order to model coefficient correlation in time and frequency.
ER  -