A Generative Product-of-Filters Model of Audio
Abstract
We propose the product-of-filters (PoF) model, a generative model that decomposes audio spectra as sparse linear combinations of "filters" in the log-spectral domain. PoF makes similar assumptions to those used in the classic homomorphic filtering approach to signal processing, but replaces hand-designed decompositions built of basic signal processing operations with a learned decomposition based on statistical inference. This paper formulates the PoF model and derives a mean-field method for posterior inference and a variational EM algorithm to estimate the model's free parameters. We demonstrate PoF's potential for audio processing on a bandwidth expansion task, and show that PoF can serve as an effective unsupervised feature extractor for a speaker identification task.
Cite
Text
Liang et al. "A Generative Product-of-Filters Model of Audio." International Conference on Learning Representations, 2014.Markdown
[Liang et al. "A Generative Product-of-Filters Model of Audio." International Conference on Learning Representations, 2014.](https://mlanthology.org/iclr/2014/liang2014iclr-generative/)BibTeX
@inproceedings{liang2014iclr-generative,
title = {{A Generative Product-of-Filters Model of Audio}},
author = {Liang, Dawen and Hoffman, Matthew D. and Mysore, Gautham J.},
booktitle = {International Conference on Learning Representations},
year = {2014},
url = {https://mlanthology.org/iclr/2014/liang2014iclr-generative/}
}