Matrix of Polynomials Model Based Polynomial Dictionary Learning Method for Acoustic Impulse Response Modeling

Jian Guan, Xuan Wang, Pengming Feng, Jing Dong, Wenwu Wang


We study the problem of dictionary learning for signals that can be represented as polynomials or polynomial matrices, such as convolutive signals with time delays or acoustic impulse responses. Recently, we developed a method for polynomial dictionary learning based on the fact that a polynomial matrix can be expressed as a polynomial with matrix coefficients, where the coefficient of the polynomial at each time lag is a scalar matrix. However, a polynomial matrix can be also equally represented as a matrix with polynomial elements. In this paper, we develop an alternative method for learning a polynomial dictionary and a sparse representation method for polynomial signal reconstruction based on this model. The proposed methods can be used directly to operate on the polynomial matrix without having to access its coefficients matrices. We demonstrate the performance of the proposed method for acoustic impulse response modeling.


 DOI: 10.21437/Interspeech.2017-395

Cite as: Guan, J., Wang, X., Feng, P., Dong, J., Wang, W. (2017) Matrix of Polynomials Model Based Polynomial Dictionary Learning Method for Acoustic Impulse Response Modeling. Proc. Interspeech 2017, 3068-3072, DOI: 10.21437/Interspeech.2017-395.


@inproceedings{Guan2017,
  author={Jian Guan and Xuan Wang and Pengming Feng and Jing Dong and Wenwu Wang},
  title={Matrix of Polynomials Model Based Polynomial Dictionary Learning Method for Acoustic Impulse Response Modeling},
  year=2017,
  booktitle={Proc. Interspeech 2017},
  pages={3068--3072},
  doi={10.21437/Interspeech.2017-395},
  url={http://dx.doi.org/10.21437/Interspeech.2017-395}
}