Ask a Question

Prefer a chat interface with context about you and your work?

Reverberant Audio Source Separation via Sparse and Low-Rank Modeling

Reverberant Audio Source Separation via Sparse and Low-Rank Modeling

The performance of audio source separation from underdetermined convolutive mixture assuming known mixing filters can be significantly improved by using an analysis sparse prior optimized by a reweighting l1 scheme and a wideband datafidelity term, as demonstrated by a recent article. In this letter, we show that the performance can …