Underdetermined Methods for Multichannel Audio Enhancement with Partial Preservation of Background Sources
Multichannel audio enhancement and source separation traditionally attempt to isolate a single source and remove all background noise. In listening enhancement applications, however, a portion of the background sources should be retained to preserve the listeners spatial awareness. We describe a time-varying spatial filter designed to apply a different gain to each sound source with minimal distortion of the source spectra and background spatial cues. The filter, inspired by methods from underdetermined source separation, adjusts its parameters at each time-frequency bin to preserve dominant sources. The nonstationary model allows the filter to process more sources than it otherwise could, while partial background preservation improves robustness to noise and errors.