Total least squares based subband modelling for scalable speech representations with damped sinusoids.

Kris Hermus, Werner Verhelst, Patrick Wambacq, Philippe Lemmerling

Browse the full Interspeech paper archive.