The standard approach to speaker verification is to extract cepstral features from the speech spectrum and model them by generative or discriminative techniques. We propose a nov...
We present a novel language identification technique using our recently developed deep-structured conditional random fields (CRFs). The deep-structured CRF is a multi-layer CRF mo...
Supervised learning uses a training set of labeled examples to compute a classifier which is a mapping from feature vectors to class labels. The success of a learning algorithm i...
—The two-path algorithm is a well-known approach for overcoming the dead-lock problem in echo cancellation systems. Typically, a fixed foreground filter is producing the echo c...