Automatic talker activity labeling for co-channel talker interference suppression

A speaker activity detector is described. It takes cochannel speech as input and labeling intervals of the input as target-only, jammer-only, or two-speaker (target and jammer). The algorithms applied are borrowed primarily from speaker recognition, thereby allowing one to use speaker-dependent test-utterance-independent information in a front-end for cochannel talker interference suppression. Parameters studied include classifier choice (vector quantization vs. Gaussian), training method (unsupervised vs. supervised), test utterance segmentation (uniform vs. adaptive), and training and testing target-to-jammer ratios. Using analysis interval lengths of 100 ms, performance reached 80% correct detection