Unsupervised spoken term detection with acoustic segment model | IEEE Conference Publication | IEEE Xplore