Combining eigenvoice speaker modeling and VTS-based environment compensation for robust speech recognition | IEEE Conference Publication | IEEE Xplore