Context dependent vector quantization for continuous speech recognition | IEEE Conference Publication | IEEE Xplore