Acoustic modeling based on the MDL principle for speech recognition.

Koichi Shinoda, Takao Watanabe

Browse the full Interspeech paper archive.