A maximum likelihood training approach to irrelevant variability compensation based on piecewise linear transformations.

Qiang Huo, Donglai Zhu

Browse the full Interspeech paper archive.