Sparse smoothing of articulatory features from Gaussian mixture model based acoustic-to-articulatory inversion: benefit to speech recognition.

Prasad Sudhakar, Prasanta Kumar Ghosh

Browse the full Interspeech paper archive.