Fragmented context-dependent syllable acoustic models.

Kishan Thambiratnam, Frank Seide

Browse the full Interspeech paper archive.