Discriminative data selection for lightly supervised training of acoustic model using closed caption texts.

Sheng Li, Yuya Akita, Tatsuya Kawahara

Browse the full Interspeech paper archive.