Language model transformation applied to lightly supervised training of acoustic model for congress meetings.

Tatsuya Kawahara, Masato Mimura, Yuya Akita

Browse the full ICASSP paper archive.