Acoustic modeling with contextual additive structure for HMM-based speech recognition.

Yoshihiko Nankaku, Kazuhiro Nakamura, Heiga Zen, Keiichi Tokuda

Browse the full ICASSP paper archive.