Smoothed N-best-based speaker adaptation for speech recognition.

Tomoko Matsui, Tatsuo Matsuoka, Sadaoki Furui

Browse the full ICASSP paper archive.