Mixed-lingual spoken word recognition by using VQ codebook sequences of variable length segments.

Hiroaki Kojima, Kazuyo Tanaka

Browse the full Interspeech paper archive.