On Modular Training of Neural Acoustics-to-Word Model for LVCSR.

Zhehuai Chen, Qi Liu, Hao Li, Kai Yu

Browse the full ICASSP paper archive.