Trainable Dynamic Subsampling for End-to-End Speech Recognition.

Shucong Zhang, Erfan Loweimi, Yumo Xu, Peter Bell, Steve Renals

Browse the full Interspeech paper archive.