Language-invariant Bottleneck Features from Adversarial End-to-end Acoustic Models for Low Resource Speech Recognition.

Jiangyan Yi, Jianhua Tao, Ye Bai

Browse the full ICASSP paper archive.