Hybrid Feature Global Attention Network for Noisy-reverberant Speech Enhancement.

Zehua Zhang, Shiyun Xu, Yinghan Cao, Changjun He

Browse the full ICASSP paper archive.