Lookup NU author(s): Yang Sun,
Professor Jonathon Chambers,
Dr Mohsen Naqvi
This is the authors' accepted manuscript of an article that has been published in its final definitive form by IEEE, 2019.
For re-use rights please refer to the publisher's terms and conditions.
Deep neural networks (DNNs) have been used for dereverberation and separation in the monaural source sepa-ration problem. However, the performance of current state-of-the-art methods is limited, particularly when applied in highly reverberant room environments. In this paper, we propose a two-stage approach with two DNN-based methods to address this problem. In the ﬁrst stage, the dereverberation of the speech mixture is achieved with the proposed dereverberation mask (DM). In the second stage, the dereverberant speech mixture is separated with the ideal ratio mask (IRM). To realize this two-stage approach, in the ﬁrst DNN-based method, the DM is integrated with the IRM to generate the enhanced time-frequency (T-F) mask, namely the ideal enhanced mask (IEM), as the training target for the single DNN. In the second DNN-based method, the DM and the IRM are predicted with two individual DNNs. The IEEE and the TIMIT corpora with real room impulse responses (RIRs) and noise from the NOISEX dataset are used to generate speech mixtures for evaluations. The proposed methods outperform the state-of-the-art speciﬁcally in highly reverberant room environments.
Author(s): Sun Y, Wang W, Chambers J, Naqvi MN
Publication type: Article
Publication status: Published
Journal: IEEE/ACM Transactions on Audio Speech and Language Processing
Print publication date: 01/01/2019
Online publication date: 17/10/2018
Acceptance date: 01/10/2018
ISSN (print): 2329-9290
ISSN (electronic): 2329-9304
Altmetrics provided by Altmetric