Academic Journal
View in EDS

Incoherent Discriminative Dictionary Learning for Speech Enhancement

Bibliographic Details
Title: Incoherent Discriminative Dictionary Learning for Speech Enhancement
Authors: Dima Shaheen, Mohiedin Wainakh, Oumayma Al Dakkak
Source: Journal of Telecommunications and Information Technology, Iss 3 (2018)
Publisher Information: National Institute of Telecommunications, 2018.
Publication Year: 2018
Subject Terms: l1 minimization algorithms, 03 medical and health sciences, sparse coding, Telecommunication, speech enhancement, supervised dictionary learning, TK5101-6720, Information technology, ADMM, T58.5-58.64, 0305 other medical science
Description: Speech enhancement is one of the many challenging tasks in signal processing, especially in the case of nonstationary speech-like noise. In this paper a new incoherent discriminative dictionary learning algorithm is proposed to model both speech and noise, where the cost function accounts for both “source confusion” and “source distortion” errors, with a regularization term that penalizes the coherence between speech and noise sub-dictionaries. At the enhancement stage, we use sparse coding on the learnt dictionary to find an estimate for both clean speech and noise amplitude spectrum. In the final phase, the Wiener filter is used to refine the clean speech estimate. Experiments on the Noizeus dataset, using two objective speech enhancement measures: frequency-weighted segmental SNR and Perceptual Evaluation of Speech Quality (PESQ) demonstrate that the proposed algorithm outperforms other speech enhancement methods tested.
Document Type: Article
ISSN: 1899-8852
1509-4553
DOI: 10.26636/jtit.2018.121317
Access URL: https://doaj.org/article/ac167b00d0de4bcd8d1543c78bc5a9f3
https://www.itl.waw.pl/czasopisma/JTIT/2018/3/42.pdf
Accession Number: edsair.doi.dedup.....d40ca58c3f0d75973b7df35d5e242f9b
Database: OpenAIRE
Description
ISSN:18998852
15094553
DOI:10.26636/jtit.2018.121317