Speech enhancement based on nonnegative matrix factorization with mixed group sparsity constraint

Hien-Thanh Duong 1, 2 Quoc-Cuong Nguyen 2, 3 Cong-Phuong Nguyen 3 Thanh-Huan Tran 4 Ngoc Q. K. Duong 5
1 HUMG - Hanoi University of Mining and Geology
2 MICA - International Research Institute MICA
3 HUST - Hanoi University of Science and Technology
4 Hanoi University of Industry
5 Technicolor R & I [Cesson Sévigné]
Abstract : This paper addresses a challenging single-channel speech enhancement problem in real-world environment where speech signal is corrupted by high level background noise. While most state-of-the-art algorithms tries to estimate noise spectral power and filter it from the observed one to obtain enhanced speech, the paper discloses another approach inspired from audio source separation technique. In the considered method, generic spectral characteristics of speech and noise are first learned from various training signals by non-negative matrix factorization (NMF). They are then used to guide the similar factorization of the observed power spectrogram into speech part and noise part. Additionally, we propose to combine two existing group sparsity-inducing penalties in the optimization process and adapt the corresponding algorithm for parameter estimation based on mul-tiplicative update (MU) rule. Experiment results over different settings confirm the effectiveness of the proposed approach .
Keywords : spectral model group sparsity Speech enhancement audio source separation nonnegative matrix factorization multiplicative update
Type de document :
Communication dans un congrès
6th ACM International Symposium on Information and Communication Technology, Dec 2015, Hanoi, Vietnam. 〈10.1145/2833258.2833276〉
Liste complète des métadonnées

Littérature citée [17 références]  Voir  Masquer  Télécharger
https://hal.inria.fr/hal-01288291
Contributeur : Ngoc Duong <>
Soumis le : mardi 15 mars 2016 - 11:06:10
Dernière modification le : jeudi 7 février 2019 - 17:55:57
Document(s) archivé(s) le : jeudi 16 juin 2016 - 10:34:33

Fichier

paper_review.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Hien-Thanh Duong, Quoc-Cuong Nguyen, Cong-Phuong Nguyen, Thanh-Huan Tran, Ngoc Q. K. Duong. Speech enhancement based on nonnegative matrix factorization with mixed group sparsity constraint. 6th ACM International Symposium on Information and Communication Technology, Dec 2015, Hanoi, Vietnam. 〈10.1145/2833258.2833276〉. 〈hal-01288291〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

98

Téléchargements de fichiers

248

Contact


archive-ouverte@inria.fr