Skip to Main content Skip to Navigation
Conference papers

SINGLE-CHANNEL SPEAKER-DEPENDENT SPEECH ENHANCEMENT EXPLOITING GENERIC NOISE MODEL LEARNED BY NON-NEGATIVE MATRIX FACTORIZATION

Abstract : This paper considers the single-channel speech separation problem given a noisy observation recorded by a microphone. More precisely, we focus on the speaker-dependent approach where spectral characteristic of target speech is learned in advance from a clean example. In training process, we propose to learn a generic spectral model for noise source by collecting various types of environmental noise via the established non-negative matrix factorization framework. In speech enhancement process, we propose to combine two existing group sparsity-inducing penalties in the optimization function and derive the corresponding algorithm for parameter estimation based on multiplicative update (MU) rule. Experiment result over mixtures containing different real-world noises confirms the effectiveness of our approach.
Complete list of metadata

Cited literature [7 references]  Display  Hide  Download

https://hal.inria.fr/hal-01288277
Contributor : Ngoc Duong <>
Submitted on : Tuesday, March 15, 2016 - 11:07:10 AM
Last modification on : Friday, November 6, 2020 - 4:38:54 AM
Long-term archiving on: : Thursday, June 16, 2016 - 10:39:47 AM

File

iceic2016_Edit2.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01288277, version 1

Collections

CNRS | UGA

Citation

Hien-Thanh Duong, Quoc-Cuong Nguyen, Cong-Phuong Nguyen, Ngoc Q. K. Duong. SINGLE-CHANNEL SPEAKER-DEPENDENT SPEECH ENHANCEMENT EXPLOITING GENERIC NOISE MODEL LEARNED BY NON-NEGATIVE MATRIX FACTORIZATION. IEEE International Conference on Electronics, Information and Communication, Jan 2016, Da Nang, Vietnam. ⟨hal-01288277⟩

Share

Metrics

Record views

151

Files downloads

512