Abstract
In this paper, we propose a speech enhancement method based on non-negative matrix factorization (NMF) techniques. NMF techniques allow us to approximate the power spectral density (PSD) of the noisy signal as a weighted linear combination of trained speech and noise basis vectors arranged as the columns of a matrix. In this work, we propose to use basis vectors that are parameterised by autoregressive (AR) coefficients. Parametric representation of the spectral basis is beneficial as it can encompass the signal characteristics like, e.g. the speech production model. It is observed that the parametric representation of basis vectors is beneficial while performing online speech enhancement in low delay scenarios.
Original language | English |
---|---|
Title of host publication | 2018 26th European Signal Processing Conference (EUSIPCO) |
Number of pages | 5 |
Publisher | IEEE |
Publication date | 2018 |
Article number | 8553039 |
ISBN (Print) | 978-90-827970-0-8 |
ISBN (Electronic) | 978-9-0827-9701-5 |
DOIs | |
Publication status | Published - 2018 |
Event | 26th European Signal Processing Conference (EUSIPCO 2018) - Rome, Italy Duration: 3 Sept 2018 → 7 Sept 2018 Conference number: 26 http://www.eusipco2018.org |
Conference
Conference | 26th European Signal Processing Conference (EUSIPCO 2018) |
---|---|
Number | 26 |
Country/Territory | Italy |
City | Rome |
Period | 03/09/2018 → 07/09/2018 |
Internet address |
Series | Proceedings of the European Signal Processing Conference |
---|---|
ISSN | 2076-1465 |
Keywords
- autoregressive modelling, speech enhancement, NMF