Multichannel audio source separation exploiting NMF-based generic source spectral model in Gaussian modeling framework
| Autor: | Quoc Cuong Nguyen, Thanh Thi Hien Duong, Ngoc Q. K. Duong, Cong-Phuong Nguyen |
|---|---|
| Přispěvatelé: | International Research Institute MICA (MICA), Institut National Polytechnique de Grenoble (INPG)-Hanoi University of Science and Technology (HUST)-Centre National de la Recherche Scientifique (CNRS), Technicolor R & I [Cesson Sévigné], Technicolor, Hanoi University of Science and Technology (HUST), Duong, Ngoc |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
Audio signal
Covariance function Computer science Estimation theory Gaussian modeling Gaussian SIGNAL (programming language) 020206 networking & telecommunications generic spectral model 02 engineering and technology spatial covariance model [STAT.ML] Statistics [stat]/Machine Learning [stat.ML] Non-negative matrix factorization 030507 speech-language pathology & audiology 03 medical and health sciences symbols.namesake [STAT.ML]Statistics [stat]/Machine Learning [stat.ML] non- negative matrix factorization Multichannel audio source separation 0202 electrical engineering electronic engineering information engineering symbols Benchmark (computing) Source separation 0305 other medical science Algorithm |
| Zdroj: | 14th Int. Conf. on Latent Variable Analysis and Signal Separation (LVA ICA) 14th Int. Conf. on Latent Variable Analysis and Signal Separation (LVA ICA), Jul 2018, London, United Kingdom Latent Variable Analysis and Signal Separation ISBN: 9783319937632 LVA/ICA |
| Popis: | International audience; Nonnegative matrix factorization (NMF) has been well-known as a powerful spectral model for audio signals. Existing work, including ours, has investigated the use of generic source spectral models (GSSM) based on NMF for single-channel audio source separation and shown its efficiency in different settings. This paper extends the work to multichannel case where the GSSM is combined with the source spatial covariance model within a unified Gaussian modeling framework. Specially, unlike a conventional combination where the estimated variances of each source are further constrained by NMF separately, we propose to constrain the total variances of all sources altogether and found a better separation performance. We present the expectation-maximization (EM) algorithm for the parameter estimation. We demonstrate the effectiveness of the proposed approach by using a benchmark dataset provided within the 2016 Signal Separation Evaluation Campaign. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|