Supervised and unsupervised speech enhancement using nonnegative matrix factorization

Nasser Mohammadiha, Paris Smaragdis, Arne Leijon

Research output: Contribution to journalArticlepeer-review

Abstract

Reducing the interference noise in a monaural noisy speech signal has been a challenging task for many years. Compared to traditional unsupervised speech enhancement methods, e.g., Wiener filtering, supervised approaches, such as algorithms based on hidden Markov models (HMM), lead to higher-quality enhanced speech signals. However, the main practical difficulty of these approaches is that for each noise type a model is required to be trained a priori. In this paper, we investigate a new class of supervised speech denoising algorithms using nonnegative matrix factorization (NMF). We propose a novel speech enhancement method that is based on a Bayesian formulation of NMF (BNMF). To circumvent the mismatch problem between the training and testing stages, we propose two solutions. First, we use an HMM in combination with BNMF (BNMF-HMM) to derive a minimum mean square error (MMSE) estimator for the speech signal with no information about the underlying noise type. Second, we suggest a scheme to learn the required noise BNMF model online, which is then used to develop an unsupervised speech enhancement system. Extensive experiments are carried out to investigate the performance of the proposed methods under different conditions. Moreover, we compare the performance of the developed algorithms with state-of-the-art speech enhancement schemes using various objective measures. Our simulations show that the proposed BNMF-based methods outperform the competing algorithms substantially.

Original languageEnglish (US)
Article number2270369
Pages (from-to)2140-2151
Number of pages12
JournalIEEE Transactions on Audio, Speech and Language Processing
Volume21
Issue number10
DOIs
StatePublished - Aug 8 2013

Keywords

  • Bayesian inference
  • HMM
  • PLCA
  • nonnegative matrix factorization (NMF)
  • speech enhancement

ASJC Scopus subject areas

  • Acoustics and Ultrasonics
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Supervised and unsupervised speech enhancement using nonnegative matrix factorization'. Together they form a unique fingerprint.

Cite this