Speaker and noise independent online single-channel speech enhancement

Abstract

Desirable properties of real-world speech enhancement methods include online operation, single-channel operation, operation in the presence of a variety of noise types including non-stationary noise, and no requirement for isolated training examples of the specific speaker and noise type at hand. Methods in the literature typically possess only a subset of these properties. Source separation methods particularly rarely simultaneously possess the first and last properties. We extend universal speech model-based speech enhancement to adaptively learn a noise model in an online fashion. We learn a model from a general corpus of speech in place of speakerdependent training examples before deployment. This setup provides all of these desirable properties, making it easy to deploy in real-world systems without the need to provide additional training examples, while explicitly modeling speech. Our experimental results show that our method achieves the same performance as in the case in which speaker-dependent training data is available.Index Termsonline speech enhancement, non-negative matrix factorization, universal speech models

Robust Non-negative Matrix Factorization with β-Divergence for Speech Separation

“…1 is only describing a basic separation system to help focus on the selection of the divergence cost function to be used under the sparse and lowrank framework. The obtained performance can, however, be further improved through techniques such as adopting a universal speech dictionary [23], imposing temporal continuity to the sparse matrix [24], using an information fusion strategy [25], or a combination with autocorrelation [26]. The use of these techniques for performance improvement is beyond the scope of this paper, and will be explored in future works.…”