Abstract
In this letter, we describe a single-channel speech dereverberation method in the short-time Fourier transform domain by using non-negative blind deconvolution. Robust decomposition of the magnitude spectra of the reverberated speech into its clean speech and a reverberation filter can be achieved by imposing a sparse and frequency-dependent prior model on the speech and an exponentially decaying envelope on the filter. Subsequently improved dereverberated speech is estimated without crude speech prior imposition for the fixed reverberation filter. The effectiveness of the algorithm was demonstrated with experimental results on speech reverberated by room impulse responses.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Nakatani, T., Yoshioka, T., Kinoshita, K., Miyoshi, M., Juang, B.-H.: Blind speech dereverberation with multi-channel linear prediction based on short time Fourier transform representation. In: IEEE ICASSP, pp. 85–88 (2008)
Kameoka, H., Nakatani, T., Yoshioka, T.: Robust speech dereverberation based on non-negativity and sparse nature of speech spectrograms. In: IEEE ICASSP, pp. 45–48 (2009)
Singh, R., Raj, B., Smaragdis, P.: Latent-variable decomposition based dereverberation of monaural and multi-channel signals. In: IEEE ICASSP, pp. 1914–1917 (2010)
Mysore, G.J., Smaragdis, P.: A convolutive spectral decomposition approach to the separation of feedback from target speech. In: IEEE Int. workshop on Machine Learning for Signal Processing (2011)
Krueger, A., Haeb-Umbach, R.: Model-based feature enhancement for reverberant speech recognition. IEEE Trans. Audio, Speech, Language Process. 18(7), 1692–1707 (2010)
Smaragdis, P.: Non-negative matrix factor deconvolution, extraction of multiple sound sources from monophonic inputs. In: Int. Workshop on ICA and BSS, pp. 494–499 (2004)
Lee, D.D., Seung, H.S.: Algorithms for non-negative matrix factorization. In: Neural Information Processing Systems, pp. 556–562. MIT Press (2001)
Kim, T., Attias, H.T., Lee, S.-Y., Lee, T.-W.: Blind source separation exploiting higher-order frequency dependencies. IEEE Trans. Audio, Speech, Language Process. 15(1), 70–79 (2007)
Nakamura, S., Hiyane, K., Asano, F., Nishiura, T., Yamada, T.: Acoustical sound database in real environments for sound scene understanding and hands-free speech recognition. In: Int. Conf. on Language Resources and Evaluation (2000)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Jeong, IY., Kim, B., Park, HM. (2013). Single-Channel Speech Dereverberation Based on Non-negative Blind Deconvolution and Prior Imposition on Speech and Filter. In: Lee, M., Hirose, A., Hou, ZG., Kil, R.M. (eds) Neural Information Processing. ICONIP 2013. Lecture Notes in Computer Science, vol 8228. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-42051-1_58
Download citation
DOI: https://doi.org/10.1007/978-3-642-42051-1_58
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-42050-4
Online ISBN: 978-3-642-42051-1
eBook Packages: Computer ScienceComputer Science (R0)