Single-channel noise suppression based on a statistical source-model for speech

Forfattere

  • Niklas Harlander Medizinische Physik, Carl-von-Ossietzky Universität Oldenburg, Germany
  • Thomas Rohdenburg Medizinische Physik, Carl-von-Ossietzky Universität Oldenburg, Germany
  • Volker Hohmann Medizinische Physik, Carl-von-Ossietzky Universität Oldenburg, Germany

Resumé

We propose a single-channel noise suppression scheme based on a statistical source-model for speech. The scheme is adapted from Ephraim and Malah (1984) and Tchorz and Kollmeier (2003) and aims at improving short-time signal-to-noise ratio (SNR) estimates in different frequency subbands by learning and classifying auditory-model based speech signal features. First, the speech signal is transformed into so-called Amplitude-Modulation-Spectrograms (AMS) rstly described in Kollmeier and Koch (1994), which include information of both center frequencies and modulation frequencies within 32-ms analysis frames. Second, the short-time subband SNR is estimated from the AMS patterns by a neural network, which was trained based on a large speech database. A second neural net obtains nal SNR estimates from (i) the AMS-based SNR estimates by Tchorz and Kollmeier (2003), and (ii) the estimates derived from the traditional approach by Ephraim and Malah (1984). The final SNR estimates can be used to steer a Wiener filter for noise suppression. Experimental results indicate a reasonable SNR-estimation accuracy.

Referencer

Ephraim, Y., and Malah, D. (1984). “Speech enhancement using a minimum mean-square error short-time spectral amplitude estimator“. IEEE Signal Proc. Letters, ASSP-32,1109 – 1121.

Kollmeier, B., and Koch, R. (1994). "Speech enhancement based on physiological and psychoacoustical models of modulation perception and binaural interaction.“ J. Acoust. Soc. Am., 95, 1593 – 1602.

Martin, R. (1994). “Spectral subtraction based on minimum statistics”. Proc. European Signal Processing Conference, 1, 1182 – 1185.

Marzinzik, M., and Kollmeier, B. (2002). “Speech pause detection for noise spectrum estimation by tracking power envelope dynamics.” IEEE Transactions on Speech and Audio Processing ASSP-10, 109 – 118.

Marzinzik, M., and Kollmeier, B. (2001). “A review of the Ephraim-Malah noise reduction algorithms,” Zeitschrift für Audiologie/Audiological Acoustics ASSP-40, 4–15.

Nabney, I. T. (2004). “NETLAB. Algorithms for Pattern Recognition.” Springer Verlag, Berlin, 3rd edition.

Universität München, Institut für Phonetik und sprachliche Kommunikation (ipsk). (1995). BAS (Bayerisches Archiv für Sprachsignale) Phondat 1 - pd1 und Phondat 2 - pd2 (Phonitatrische Datenbank). http://www.bas.uni-muenchen.de/Bas.

Tchorz, J., and Kollmeier, B. (2003). “SNR Estimation Based on Amplitude Modulation Analysis With Applications to Noise Suppression,” IEEE Trans. Speech and Audio Processing, 11, 184-192.

Yderligere filer

Publiceret

2007-12-15

Citation/Eksport

Harlander, N., Rohdenburg, T., & Hohmann, V. (2007). Single-channel noise suppression based on a statistical source-model for speech. Proceedings of the International Symposium on Auditory and Audiological Research, 1, 433–440. Hentet fra https://proceedings.isaar.eu/index.php/isaarproc/article/view/2007-42

Nummer

Årg. 1 (2007): Auditory Signal Processing in Hearing Impaired Listeners

Sektion

2007/4. Speech perception and processing

Licens

Authors who publish with this journal agree to the following terms:

a. Authors retain copyright* and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.

b. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.

c. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).

*From the 2017 issue onward. The Danavox Jubilee Foundation owns the copyright of all articles published in the 1969-2015 issues. However, authors are still allowed to share the work with an acknowledgement of the work's authorship and initial publication in this journal.