[go: up one dir, main page]
More Web Proxy on the site http://driver.im/ skip to main content
article

Speech intelligibility improvement using convolutive blind source separation assisted by denoising algorithms

Published: 01 January 2008 Publication History

Abstract

The present study is concerned with the blind source separation (BSS) of speech and speech-shaped noise sources. All recordings were carried out in an anechoic chamber using a dummy head (two microphones, one in each ear). The program which implements the algorithm for BSS of convolutive mixtures introduced by Parra and Spence [Parra, L., Spence, C., 2000a. Convolutive blind source separation of non-stationary sources. IEEE Trans. Speech Audio Process. 8(3), 320-327 (US Patent US6167417)] was used to separate out the signals. In the postprocessing phase two different denoising algorithms were used. The first was based on a minimum mean-square error log-spectral amplitude estimator [Ephraim, E., Malah, D., 1985. Speech enhancement using a minimum mean-square error log-spectral amplitude estimator. IEEE Trans. Speech Audio Process. ASSP-33(2), 443-445], while the second one was based on Wiener filter in which the concept of an a priori signal-to-noise estimation presented by Ephraim (as mentioned above) was applied [Scalart, P., Filho, J.V., 1996. Speech enhancement based on a priori signal to noise estimation. IEEE Internat. Conf. Acoust. Speech Signal Process. 1, 629-632]. Non-sense word tests were used as a target speech in both cases while one or two disturbing sources were used as interferences. The speech intelligibility before and after the BSS was measured for three subjects with audiologically normal hearing. Next the speech signal after BSS was denoised and presented to the same listeners. The results revealed some ambiguities caused by the insufficient number of microphones compared to the number of sound sources. For one disturbance only, the intelligibility improvement was significant. However, when there were two disturbances and the target speech, the separation was much poorer. The additional denoising, as could be expected, raises the intelligibility slightly. Although the BSS method requires more research on optimization, the results of the investigation imply that it may be applied to hearing aids in the future.

References

[1]
Aichner, R., Buchner, H., et al. 2003. On-line time-domain blind source separation of nonstationary convolved signals. In: 4th Internat. Symposium on Independent Component Analysis and Blind Signal Separation (ICA2003), Nara, Japan.
[2]
Amari, S., Douglas, SC., et al. 1997. Multichannel blind deconvolution and equalization using the natural gradient. In: 1st IEEE Workshop on Signal Processing Advances in Wireless Communications.
[3]
Anemueller, J., Kollmeier, B., 2000. Amplitude modulation decorrelation for convolutive blind source separation. ICA 2000.
[4]
Real-time sound source localization and separation system and its application to automatic speech recognition. Eurospeech.
[5]
Belouchrani, A., Amin, M.G., 1996. A new approach for blind source separation using time-frequency distributions. In: Proc. SPIE.
[6]
Phonetic structure of a test material used in subjective measurements of speech quality (in Polish). Speech Language Technol. Poznan. v3. 71-80.
[7]
Blind source separation for convolutive mixtures: a unified treatment. In: Huang, Y., Benesty, J. (Eds.), Audio Signal Processing for Next Generation Multimedia Communication Systems, Kluwer Academic Publishers. pp. 255-293.
[8]
Cardoso, J.-F., 1989. Eigenstructure of the 4th-order cumulant tensor with application to the blind source separation problem. In: Proc. ICASSP 89.
[9]
Second order nonstationary source separation. J. VLSI Signal Process. v32 i1-2. 93-104.
[10]
Blind source separation and independent component analysis: a review. Neural Inf. Process. - Lett. Rev. v6 i1. 1-57.
[11]
Approximate maximum likelihood source separation using the natural gradient. IEICE Trans. Fundam. Electron. Commun. Comput. Sci. v86 i1. 198-205.
[12]
Adaptive Blind Signal and Image Processing Learning Algorithms and Applications. Wiley, Chichester/New York/Weinheim/Brisbane/Singapore/Toronto.
[13]
Cichocki, A., Belouchrani, A., 2001. Sources separation of temporally correlated sources from noisy data using bank of band-pass filters. In: Third International Conference on Independent Component Analysis and Signal Separation (ICA-2001), San Diego, USA.
[14]
Blind Separation of sources: problem statement. Signal Process. v24 i1. 11-20.
[15]
Independent component analysis, a new concept?. Signal Process. v36 i3. 287-314.
[16]
Binaural sluggishness in the perception of tone sequences and speech in noise. J. Acoust. Soc. Amer. v107 i1. 517-527.
[17]
Convolutive blind separation of speech mixtures using natural gradient. Speech Commun. v39. 65-78.
[18]
The effect of a hearing-aid on the speech-reception threshold of hearing-impaired listeners in quiet and in noise. J. Acoust. Soc. Amer. v73. 2166-2173.
[19]
Speech enhancement using a minimum mean-square error log-spectral amplitude estimator. IEEE Trans. Speech Audio Process. vASSP-33 i2. 443-445.
[20]
Harmeling, S., 2001. convbss. Berlin, Fraunhofer First Berlin.
[21]
Independent Component Analysis. Wiley, New York.
[22]
Blind separation of sources part I: an adaptive algorithm based on neuromimetic architecture. Signal Process. v24 i1. 1-10.
[23]
A method of blind separation for convolved nonstationary signals. Neurocomputing. v22 i1-3. 157-171.
[24]
Kocinski, J., 2005. Blind source separation (BSS) of sound sources. ForumAcusticum 2005, Budapest.
[25]
Speech intelligibility in various spatial configurations of background noise. Arch. Acoust. v30 i2. 173-191.
[26]
Blind source separation of convolutive mixtures of speech in frequency domain. IEICE Trans. Fundam. vE88 i7. 1640-1655.
[27]
A neural net for blind separation of nonstationary signals. Neural Networks. v8 i3. 411-419.
[28]
Separation of mixture of independent signals using time delayed correlations. Phys. Rev. Lett. v72 i23. 3634-3637.
[29]
An Introduction to the Psychology of Hearing. 4th ed. Academic Press, London.
[30]
An adaptive beamforming perspective on convolutive blind source separation. In: Davis, G. (Ed.), Noise Reduction in Speech Applications, CRC Press LLC.
[31]
Convolutive blind source separation of non-stationary sources. IEEE Trans. Speech Audio Process. v8 i3. 320-327.
[32]
On-line blind source separation of non-stationary signals. J. VLSI Signal Process. v26 i1/2. 39-46.
[33]
Pham, D.-T., Serviere, C., et al., 2003. Blind separation of convolutive audio mixtures using nonstationarity. ICA 2003, Nara, Japan.
[34]
Blind source separation combining independent component analysis and beamforming. EURASIP J. Appl. Signal Process. v11. 1135-1146.
[35]
Frequency-domain blind source separation. In: Benesty, J., Makino, S., Chen, J. (Eds.), Speech Enhancement, Springer.
[36]
Speech enhancement based on a priori signal to noise estimation. IEEE Internat. Conf. Acoust. Speech Signal Process. v1. 629-632.
[37]
A frequency domain blind signal separation method based on decorrelation. IEEE Trans. Signal Process. v50 i8. 1855-1865.
[38]
Smaragdis, P., 1997. Efficient blind separation of convolved sound mixtures. In: EEE ASSP Workshop on Applications of Signal Processing to Audio and Acoustics, New Paltz, NY.
[39]
Blind separation of convolved mixtures in the frequency domain. Neurocomputing. v22. 21-34.
[40]
Zavarehei, E., 2005a. MMSESTSA85.m.
[41]
Zavarehei, E., 2005b. WienerScalart96.m.
[42]
Blind source separation in frequency domain. Signal Process. v83 i9. 2037-2046.
[43]
Artifact reduction in biomagnetic recordings based on time-delayed second order correlations. IEEE Trans. Biomed. Eng. v47. 75-87.

Cited By

View all
  • (2019)A new dual subband fast NLMS adaptive filtering algorithm for blind speech quality enhancement and acoustic noise reductionInternational Journal of Speech Technology10.1007/s10772-019-09614-922:2(391-406)Online publication date: 19-Jul-2019
  • (2019)A new robust forward BSS adaptive algorithm based on automatic voice activity detector for speech quality enhancementInternational Journal of Speech Technology10.1007/s10772-018-9555-021:4(1007-1020)Online publication date: 9-Feb-2019
  • (2013)Clinical evaluation of the performance of a blind source separation algorithm combining beamforming and independent component analysis in hearing aid useSpeech Communication10.1016/j.specom.2012.11.00255:4(544-552)Online publication date: 1-May-2013
  • Show More Cited By
  1. Speech intelligibility improvement using convolutive blind source separation assisted by denoising algorithms

      Recommendations

      Comments

      Please enable JavaScript to view thecomments powered by Disqus.

      Information & Contributors

      Information

      Published In

      cover image Speech Communication
      Speech Communication  Volume 50, Issue 1
      January, 2008
      81 pages

      Publisher

      Elsevier Science Publishers B. V.

      Netherlands

      Publication History

      Published: 01 January 2008

      Author Tags

      1. Blind source separation
      2. Denoising
      3. Speech enhancement
      4. Speech intelligibility

      Qualifiers

      • Article

      Contributors

      Other Metrics

      Bibliometrics & Citations

      Bibliometrics

      Article Metrics

      • Downloads (Last 12 months)0
      • Downloads (Last 6 weeks)0
      Reflects downloads up to 20 Dec 2024

      Other Metrics

      Citations

      Cited By

      View all
      • (2019)A new dual subband fast NLMS adaptive filtering algorithm for blind speech quality enhancement and acoustic noise reductionInternational Journal of Speech Technology10.1007/s10772-019-09614-922:2(391-406)Online publication date: 19-Jul-2019
      • (2019)A new robust forward BSS adaptive algorithm based on automatic voice activity detector for speech quality enhancementInternational Journal of Speech Technology10.1007/s10772-018-9555-021:4(1007-1020)Online publication date: 9-Feb-2019
      • (2013)Clinical evaluation of the performance of a blind source separation algorithm combining beamforming and independent component analysis in hearing aid useSpeech Communication10.1016/j.specom.2012.11.00255:4(544-552)Online publication date: 1-May-2013
      • (2011)Blind source separation algorithm based on PSO and algebraic equations of order twoProceedings of the Third international conference on Artificial intelligence and computational intelligence - Volume Part III10.5555/2045921.2045982(444-450)Online publication date: 24-Sep-2011
      • (2009)Semi-blind suppression of internal noise for hands-free robot spoken dialog systemProceedings of the 2009 IEEE/RSJ international conference on Intelligent robots and systems10.5555/1733343.1733480(658-663)Online publication date: 10-Oct-2009

      View Options

      View options

      Media

      Figures

      Other

      Tables

      Share

      Share

      Share this Publication link

      Share on social media