Prof. Robert Nickel, Ph.D.

Ruhr-Universität Bochum
Institut für Kommunikationsakustik
Fakultät für Elektrotechnik und Informationstechnik
Raum: ID/2/221
Universitätsstr. 150
D-44780 Bochum
Email: robert.nickel@rub.de
Tel.: +49 234 32 25872

Project Summary:

The goal of the proposed work is the development of a new and improved method for the de-noising and de-reverberation of speech signals in hearing instruments. The proposed new method was pioneered by the research team of Prof. Robert Nickel in the context of (non-reverberant) monaural speech enhancement [Xiao et al. 2008, 2009a, 2009b, Nickel et al. 2009]. The approach is fundamentally different from most conventional and state-of-the-art de-noising and de-reverberation schemes. Instead of filtering a distorted signal, a new “clean” signal is re-synthesized based on its likely characteristics. These characteristics are estimated from the distorted signal. Recently, [Xiao, Lee, and Nickel 2008, 2009a, 2009b] presented a first successful monaural (single-channel) implementation of the proposed method for mobile communication systems. In addition, the new approach also offers great potential to improve existing de-noising and de-reverberation schemes in hearing instruments. To cope with the specific requirements of hearing instruments, however, the general approach must be substantially modified and extended.

Project Publications:

The project has led to the following publications:

  1. “Inventory-Style Speech Enhancement with Uncertainty-of-Observation Techniques,” by R. M. Nickel, R. F. Astudillo, D. Kolossa, S. Zeiler, and R. Martin, Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2012), Kyoto, Japan, March 25 - March 30, 2012.

  2. “Memory and Complexity Reduction for Inventory-Style Speech Enhancement Systems,” by R. M. Nickel and R. Martin, Proceedings of the 19th European Signal Processing Conference (EUSIPCO 2011), Barcelona, Spain, August 29 - September 2, 2011, pp. 196-200.
  3. “Speech Inventory Based Discriminative Training for Joint Speech Enhancement and Low-Rate Speech Coding,” X. Xiao, R. M. Nickel, Proceedings of INTERSPEECH 2010 (International Speech Communication Association), Makuhari, Japan, September 26-30, 2010, pp. 2398-2401.

  4. “Joint Noise Reduction and Bandwidth Extension via Inventory Based Speech Analysis/Resynthesis,” X. Xiao, R. M. Nickel, Proceedings of the 9th Speech Communication Conference of the Information Technology Society of the Association of German Engineers (9. ITG-VDE Fachtagung Sprachkommu-nikation), Bochum, Germany, October 6-8, 2010.

There are several other papers still in preparation. These are either directly linked to the project work or grew out of collaborations that resulted from the project work.

 

A Short CV of Assist. Prof. Robert M. Nickel, Ph.D.

Professional Preparation:

Degree: Ph.D. in Electrical Engineering in 2001.
School: University of Michigan, Ann Arbor, Michigan, 1997-2001.
Major:   Signal Processing.
Dissertation:  Generalized Scale Transforms, Theory and Applications.
Degree:    M.S.E. in Electrical Engineering in 1997.
School:  University of Michigan, Ann Arbor, Michigan, 1995-1997.
Major:  Signal Processing.
Degree:  Diplom Ingenieur in Electrical Engineering in 1995.
School: Rheinisch-Westfälische Technische Hochschule (RWTH),
Aachen, Germany, 1988-1995.
Major:  Communications.
Thesis: Unsupervised System Identification for Plastic Molding Machines.
                                                                                            

Academic Appointments:

Appointment: Assistant Professor since July 2007.
Institution: Department of Electrical Engineering,
Bucknell University, Lewisburg, Pennsylvania.
Appointment: Assistant Professor, August 2002 – June 2007.
Institution:  Electrical Engineering Department,
Penn State University, University Park, Pennsylvania.
Appointment: Adjunct Faculty, Academic Year 2001/2002.
Institution: Electrical Engineering and Computer Science Department,
University of Michigan, Ann Arbor, Michigan.

Current Research Interests:

- General Speech Signal Processing.
- Speech Modeling, Coding, Synthesis, and Enhancement.
- Speech/Speaker/Language Recognition.
- Detection, Estimation, and Classification.
- General Signal Theory.
- Time-Frequency Analysis and Generalized Scale Transforms.

List of Selected Publications (other than Project Publications):

  1. “Speech Enhancement with Inventory Style Speech Resynthesis,” by X. Xiao and R. M. Nickel, IEEE Transactions on Audio, Speech, and Language Processing, Vol. 18, No. 6, Aug. 2010, pp. 1243-1257.
  2. "Automatic Speech Character Identification," R. M. Nickel, A Tutorial Article for the IEEE Circuits and Systems Magazine, Vol. 6, No. 4, pp. 8-29, 2006.
  3. "Noise Robust Speech Coding at Very Low Bit Rates," X. Xiao, R. M. Nickel, Proceedings of the 17th European Signal Processing Conference (EUSIPCO), Glasgow, Scotland, August 24-28, 2009, pp. 2529-2533.
  4. "Inventory Based Speech Enhancement for Speaker Dedicated Speech Communication Systems," X. Xiao, P. Lee, R. M. Nickel, Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Taipei, Taiwan, April 2009, pp. 3877-3880.
  5. "Inventory Based Speech Denoising with Hidden Markov Models," X. Xiao, P. Lee, R. M. Nickel, Proceedings of the 16th European Signal Processing Conference (EUSIPCO), Lausanne, Switzerland, August 25-29, 2008.
  6. "A Least-Squares Approach to Blind Source Separation in Multispeaker Environments," R. M. Nickel, A. N. Iyer, Journal of Computers, Vol. 2, No. 7, pp. 59-67, September, 2007.
  7. "Robust Speaker Verification with Principal Pitch Components," R. M. Nickel, S. P. Oswal, and A. N. Iyer, International Journal of Speech Technology (Springer), Vol. 8, No. 4, pp. 323-339, December, 2005.
  8. "A Novel Approach to Automated Source Separation in Multispeaker Environments," R. M. Nickel, A. N. Iyer, Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Vol. 5, pp. 629-632, May 14-19, Toulouse, France, 2006.
  9. "Robust Speaker Verification with Principal Pitch Components," R. M. Nickel, S. P. Oswal, A. N. Iyer, Proceedings of the XII. European Signal Processing Conference (EUSIPCO), pp. 1024-1026, September 6-10, Vienna, Austria, 2004.

Links and Related Sites:

Foto Robert Nickel