---
_id: '11846'
abstract:
- lang: eng
  text: In this paper, we present a new technique for automatic speech recognition
    (ASR) in reverberant environments. Our approach is aimed at the enhancement of
    the logarithmic Mel power spectrum, which is computed at an intermediate stage
    to obtain the widely used Mel frequency cepstral coefficients (MFCCs). Given the
    reverberant logarithmic Mel power spectral coefficients (LMPSCs), a minimum mean
    square error estimate of the clean LMPSCs is computed by carrying out Bayesian
    inference. We employ switching linear dynamical models as an a priori model for
    the dynamics of the clean LMPSCs. Further, we derive a stochastic observation
    model which relates the clean to the reverberant LMPSCs through a simplified model
    of the room impulse response (RIR). This model requires only two parameters, namely
    RIR energy and reverberation time, which can be estimated from the captured microphone
    signal. The performance of the proposed enhancement technique is studied on the
    AURORA5 database and compared to that of constrained maximum-likelihood linear
    regression (CMLLR). It is shown by experimental results that our approach significantly
    outperforms CMLLR and that up to 80\% of the errors caused by the reverberation
    are recovered. In addition to the fact that the approach is compatible with the
    standard MFCC feature vectors, it leaves the ASR back-end unchanged. It is of
    moderate computational complexity and suitable for real time applications.
author:
- first_name: Alexander
  full_name: Krueger, Alexander
  last_name: Krueger
- first_name: Reinhold
  full_name: Haeb-Umbach, Reinhold
  id: '242'
  last_name: Haeb-Umbach
citation:
  ama: Krueger A, Haeb-Umbach R. Model-Based Feature Enhancement for Reverberant Speech
    Recognition. <i>IEEE Transactions on Audio, Speech, and Language Processing</i>.
    2010;18(7):1692-1707. doi:<a href="https://doi.org/10.1109/TASL.2010.2049684">10.1109/TASL.2010.2049684</a>
  apa: Krueger, A., &#38; Haeb-Umbach, R. (2010). Model-Based Feature Enhancement
    for Reverberant Speech Recognition. <i>IEEE Transactions on Audio, Speech, and
    Language Processing</i>, <i>18</i>(7), 1692–1707. <a href="https://doi.org/10.1109/TASL.2010.2049684">https://doi.org/10.1109/TASL.2010.2049684</a>
  bibtex: '@article{Krueger_Haeb-Umbach_2010, title={Model-Based Feature Enhancement
    for Reverberant Speech Recognition}, volume={18}, DOI={<a href="https://doi.org/10.1109/TASL.2010.2049684">10.1109/TASL.2010.2049684</a>},
    number={7}, journal={IEEE Transactions on Audio, Speech, and Language Processing},
    author={Krueger, Alexander and Haeb-Umbach, Reinhold}, year={2010}, pages={1692–1707}
    }'
  chicago: 'Krueger, Alexander, and Reinhold Haeb-Umbach. “Model-Based Feature Enhancement
    for Reverberant Speech Recognition.” <i>IEEE Transactions on Audio, Speech, and
    Language Processing</i> 18, no. 7 (2010): 1692–1707. <a href="https://doi.org/10.1109/TASL.2010.2049684">https://doi.org/10.1109/TASL.2010.2049684</a>.'
  ieee: A. Krueger and R. Haeb-Umbach, “Model-Based Feature Enhancement for Reverberant
    Speech Recognition,” <i>IEEE Transactions on Audio, Speech, and Language Processing</i>,
    vol. 18, no. 7, pp. 1692–1707, 2010.
  mla: Krueger, Alexander, and Reinhold Haeb-Umbach. “Model-Based Feature Enhancement
    for Reverberant Speech Recognition.” <i>IEEE Transactions on Audio, Speech, and
    Language Processing</i>, vol. 18, no. 7, 2010, pp. 1692–707, doi:<a href="https://doi.org/10.1109/TASL.2010.2049684">10.1109/TASL.2010.2049684</a>.
  short: A. Krueger, R. Haeb-Umbach, IEEE Transactions on Audio, Speech, and Language
    Processing 18 (2010) 1692–1707.
date_created: 2019-07-12T05:29:23Z
date_updated: 2022-01-06T06:51:11Z
department:
- _id: '54'
doi: 10.1109/TASL.2010.2049684
intvolume: '        18'
issue: '7'
keyword:
- ASR
- AURORA5 database
- automatic speech recognition
- Bayesian inference
- belief networks
- CMLLR
- computational complexity
- constrained maximum likelihood linear regression
- least mean squares methods
- LMPSC computation
- logarithmic Mel power spectrum
- maximum likelihood estimation
- Mel frequency cepstral coefficients
- MFCC feature vectors
- microphone signal
- minimum mean square error estimation
- model-based feature enhancement
- regression analysis
- reverberant speech recognition
- reverberation
- RIR energy
- room impulse response
- speech recognition
- stochastic observation model
- stochastic processes
language:
- iso: eng
main_file_link:
- open_access: '1'
  url: https://groups.uni-paderborn.de/nt/pubs/2010/KrHa10.pdf
oa: '1'
page: 1692-1707
publication: IEEE Transactions on Audio, Speech, and Language Processing
status: public
title: Model-Based Feature Enhancement for Reverberant Speech Recognition
type: journal_article
user_id: '44006'
volume: 18
year: '2010'
...
---
_id: '11913'
abstract:
- lang: eng
  text: In this paper we propose to employ directional statistics in a complex vector
    space to approach the problem of blind speech separation in the presence of spatially
    correlated noise. We interpret the values of the short time Fourier transform
    of the microphone signals to be draws from a mixture of complex Watson distributions,
    a probabilistic model which naturally accounts for spatial aliasing. The parameters
    of the density are related to the a priori source probabilities, the power of
    the sources and the transfer function ratios from sources to sensors. Estimation
    formulas are derived for these parameters by employing the Expectation Maximization
    (EM) algorithm. The E-step corresponds to the estimation of the source presence
    probabilities for each time-frequency bin, while the M-step leads to a maximum
    signal-to-noise ratio (MaxSNR) beamformer in the presence of uncertainty about
    the source activity. Experimental results are reported for an implementation in
    a generalized sidelobe canceller (GSC) like spatial beamforming configuration
    for 3 speech sources with significant coherent noise in reverberant environments,
    demonstrating the usefulness of the novel modeling framework.
author:
- first_name: Dang Hai
  full_name: Tran Vu, Dang Hai
  last_name: Tran Vu
- first_name: Reinhold
  full_name: Haeb-Umbach, Reinhold
  id: '242'
  last_name: Haeb-Umbach
citation:
  ama: 'Tran Vu DH, Haeb-Umbach R. Blind speech separation employing directional statistics
    in an Expectation Maximization framework. In: <i>IEEE International Conference
    on Acoustics, Speech and Signal Processing (ICASSP 2010)</i>. ; 2010:241-244.
    doi:<a href="https://doi.org/10.1109/ICASSP.2010.5495994">10.1109/ICASSP.2010.5495994</a>'
  apa: Tran Vu, D. H., &#38; Haeb-Umbach, R. (2010). Blind speech separation employing
    directional statistics in an Expectation Maximization framework. In <i>IEEE International
    Conference on Acoustics, Speech and Signal Processing (ICASSP 2010)</i> (pp. 241–244).
    <a href="https://doi.org/10.1109/ICASSP.2010.5495994">https://doi.org/10.1109/ICASSP.2010.5495994</a>
  bibtex: '@inproceedings{Tran Vu_Haeb-Umbach_2010, title={Blind speech separation
    employing directional statistics in an Expectation Maximization framework}, DOI={<a
    href="https://doi.org/10.1109/ICASSP.2010.5495994">10.1109/ICASSP.2010.5495994</a>},
    booktitle={IEEE International Conference on Acoustics, Speech and Signal Processing
    (ICASSP 2010)}, author={Tran Vu, Dang Hai and Haeb-Umbach, Reinhold}, year={2010},
    pages={241–244} }'
  chicago: Tran Vu, Dang Hai, and Reinhold Haeb-Umbach. “Blind Speech Separation Employing
    Directional Statistics in an Expectation Maximization Framework.” In <i>IEEE International
    Conference on Acoustics, Speech and Signal Processing (ICASSP 2010)</i>, 241–44,
    2010. <a href="https://doi.org/10.1109/ICASSP.2010.5495994">https://doi.org/10.1109/ICASSP.2010.5495994</a>.
  ieee: D. H. Tran Vu and R. Haeb-Umbach, “Blind speech separation employing directional
    statistics in an Expectation Maximization framework,” in <i>IEEE International
    Conference on Acoustics, Speech and Signal Processing (ICASSP 2010)</i>, 2010,
    pp. 241–244.
  mla: Tran Vu, Dang Hai, and Reinhold Haeb-Umbach. “Blind Speech Separation Employing
    Directional Statistics in an Expectation Maximization Framework.” <i>IEEE International
    Conference on Acoustics, Speech and Signal Processing (ICASSP 2010)</i>, 2010,
    pp. 241–44, doi:<a href="https://doi.org/10.1109/ICASSP.2010.5495994">10.1109/ICASSP.2010.5495994</a>.
  short: 'D.H. Tran Vu, R. Haeb-Umbach, in: IEEE International Conference on Acoustics,
    Speech and Signal Processing (ICASSP 2010), 2010, pp. 241–244.'
date_created: 2019-07-12T05:30:40Z
date_updated: 2022-01-06T06:51:12Z
department:
- _id: '54'
doi: 10.1109/ICASSP.2010.5495994
keyword:
- array signal processing
- blind source separation
- blind speech separation
- complex vector space
- complex Watson distribution
- directional statistics
- expectation-maximisation algorithm
- expectation maximization algorithm
- Fourier transform
- Fourier transforms
- generalized sidelobe canceller
- interference suppression
- maximum signal-to-noise ratio beamformer
- microphone signal
- probabilistic model
- spatial aliasing
- spatial beamforming configuration
- speech enhancement
- statistical distributions
language:
- iso: eng
main_file_link:
- open_access: '1'
  url: https://groups.uni-paderborn.de/nt/pubs/2010/DaHa10-2.pdf
oa: '1'
page: 241-244
publication: IEEE International Conference on Acoustics, Speech and Signal Processing
  (ICASSP 2010)
status: public
title: Blind speech separation employing directional statistics in an Expectation
  Maximization framework
type: conference
user_id: '44006'
year: '2010'
...
---
_id: '11931'
abstract:
- lang: eng
  text: The paper is concerned with binaural signal processing for a bimodal human-robot
    interface with hearing and vision. The two microphone signals are processed to
    obtain an enhanced single-channel input signal for the subsequent speech recognizer
    and to localize the acoustic source, an important information for establishing
    a natural human-robot communication. We utilize a robust adaptive algorithm for
    filter-and-sum beamforming (FSB) and extract speaker direction information from
    the resulting FIR filter coefficients. Further, particle filtering is applied
    which conducts a nonlinear Bayesian tracking of speaker movement. Good location
    accuracy can be achieved even in highly reverberant environments. The results
    obtained outperform the conventional generalized cross correlation (GCC) method.
author:
- first_name: Ernst
  full_name: Warsitz, Ernst
  last_name: Warsitz
- first_name: Reinhold
  full_name: Haeb-Umbach, Reinhold
  id: '242'
  last_name: Haeb-Umbach
citation:
  ama: 'Warsitz E, Haeb-Umbach R. Robust speaker direction estimation with particle
    filtering. In: <i>IEEE Workshop on Multimedia Signal Processing (MMSP 2004)</i>.
    ; 2004:367-370. doi:<a href="https://doi.org/10.1109/MMSP.2004.1436569">10.1109/MMSP.2004.1436569</a>'
  apa: Warsitz, E., &#38; Haeb-Umbach, R. (2004). Robust speaker direction estimation
    with particle filtering. In <i>IEEE Workshop on Multimedia Signal Processing (MMSP
    2004)</i> (pp. 367–370). <a href="https://doi.org/10.1109/MMSP.2004.1436569">https://doi.org/10.1109/MMSP.2004.1436569</a>
  bibtex: '@inproceedings{Warsitz_Haeb-Umbach_2004, title={Robust speaker direction
    estimation with particle filtering}, DOI={<a href="https://doi.org/10.1109/MMSP.2004.1436569">10.1109/MMSP.2004.1436569</a>},
    booktitle={IEEE Workshop on Multimedia Signal Processing (MMSP 2004)}, author={Warsitz,
    Ernst and Haeb-Umbach, Reinhold}, year={2004}, pages={367–370} }'
  chicago: Warsitz, Ernst, and Reinhold Haeb-Umbach. “Robust Speaker Direction Estimation
    with Particle Filtering.” In <i>IEEE Workshop on Multimedia Signal Processing
    (MMSP 2004)</i>, 367–70, 2004. <a href="https://doi.org/10.1109/MMSP.2004.1436569">https://doi.org/10.1109/MMSP.2004.1436569</a>.
  ieee: E. Warsitz and R. Haeb-Umbach, “Robust speaker direction estimation with particle
    filtering,” in <i>IEEE Workshop on Multimedia Signal Processing (MMSP 2004)</i>,
    2004, pp. 367–370.
  mla: Warsitz, Ernst, and Reinhold Haeb-Umbach. “Robust Speaker Direction Estimation
    with Particle Filtering.” <i>IEEE Workshop on Multimedia Signal Processing (MMSP
    2004)</i>, 2004, pp. 367–70, doi:<a href="https://doi.org/10.1109/MMSP.2004.1436569">10.1109/MMSP.2004.1436569</a>.
  short: 'E. Warsitz, R. Haeb-Umbach, in: IEEE Workshop on Multimedia Signal Processing
    (MMSP 2004), 2004, pp. 367–370.'
date_created: 2019-07-12T05:31:01Z
date_updated: 2022-01-06T06:51:12Z
department:
- _id: '54'
doi: 10.1109/MMSP.2004.1436569
keyword:
- bimodal human-robot interface
- binaural signal processing
- enhanced single-channel input signal
- filter-and-sum beamforming
- filtering theory
- FIR filter coefficient
- generalized cross correlation method
- microphones
- microphone signal
- nonlinear Bayesian tracking
- particle filtering
- robust adaptive algorithm
- robust speaker direction estimation
- signal processing
- speech enhancement
- speech recognition
- speech recognizer
- user interfaces
language:
- iso: eng
main_file_link:
- open_access: '1'
  url: https://groups.uni-paderborn.de/nt/pubs/2004/WaHa04.pdf
oa: '1'
page: 367-370
publication: IEEE Workshop on Multimedia Signal Processing (MMSP 2004)
status: public
title: Robust speaker direction estimation with particle filtering
type: conference
user_id: '44006'
year: '2004'
...