---
_id: '26770'
abstract:
- lang: eng
text: "Automatic transcription of meetings requires handling of overlapped speech,
which calls for continuous speech separation (CSS) systems. The uPIT criterion
was proposed for utterance-level separation with neural networks and introduces
the constraint that the total number of speakers must not exceed the number of
output channels. When processing meeting-like data in a segment-wise manner, i.e.,
by separating overlapping segments independently and stitching adjacent segments
to continuous output streams, this constraint has to be fulfilled for any segment.
In this contribution, we show that this constraint can be significantly relaxed.
We propose a novel graph-based PIT criterion, which casts the assignment of utterances
to output channels in a graph coloring problem. It only requires that the number
of concurrently active speakers must not exceed the number of output channels.
As a consequence, the system can process an arbitrary number of speakers and arbitrarily
long segments and thus can handle more diverse scenarios.\r\nFurther, the stitching
algorithm for obtaining a consistent output order in neighboring segments is of
less importance and can even be eliminated completely, not the least reducing
the computational effort. Experiments on meeting-style WSJ data show improvements
in recognition performance over using the uPIT criterion. "
author:
- first_name: Thilo
full_name: von Neumann, Thilo
id: '49870'
last_name: von Neumann
orcid: https://orcid.org/0000-0002-7717-8670
- first_name: Keisuke
full_name: Kinoshita, Keisuke
last_name: Kinoshita
- first_name: Christoph
full_name: Boeddeker, Christoph
id: '40767'
last_name: Boeddeker
- first_name: Marc
full_name: Delcroix, Marc
last_name: Delcroix
- first_name: Reinhold
full_name: Haeb-Umbach, Reinhold
id: '242'
last_name: Haeb-Umbach
citation:
ama: 'von Neumann T, Kinoshita K, Boeddeker C, Delcroix M, Haeb-Umbach R. Graph-PIT:
Generalized Permutation Invariant Training for Continuous Separation of Arbitrary
Numbers of Speakers. In: Interspeech 2021. ; 2021. doi:10.21437/interspeech.2021-1177'
apa: 'von Neumann, T., Kinoshita, K., Boeddeker, C., Delcroix, M., & Haeb-Umbach,
R. (2021). Graph-PIT: Generalized Permutation Invariant Training for Continuous
Separation of Arbitrary Numbers of Speakers. Interspeech 2021. Interspeech.
https://doi.org/10.21437/interspeech.2021-1177'
bibtex: '@inproceedings{von Neumann_Kinoshita_Boeddeker_Delcroix_Haeb-Umbach_2021,
title={Graph-PIT: Generalized Permutation Invariant Training for Continuous Separation
of Arbitrary Numbers of Speakers}, DOI={10.21437/interspeech.2021-1177},
booktitle={Interspeech 2021}, author={von Neumann, Thilo and Kinoshita, Keisuke
and Boeddeker, Christoph and Delcroix, Marc and Haeb-Umbach, Reinhold}, year={2021}
}'
chicago: 'Neumann, Thilo von, Keisuke Kinoshita, Christoph Boeddeker, Marc Delcroix,
and Reinhold Haeb-Umbach. “Graph-PIT: Generalized Permutation Invariant Training
for Continuous Separation of Arbitrary Numbers of Speakers.” In Interspeech
2021, 2021. https://doi.org/10.21437/interspeech.2021-1177.'
ieee: 'T. von Neumann, K. Kinoshita, C. Boeddeker, M. Delcroix, and R. Haeb-Umbach,
“Graph-PIT: Generalized Permutation Invariant Training for Continuous Separation
of Arbitrary Numbers of Speakers,” presented at the Interspeech, 2021, doi: 10.21437/interspeech.2021-1177.'
mla: 'von Neumann, Thilo, et al. “Graph-PIT: Generalized Permutation Invariant Training
for Continuous Separation of Arbitrary Numbers of Speakers.” Interspeech 2021,
2021, doi:10.21437/interspeech.2021-1177.'
short: 'T. von Neumann, K. Kinoshita, C. Boeddeker, M. Delcroix, R. Haeb-Umbach,
in: Interspeech 2021, 2021.'
conference:
name: Interspeech
date_created: 2021-10-25T08:50:01Z
date_updated: 2023-11-15T12:14:40Z
ddc:
- '000'
department:
- _id: '54'
doi: 10.21437/interspeech.2021-1177
file:
- access_level: open_access
content_type: video/mp4
creator: tvn
date_created: 2021-12-06T10:39:13Z
date_updated: 2021-12-06T10:48:30Z
file_id: '28327'
file_name: Interspeech 2021 voiceover-002-compressed.mp4
file_size: 9550220
relation: supplementary_material
title: Video for INTERSPEECH 2021
- access_level: open_access
content_type: application/vnd.openxmlformats-officedocument.presentationml.presentation
creator: tvn
date_created: 2021-12-06T10:47:01Z
date_updated: 2021-12-06T10:47:01Z
file_id: '28328'
file_name: Graph-PIT-poster-presentation.pptx
file_size: 1337297
relation: slides
title: Slides from INTERSPEECH 2021
- access_level: open_access
content_type: application/pdf
creator: tvn
date_created: 2021-12-06T10:48:21Z
date_updated: 2021-12-06T10:48:21Z
file_id: '28329'
file_name: INTERSPEECH2021_Graph_PIT.pdf
file_size: 226589
relation: main_file
file_date_updated: 2021-12-06T10:48:30Z
has_accepted_license: '1'
keyword:
- Continuous speech separation
- automatic speech recognition
- overlapped speech
- permutation invariant training
language:
- iso: eng
oa: '1'
project:
- _id: '52'
name: 'PC2: Computing Resources Provided by the Paderborn Center for Parallel Computing'
publication: Interspeech 2021
publication_status: published
quality_controlled: '1'
related_material:
link:
- relation: software
url: https://github.com/fgnt/graph_pit
status: public
title: 'Graph-PIT: Generalized Permutation Invariant Training for Continuous Separation
of Arbitrary Numbers of Speakers'
type: conference
user_id: '49870'
year: '2021'
...
---
_id: '11861'
abstract:
- lang: eng
text: 'In this contribution we present a theoretical and experimental investigation
into the effects of reverberation and noise on features in the logarithmic mel
power spectral domain, an intermediate stage in the computation of the mel frequency
cepstral coefficients, prevalent in automatic speech recognition (ASR). Gaining
insight into the complex interaction between clean speech, noise, and noisy reverberant
speech features is essential for any ASR system to be robust against noise and
reverberation present in distant microphone input signals. The findings are gathered
in a probabilistic formulation of an observation model which may be used in model-based
feature compensation schemes. The proposed observation model extends previous
models in three major directions: First, the contribution of additive background
noise to the observation error is explicitly taken into account. Second, an energy
compensation constant is introduced which ensures an unbiased estimate of the
reverberant speech features, and, third, a recursive variant of the observation
model is developed resulting in reduced computational complexity when used in
model-based feature compensation. The experimental section is used to evaluate
the accuracy of the model and to describe how its parameters can be determined
from test data.'
author:
- first_name: Volker
full_name: Leutnant, Volker
last_name: Leutnant
- 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: Leutnant V, Krueger A, Haeb-Umbach R. A New Observation Model in the Logarithmic
Mel Power Spectral Domain for the Automatic Recognition of Noisy Reverberant Speech.
IEEE/ACM Transactions on Audio, Speech, and Language Processing. 2014;22(1):95-109.
doi:10.1109/TASLP.2013.2285480
apa: Leutnant, V., Krueger, A., & Haeb-Umbach, R. (2014). A New Observation
Model in the Logarithmic Mel Power Spectral Domain for the Automatic Recognition
of Noisy Reverberant Speech. IEEE/ACM Transactions on Audio, Speech, and Language
Processing, 22(1), 95–109. https://doi.org/10.1109/TASLP.2013.2285480
bibtex: '@article{Leutnant_Krueger_Haeb-Umbach_2014, title={A New Observation Model
in the Logarithmic Mel Power Spectral Domain for the Automatic Recognition of
Noisy Reverberant Speech}, volume={22}, DOI={10.1109/TASLP.2013.2285480},
number={1}, journal={IEEE/ACM Transactions on Audio, Speech, and Language Processing},
author={Leutnant, Volker and Krueger, Alexander and Haeb-Umbach, Reinhold}, year={2014},
pages={95–109} }'
chicago: 'Leutnant, Volker, Alexander Krueger, and Reinhold Haeb-Umbach. “A New
Observation Model in the Logarithmic Mel Power Spectral Domain for the Automatic
Recognition of Noisy Reverberant Speech.” IEEE/ACM Transactions on Audio, Speech,
and Language Processing 22, no. 1 (2014): 95–109. https://doi.org/10.1109/TASLP.2013.2285480.'
ieee: V. Leutnant, A. Krueger, and R. Haeb-Umbach, “A New Observation Model in the
Logarithmic Mel Power Spectral Domain for the Automatic Recognition of Noisy Reverberant
Speech,” IEEE/ACM Transactions on Audio, Speech, and Language Processing,
vol. 22, no. 1, pp. 95–109, 2014.
mla: Leutnant, Volker, et al. “A New Observation Model in the Logarithmic Mel Power
Spectral Domain for the Automatic Recognition of Noisy Reverberant Speech.” IEEE/ACM
Transactions on Audio, Speech, and Language Processing, vol. 22, no. 1, 2014,
pp. 95–109, doi:10.1109/TASLP.2013.2285480.
short: V. Leutnant, A. Krueger, R. Haeb-Umbach, IEEE/ACM Transactions on Audio,
Speech, and Language Processing 22 (2014) 95–109.
date_created: 2019-07-12T05:29:41Z
date_updated: 2022-01-06T06:51:11Z
department:
- _id: '54'
doi: 10.1109/TASLP.2013.2285480
intvolume: ' 22'
issue: '1'
keyword:
- computational complexity
- reverberation
- speech recognition
- automatic speech recognition
- background noise
- clean speech
- computational complexity
- energy compensation
- logarithmic mel power spectral domain
- mel frequency cepstral coefficients
- microphone input signals
- model-based feature compensation schemes
- noisy reverberant speech automatic recognition
- noisy reverberant speech features
- reverberation
- Atmospheric modeling
- Computational modeling
- Noise
- Noise measurement
- Reverberation
- Speech
- Vectors
- Model-based feature compensation
- observation model for reverberant and noisy speech
- recursive observation model
- robust automatic speech recognition
language:
- iso: eng
page: 95-109
publication: IEEE/ACM Transactions on Audio, Speech, and Language Processing
publication_identifier:
issn:
- 2329-9290
status: public
title: A New Observation Model in the Logarithmic Mel Power Spectral Domain for the
Automatic Recognition of Noisy Reverberant Speech
type: journal_article
user_id: '44006'
volume: 22
year: '2014'
...
---
_id: '11716'
abstract:
- lang: eng
text: The accuracy of automatic speech recognition systems in noisy and reverberant
environments can be improved notably by exploiting the uncertainty of the estimated
speech features using so-called uncertainty-of-observation techniques. In this
paper, we introduce a new Bayesian decision rule that can serve as a mathematical
framework from which both known and new uncertainty-of-observation techniques
can be either derived or approximated. The new decision rule in its direct form
leads to the new significance decoding approach for Gaussian mixture models, which
results in better performance compared to standard uncertainty-of-observation
techniques in different additive and convolutive noise scenarios.
author:
- first_name: Ahmed H.
full_name: Abdelaziz, Ahmed H.
last_name: Abdelaziz
- first_name: Steffen
full_name: Zeiler, Steffen
last_name: Zeiler
- first_name: Dorothea
full_name: Kolossa, Dorothea
last_name: Kolossa
- first_name: Volker
full_name: Leutnant, Volker
last_name: Leutnant
- first_name: Reinhold
full_name: Haeb-Umbach, Reinhold
id: '242'
last_name: Haeb-Umbach
citation:
ama: 'Abdelaziz AH, Zeiler S, Kolossa D, Leutnant V, Haeb-Umbach R. GMM-based significance
decoding. In: Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International
Conference On. ; 2013:6827-6831. doi:10.1109/ICASSP.2013.6638984'
apa: Abdelaziz, A. H., Zeiler, S., Kolossa, D., Leutnant, V., & Haeb-Umbach,
R. (2013). GMM-based significance decoding. In Acoustics, Speech and Signal
Processing (ICASSP), 2013 IEEE International Conference on (pp. 6827–6831).
https://doi.org/10.1109/ICASSP.2013.6638984
bibtex: '@inproceedings{Abdelaziz_Zeiler_Kolossa_Leutnant_Haeb-Umbach_2013, title={GMM-based
significance decoding}, DOI={10.1109/ICASSP.2013.6638984},
booktitle={Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International
Conference on}, author={Abdelaziz, Ahmed H. and Zeiler, Steffen and Kolossa, Dorothea
and Leutnant, Volker and Haeb-Umbach, Reinhold}, year={2013}, pages={6827–6831}
}'
chicago: Abdelaziz, Ahmed H., Steffen Zeiler, Dorothea Kolossa, Volker Leutnant,
and Reinhold Haeb-Umbach. “GMM-Based Significance Decoding.” In Acoustics,
Speech and Signal Processing (ICASSP), 2013 IEEE International Conference On,
6827–31, 2013. https://doi.org/10.1109/ICASSP.2013.6638984.
ieee: A. H. Abdelaziz, S. Zeiler, D. Kolossa, V. Leutnant, and R. Haeb-Umbach, “GMM-based
significance decoding,” in Acoustics, Speech and Signal Processing (ICASSP),
2013 IEEE International Conference on, 2013, pp. 6827–6831.
mla: Abdelaziz, Ahmed H., et al. “GMM-Based Significance Decoding.” Acoustics,
Speech and Signal Processing (ICASSP), 2013 IEEE International Conference On,
2013, pp. 6827–31, doi:10.1109/ICASSP.2013.6638984.
short: 'A.H. Abdelaziz, S. Zeiler, D. Kolossa, V. Leutnant, R. Haeb-Umbach, in:
Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference
On, 2013, pp. 6827–6831.'
date_created: 2019-07-12T05:26:53Z
date_updated: 2022-01-06T06:51:07Z
department:
- _id: '54'
doi: 10.1109/ICASSP.2013.6638984
keyword:
- Bayes methods
- Gaussian processes
- convolution
- decision theory
- decoding
- noise
- reverberation
- speech coding
- speech recognition
- Bayesian decision rule
- GMM
- Gaussian mixture models
- additive noise scenarios
- automatic speech recognition systems
- convolutive noise scenarios
- decoding approach
- mathematical framework
- reverberant environments
- significance decoding
- speech feature estimation
- uncertainty-of-observation techniques
- Hidden Markov models
- Maximum likelihood decoding
- Noise
- Speech
- Speech recognition
- Uncertainty
- Uncertainty-of-observation
- modified imputation
- noise robust speech recognition
- significance decoding
- uncertainty decoding
language:
- iso: eng
page: 6827-6831
publication: Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International
Conference on
publication_identifier:
issn:
- 1520-6149
status: public
title: GMM-based significance decoding
type: conference
user_id: '44006'
year: '2013'
...
---
_id: '11862'
abstract:
- lang: eng
text: In this contribution we extend a previously proposed Bayesian approach for
the enhancement of reverberant logarithmic mel power spectral coefficients for
robust automatic speech recognition to the additional compensation of background
noise. A recently proposed observation model is employed whose time-variant observation
error statistics are obtained as a side product of the inference of the a posteriori
probability density function of the clean speech feature vectors. Further a reduction
of the computational effort and the memory requirements are achieved by using
a recursive formulation of the observation model. The performance of the proposed
algorithms is first experimentally studied on a connected digits recognition task
with artificially created noisy reverberant data. It is shown that the use of
the time-variant observation error model leads to a significant error rate reduction
at low signal-to-noise ratios compared to a time-invariant model. Further experiments
were conducted on a 5000 word task recorded in a reverberant and noisy environment.
A significant word error rate reduction was obtained demonstrating the effectiveness
of the approach on real-world data.
author:
- first_name: Volker
full_name: Leutnant, Volker
last_name: Leutnant
- 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: Leutnant V, Krueger A, Haeb-Umbach R. Bayesian Feature Enhancement for Reverberation
and Noise Robust Speech Recognition. IEEE Transactions on Audio, Speech, and
Language Processing. 2013;21(8):1640-1652. doi:10.1109/TASL.2013.2258013
apa: Leutnant, V., Krueger, A., & Haeb-Umbach, R. (2013). Bayesian Feature Enhancement
for Reverberation and Noise Robust Speech Recognition. IEEE Transactions on
Audio, Speech, and Language Processing, 21(8), 1640–1652. https://doi.org/10.1109/TASL.2013.2258013
bibtex: '@article{Leutnant_Krueger_Haeb-Umbach_2013, title={Bayesian Feature Enhancement
for Reverberation and Noise Robust Speech Recognition}, volume={21}, DOI={10.1109/TASL.2013.2258013},
number={8}, journal={IEEE Transactions on Audio, Speech, and Language Processing},
author={Leutnant, Volker and Krueger, Alexander and Haeb-Umbach, Reinhold}, year={2013},
pages={1640–1652} }'
chicago: 'Leutnant, Volker, Alexander Krueger, and Reinhold Haeb-Umbach. “Bayesian
Feature Enhancement for Reverberation and Noise Robust Speech Recognition.” IEEE
Transactions on Audio, Speech, and Language Processing 21, no. 8 (2013): 1640–52.
https://doi.org/10.1109/TASL.2013.2258013.'
ieee: V. Leutnant, A. Krueger, and R. Haeb-Umbach, “Bayesian Feature Enhancement
for Reverberation and Noise Robust Speech Recognition,” IEEE Transactions on
Audio, Speech, and Language Processing, vol. 21, no. 8, pp. 1640–1652, 2013.
mla: Leutnant, Volker, et al. “Bayesian Feature Enhancement for Reverberation and
Noise Robust Speech Recognition.” IEEE Transactions on Audio, Speech, and Language
Processing, vol. 21, no. 8, 2013, pp. 1640–52, doi:10.1109/TASL.2013.2258013.
short: V. Leutnant, A. Krueger, R. Haeb-Umbach, IEEE Transactions on Audio, Speech,
and Language Processing 21 (2013) 1640–1652.
date_created: 2019-07-12T05:29:42Z
date_updated: 2022-01-06T06:51:11Z
department:
- _id: '54'
doi: 10.1109/TASL.2013.2258013
intvolume: ' 21'
issue: '8'
keyword:
- Bayes methods
- compensation
- error statistics
- reverberation
- speech recognition
- Bayesian feature enhancement
- background noise
- clean speech feature vectors
- compensation
- connected digits recognition task
- error statistics
- memory requirements
- noisy reverberant data
- posteriori probability density function
- recursive formulation
- reverberant logarithmic mel power spectral coefficients
- robust automatic speech recognition
- signal-to-noise ratios
- time-variant observation
- word error rate reduction
- Robust automatic speech recognition
- model-based Bayesian feature enhancement
- observation model for reverberant and noisy speech
- recursive observation model
language:
- iso: eng
page: 1640-1652
publication: IEEE Transactions on Audio, Speech, and Language Processing
status: public
title: Bayesian Feature Enhancement for Reverberation and Noise Robust Speech Recognition
type: journal_article
user_id: '44006'
volume: 21
year: '2013'
...
---
_id: '11864'
abstract:
- lang: eng
text: In this work, an observation model for the joint compensation of noise and
reverberation in the logarithmic mel power spectral density domain is considered.
It relates the features of the noisy reverberant speech to those of the non-reverberant
speech and the noise. In contrast to enhancement of features only corrupted by
reverberation (reverberant features), enhancement of noisy reverberant features
requires a more sophisticated model for the error introduced by the proposed observation
model. In a first consideration, it will be shown that this error is highly dependent
on the instantaneous ratio of the power of reverberant speech to the power of
the noise and, moreover, sensitive to the phase between reverberant speech and
noise in the short-time discrete Fourier domain. Afterwards, a statistically motivated
approach will be presented allowing for the model of the observation error to
be inferred from the error model previously used for the reverberation only case.
Finally, the developed observation error model will be utilized in a Bayesian
feature enhancement scheme, leading to improvements in word accuracy on the AURORA5
database.
author:
- first_name: Volker
full_name: Leutnant, Volker
last_name: Leutnant
- 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: 'Leutnant V, Krueger A, Haeb-Umbach R. A Statistical Observation Model For
Noisy Reverberant Speech Features and its Application to Robust ASR. In: Signal
Processing, Communications and Computing (ICSPCC), 2012 IEEE International Conference
On. ; 2012.'
apa: Leutnant, V., Krueger, A., & Haeb-Umbach, R. (2012). A Statistical Observation
Model For Noisy Reverberant Speech Features and its Application to Robust ASR.
In Signal Processing, Communications and Computing (ICSPCC), 2012 IEEE International
Conference on.
bibtex: '@inproceedings{Leutnant_Krueger_Haeb-Umbach_2012, title={A Statistical
Observation Model For Noisy Reverberant Speech Features and its Application to
Robust ASR}, booktitle={Signal Processing, Communications and Computing (ICSPCC),
2012 IEEE International Conference on}, author={Leutnant, Volker and Krueger,
Alexander and Haeb-Umbach, Reinhold}, year={2012} }'
chicago: Leutnant, Volker, Alexander Krueger, and Reinhold Haeb-Umbach. “A Statistical
Observation Model For Noisy Reverberant Speech Features and Its Application to
Robust ASR.” In Signal Processing, Communications and Computing (ICSPCC), 2012
IEEE International Conference On, 2012.
ieee: V. Leutnant, A. Krueger, and R. Haeb-Umbach, “A Statistical Observation Model
For Noisy Reverberant Speech Features and its Application to Robust ASR,” in Signal
Processing, Communications and Computing (ICSPCC), 2012 IEEE International Conference
on, 2012.
mla: Leutnant, Volker, et al. “A Statistical Observation Model For Noisy Reverberant
Speech Features and Its Application to Robust ASR.” Signal Processing, Communications
and Computing (ICSPCC), 2012 IEEE International Conference On, 2012.
short: 'V. Leutnant, A. Krueger, R. Haeb-Umbach, in: Signal Processing, Communications
and Computing (ICSPCC), 2012 IEEE International Conference On, 2012.'
date_created: 2019-07-12T05:29:44Z
date_updated: 2022-01-06T06:51:11Z
department:
- _id: '54'
keyword:
- Robust Automatic Speech Recognition
- Bayesian feature enhancement
- observation model for reverberant and noisy speech
language:
- iso: eng
main_file_link:
- open_access: '1'
url: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6335731
oa: '1'
publication: Signal Processing, Communications and Computing (ICSPCC), 2012 IEEE International
Conference on
status: public
title: A Statistical Observation Model For Noisy Reverberant Speech Features and its
Application to Robust ASR
type: conference
user_id: '44006'
year: '2012'
...
---
_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. IEEE Transactions on Audio, Speech, and Language Processing.
2010;18(7):1692-1707. doi:10.1109/TASL.2010.2049684
apa: Krueger, A., & Haeb-Umbach, R. (2010). Model-Based Feature Enhancement
for Reverberant Speech Recognition. IEEE Transactions on Audio, Speech, and
Language Processing, 18(7), 1692–1707. https://doi.org/10.1109/TASL.2010.2049684
bibtex: '@article{Krueger_Haeb-Umbach_2010, title={Model-Based Feature Enhancement
for Reverberant Speech Recognition}, volume={18}, DOI={10.1109/TASL.2010.2049684},
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.” IEEE Transactions on Audio, Speech, and
Language Processing 18, no. 7 (2010): 1692–1707. https://doi.org/10.1109/TASL.2010.2049684.'
ieee: A. Krueger and R. Haeb-Umbach, “Model-Based Feature Enhancement for Reverberant
Speech Recognition,” IEEE Transactions on Audio, Speech, and Language Processing,
vol. 18, no. 7, pp. 1692–1707, 2010.
mla: Krueger, Alexander, and Reinhold Haeb-Umbach. “Model-Based Feature Enhancement
for Reverberant Speech Recognition.” IEEE Transactions on Audio, Speech, and
Language Processing, vol. 18, no. 7, 2010, pp. 1692–707, doi:10.1109/TASL.2010.2049684.
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: '11937'
abstract:
- lang: eng
text: In automatic speech recognition, hidden Markov models (HMMs) are commonly
used for speech decoding, while switching linear dynamic models (SLDMs) can be
employed for a preceding model-based speech feature enhancement. In this paper,
these model types are combined in order to obtain a novel iterative speech feature
enhancement and recognition architecture. It is shown that speech feature enhancement
with SLDMs can be improved by feeding back information from the HMM to the enhancement
stage. Two different feedback structures are derived. In the first, the posteriors
of the HMM states are used to control the model probabilities of the SLDMs, while
in the second they are employed to directly influence the estimate of the speech
feature distribution. Both approaches lead to improvements in recognition accuracy
both on the AURORA2 and AURORA4 databases compared to non-iterative speech feature
enhancement with SLDMs. It is also shown that a combination with uncertainty decoding
further enhances performance.
author:
- first_name: Stefan
full_name: Windmann, Stefan
last_name: Windmann
- first_name: Reinhold
full_name: Haeb-Umbach, Reinhold
id: '242'
last_name: Haeb-Umbach
citation:
ama: Windmann S, Haeb-Umbach R. Approaches to Iterative Speech Feature Enhancement
and Recognition. IEEE Transactions on Audio, Speech, and Language Processing.
2009;17(5):974-984. doi:10.1109/TASL.2009.2014894
apa: Windmann, S., & Haeb-Umbach, R. (2009). Approaches to Iterative Speech
Feature Enhancement and Recognition. IEEE Transactions on Audio, Speech, and
Language Processing, 17(5), 974–984. https://doi.org/10.1109/TASL.2009.2014894
bibtex: '@article{Windmann_Haeb-Umbach_2009, title={Approaches to Iterative Speech
Feature Enhancement and Recognition}, volume={17}, DOI={10.1109/TASL.2009.2014894},
number={5}, journal={IEEE Transactions on Audio, Speech, and Language Processing},
author={Windmann, Stefan and Haeb-Umbach, Reinhold}, year={2009}, pages={974–984}
}'
chicago: 'Windmann, Stefan, and Reinhold Haeb-Umbach. “Approaches to Iterative Speech
Feature Enhancement and Recognition.” IEEE Transactions on Audio, Speech, and
Language Processing 17, no. 5 (2009): 974–84. https://doi.org/10.1109/TASL.2009.2014894.'
ieee: S. Windmann and R. Haeb-Umbach, “Approaches to Iterative Speech Feature Enhancement
and Recognition,” IEEE Transactions on Audio, Speech, and Language Processing,
vol. 17, no. 5, pp. 974–984, 2009.
mla: Windmann, Stefan, and Reinhold Haeb-Umbach. “Approaches to Iterative Speech
Feature Enhancement and Recognition.” IEEE Transactions on Audio, Speech, and
Language Processing, vol. 17, no. 5, 2009, pp. 974–84, doi:10.1109/TASL.2009.2014894.
short: S. Windmann, R. Haeb-Umbach, IEEE Transactions on Audio, Speech, and Language
Processing 17 (2009) 974–984.
date_created: 2019-07-12T05:31:08Z
date_updated: 2022-01-06T06:51:12Z
department:
- _id: '54'
doi: 10.1109/TASL.2009.2014894
intvolume: ' 17'
issue: '5'
keyword:
- AURORA2 databases
- AURORA4 databases
- automatic speech recognition
- feedback structures
- hidden Markov models
- HMM
- iterative methods
- iterative speech feature enhancement
- model probabilities
- speech decoding
- speech enhancement
- speech feature distribution
- speech recognition
- switching linear dynamic models
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://groups.uni-paderborn.de/nt/pubs/2009/WiHa09-1.pdf
oa: '1'
page: 974-984
publication: IEEE Transactions on Audio, Speech, and Language Processing
status: public
title: Approaches to Iterative Speech Feature Enhancement and Recognition
type: journal_article
user_id: '44006'
volume: 17
year: '2009'
...
---
_id: '11938'
abstract:
- lang: eng
text: In this paper, parameter estimation of a state-space model of noise or noisy
speech cepstra is investigated. A blockwise EM algorithm is derived for the estimation
of the state and observation noise covariance from noise-only input data. It is
supposed to be used during the offline training mode of a speech recognizer. Further
a sequential online EM algorithm is developed to adapt the observation noise covariance
on noisy speech cepstra at its input. The estimated parameters are then used in
model-based speech feature enhancement for noise-robust automatic speech recognition.
Experiments on the AURORA4 database lead to improved recognition results with
a linear state model compared to the assumption of stationary noise.
author:
- first_name: Stefan
full_name: Windmann, Stefan
last_name: Windmann
- first_name: Reinhold
full_name: Haeb-Umbach, Reinhold
id: '242'
last_name: Haeb-Umbach
citation:
ama: Windmann S, Haeb-Umbach R. Parameter Estimation of a State-Space Model of Noise
for Robust Speech Recognition. IEEE Transactions on Audio, Speech, and Language
Processing. 2009;17(8):1577-1590. doi:10.1109/TASL.2009.2023172
apa: Windmann, S., & Haeb-Umbach, R. (2009). Parameter Estimation of a State-Space
Model of Noise for Robust Speech Recognition. IEEE Transactions on Audio, Speech,
and Language Processing, 17(8), 1577–1590. https://doi.org/10.1109/TASL.2009.2023172
bibtex: '@article{Windmann_Haeb-Umbach_2009, title={Parameter Estimation of a State-Space
Model of Noise for Robust Speech Recognition}, volume={17}, DOI={10.1109/TASL.2009.2023172},
number={8}, journal={IEEE Transactions on Audio, Speech, and Language Processing},
author={Windmann, Stefan and Haeb-Umbach, Reinhold}, year={2009}, pages={1577–1590}
}'
chicago: 'Windmann, Stefan, and Reinhold Haeb-Umbach. “Parameter Estimation of a
State-Space Model of Noise for Robust Speech Recognition.” IEEE Transactions
on Audio, Speech, and Language Processing 17, no. 8 (2009): 1577–90. https://doi.org/10.1109/TASL.2009.2023172.'
ieee: S. Windmann and R. Haeb-Umbach, “Parameter Estimation of a State-Space Model
of Noise for Robust Speech Recognition,” IEEE Transactions on Audio, Speech,
and Language Processing, vol. 17, no. 8, pp. 1577–1590, 2009.
mla: Windmann, Stefan, and Reinhold Haeb-Umbach. “Parameter Estimation of a State-Space
Model of Noise for Robust Speech Recognition.” IEEE Transactions on Audio,
Speech, and Language Processing, vol. 17, no. 8, 2009, pp. 1577–90, doi:10.1109/TASL.2009.2023172.
short: S. Windmann, R. Haeb-Umbach, IEEE Transactions on Audio, Speech, and Language
Processing 17 (2009) 1577–1590.
date_created: 2019-07-12T05:31:09Z
date_updated: 2022-01-06T06:51:12Z
department:
- _id: '54'
doi: 10.1109/TASL.2009.2023172
intvolume: ' 17'
issue: '8'
keyword:
- AURORA4 database
- blockwise EM algorithm
- covariance analysis
- linear state model
- noise covariance
- noise-robust automatic speech recognition
- noisy speech cepstra
- offline training mode
- parameter estimation
- speech recognition
- speech recognition equipment
- speech recognizer
- state-space methods
- state-space model
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://groups.uni-paderborn.de/nt/pubs/2009/WiHa09-2.pdf
oa: '1'
page: 1577-1590
publication: IEEE Transactions on Audio, Speech, and Language Processing
status: public
title: Parameter Estimation of a State-Space Model of Noise for Robust Speech Recognition
type: journal_article
user_id: '44006'
volume: 17
year: '2009'
...
---
_id: '11820'
abstract:
- lang: eng
text: In this paper, we derive an uncertainty decoding rule for automatic speech
recognition (ASR), which accounts for both corrupted observations and inter-frame
correlation. The conditional independence assumption, prevalent in hidden Markov
model-based ASR, is relaxed to obtain a clean speech posterior that is conditioned
on the complete observed feature vector sequence. This is a more informative posterior
than one conditioned only on the current observation. The novel decoding is used
to obtain a transmission-error robust remote ASR system, where the speech capturing
unit is connected to the decoder via an error-prone communication network. We
show how the clean speech posterior can be computed for communication links being
characterized by either bit errors or packet loss. Recognition results are presented
for both distributed and network speech recognition, where in the latter case
common voice-over-IP codecs are employed.
author:
- first_name: Valentin
full_name: Ion, Valentin
last_name: Ion
- first_name: Reinhold
full_name: Haeb-Umbach, Reinhold
id: '242'
last_name: Haeb-Umbach
citation:
ama: Ion V, Haeb-Umbach R. A Novel Uncertainty Decoding Rule With Applications to
Transmission Error Robust Speech Recognition. IEEE Transactions on Audio, Speech,
and Language Processing. 2008;16(5):1047-1060. doi:10.1109/TASL.2008.925879
apa: Ion, V., & Haeb-Umbach, R. (2008). A Novel Uncertainty Decoding Rule With
Applications to Transmission Error Robust Speech Recognition. IEEE Transactions
on Audio, Speech, and Language Processing, 16(5), 1047–1060. https://doi.org/10.1109/TASL.2008.925879
bibtex: '@article{Ion_Haeb-Umbach_2008, title={A Novel Uncertainty Decoding Rule
With Applications to Transmission Error Robust Speech Recognition}, volume={16},
DOI={10.1109/TASL.2008.925879},
number={5}, journal={IEEE Transactions on Audio, Speech, and Language Processing},
author={Ion, Valentin and Haeb-Umbach, Reinhold}, year={2008}, pages={1047–1060}
}'
chicago: 'Ion, Valentin, and Reinhold Haeb-Umbach. “A Novel Uncertainty Decoding
Rule With Applications to Transmission Error Robust Speech Recognition.” IEEE
Transactions on Audio, Speech, and Language Processing 16, no. 5 (2008): 1047–60.
https://doi.org/10.1109/TASL.2008.925879.'
ieee: V. Ion and R. Haeb-Umbach, “A Novel Uncertainty Decoding Rule With Applications
to Transmission Error Robust Speech Recognition,” IEEE Transactions on Audio,
Speech, and Language Processing, vol. 16, no. 5, pp. 1047–1060, 2008.
mla: Ion, Valentin, and Reinhold Haeb-Umbach. “A Novel Uncertainty Decoding Rule
With Applications to Transmission Error Robust Speech Recognition.” IEEE Transactions
on Audio, Speech, and Language Processing, vol. 16, no. 5, 2008, pp. 1047–60,
doi:10.1109/TASL.2008.925879.
short: V. Ion, R. Haeb-Umbach, IEEE Transactions on Audio, Speech, and Language
Processing 16 (2008) 1047–1060.
date_created: 2019-07-12T05:28:53Z
date_updated: 2022-01-06T06:51:10Z
department:
- _id: '54'
doi: 10.1109/TASL.2008.925879
intvolume: ' 16'
issue: '5'
keyword:
- automatic speech recognition
- bit errors
- codecs
- communication links
- corrupted observations
- decoding
- distributed speech recognition
- error-prone communication network
- feature vector sequence
- hidden Markov model-based ASR
- hidden Markov models
- inter-frame correlation
- Internet telephony
- network speech recognition
- packet loss
- speech posterior
- speech recognition
- transmission error robust speech recognition
- uncertainty decoding
- voice-over-IP codecs
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://groups.uni-paderborn.de/nt/pubs/2008/IoHa08-1.pdf
oa: '1'
page: 1047-1060
publication: IEEE Transactions on Audio, Speech, and Language Processing
status: public
title: A Novel Uncertainty Decoding Rule With Applications to Transmission Error Robust
Speech Recognition
type: journal_article
user_id: '44006'
volume: 16
year: '2008'
...
---
_id: '11869'
abstract:
- lang: eng
text: Amongst several data driven approaches for designing filters for the time
sequence of spectral parameters, the linear discriminant analysis (LDA) based
method has been proposed for automatic speech recognition. Here we apply LDA-based
filter design to cepstral features, which better match the inherent assumption
of this method that feature vector components are uncorrelated. Extensive recognition
experiments have been conducted both on the standard TIMIT phone recognition task
and on a proprietary 130-words command word task under various adverse environmental
conditions, including reverberant data with real-life room impulse responses and
data processed by acoustic echo cancellation algorithms. Significant error rate
reductions have been achieved when applying the novel long-range feature filters
compared to standard approaches employing cepstral mean normalization and delta
and delta-delta features, in particular when facing acoustic echo cancellation
scenarios and room reverberation. For example, the phone accuracy on reverberated
TIMIT data could be increased from 50.7\% to 56.0\%
author:
- first_name: M.
full_name: Lieb, M.
last_name: Lieb
- first_name: Reinhold
full_name: Haeb-Umbach, Reinhold
id: '242'
last_name: Haeb-Umbach
citation:
ama: 'Lieb M, Haeb-Umbach R. LDA derived cepstral trajectory filters in adverse
environmental conditions. In: IEEE International Conference on Acoustics, Speech,
and Signal Processing (ICASSP 2000). Vol 2. ; 2000:II1105-II1108 vol.2. doi:10.1109/ICASSP.2000.859157'
apa: Lieb, M., & Haeb-Umbach, R. (2000). LDA derived cepstral trajectory filters
in adverse environmental conditions. In IEEE International Conference on Acoustics,
Speech, and Signal Processing (ICASSP 2000) (Vol. 2, pp. II1105-II1108 vol.2).
https://doi.org/10.1109/ICASSP.2000.859157
bibtex: '@inproceedings{Lieb_Haeb-Umbach_2000, title={LDA derived cepstral trajectory
filters in adverse environmental conditions}, volume={2}, DOI={10.1109/ICASSP.2000.859157},
booktitle={IEEE International Conference on Acoustics, Speech, and Signal Processing
(ICASSP 2000)}, author={Lieb, M. and Haeb-Umbach, Reinhold}, year={2000}, pages={II1105-II1108
vol.2} }'
chicago: Lieb, M., and Reinhold Haeb-Umbach. “LDA Derived Cepstral Trajectory Filters
in Adverse Environmental Conditions.” In IEEE International Conference on Acoustics,
Speech, and Signal Processing (ICASSP 2000), 2:II1105-II1108 vol.2, 2000.
https://doi.org/10.1109/ICASSP.2000.859157.
ieee: M. Lieb and R. Haeb-Umbach, “LDA derived cepstral trajectory filters in adverse
environmental conditions,” in IEEE International Conference on Acoustics, Speech,
and Signal Processing (ICASSP 2000), 2000, vol. 2, pp. II1105-II1108 vol.2.
mla: Lieb, M., and Reinhold Haeb-Umbach. “LDA Derived Cepstral Trajectory Filters
in Adverse Environmental Conditions.” IEEE International Conference on Acoustics,
Speech, and Signal Processing (ICASSP 2000), vol. 2, 2000, pp. II1105-II1108
vol.2, doi:10.1109/ICASSP.2000.859157.
short: 'M. Lieb, R. Haeb-Umbach, in: IEEE International Conference on Acoustics,
Speech, and Signal Processing (ICASSP 2000), 2000, pp. II1105-II1108 vol.2.'
date_created: 2019-07-12T05:29:50Z
date_updated: 2022-01-06T06:51:11Z
department:
- _id: '54'
doi: 10.1109/ICASSP.2000.859157
intvolume: ' 2'
keyword:
- acoustic echo cancellation algorithms
- adverse environmental conditions
- automatic speech recognition
- cepstral analysis
- cepstral features
- cepstral mean normalization
- command word task
- delta-delta features
- delta features
- echo suppression
- error rate reductions
- feature vector components
- FIR filters
- LDA derived cepstral trajectory filters
- linear discriminant analysis
- long-range feature filters
- phone accuracy
- real-life room impulse responses
- reverberant data
- spectral parameters
- speech recognition
- standard TIMIT phone recognition task
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://groups.uni-paderborn.de/nt/pubs/2000/LiHa00.pdf
oa: '1'
page: II1105-II1108 vol.2
publication: IEEE International Conference on Acoustics, Speech, and Signal Processing
(ICASSP 2000)
status: public
title: LDA derived cepstral trajectory filters in adverse environmental conditions
type: conference
user_id: '44006'
volume: 2
year: '2000'
...