Jan Vaněk
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Jan Vaněk has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
- Jan Urban
University of South Bohemia in Ceske Budejovice
10 shared publications - Josef V. Psutka
University of West Bohemia in Pilsen
9 shared publications - Aleš Pražák
University of West Bohemia in Pilsen
7 shared publications Josef V. Psutka
University of West Bohemia in Pilsen
7 shared publicationsJosef Psutka
University of West Bohemia in Pilsen
6 shared publications- Josef Michálek
University of West Bohemia
5 shared publications
Latest Publications
- 17 October 2021
- Published by Springer Nature
- p. 85-96
Various DNN-HMM Architectures Used in Acoustic Modeling with Single-Speaker and Single-Channel
- 17 October 2021
- Published by Springer Nature
- p. 85-96
Various DNN-HMM Architectures Used in Acoustic Modeling with Single-Speaker and Single-Channel
- 22 September 2021
- Published by Springer Nature
- p. 553-564
Recognition of Heavily Accented and Emotional Speech of English and Czech Holocaust Survivors Using Various DNN Architectures
- 22 September 2021
- Published by Springer Nature
- p. 553-564
Recognition of Heavily Accented and Emotional Speech of English and Czech Holocaust Survivors Using Various DNN Architectures
- 1 September 2020
- Published by Springer Nature
- p. 465-473
Complexity of the TDNN Acoustic Model with Respect to the HMM Topology
- 1 September 2020
- Published by Springer Nature
- p. 465-473
Complexity of the TDNN Acoustic Model with Respect to the HMM Topology
- 27 September 2019
- Published by Springer Nature
- p. 235-245
Tuning of Acoustic Modeling and Adaptation Technique for a Real Speech Recognition Task
- 27 September 2019
- Published by Springer Nature
- p. 235-245
Tuning of Acoustic Modeling and Adaptation Technique for a Real Speech Recognition Task
- 19 September 2018
- Published by Springer Nature
- p. 79-90
A Comparison of Adaptation Techniques and Recurrent Neural Network Architectures
- 19 September 2018
- Published by Springer Nature
- p. 79-90
A Comparison of Adaptation Techniques and Recurrent Neural Network Architectures
- 8 September 2018
- Published by Springer Nature
- p. 436-444
A Survey of Recent DNN Architectures on the TIMIT Phone Recognition Task
- 8 September 2018
- Published by Springer Nature
- p. 436-444
A Survey of Recent DNN Architectures on the TIMIT Phone Recognition Task
- 25 August 2018
- Published by Springer Nature
- p. 728-736
Recurrent DNNs and Its Ensembles on the TIMIT Phone Recognition Task
- 25 August 2018
- Published by Springer Nature
- p. 728-736
Recurrent DNNs and Its Ensembles on the TIMIT Phone Recognition Task
- Open Access
A Survey of Recent DNN Architectures on the TIMIT Phone Recognition Task
In this survey paper, we have evaluated several recent deep neural network (DNN) architectures on a TIMIT phone recognition task. We chose the TIMIT corpus due to its popularity and broad availability in the community. It also simulates a low-resource scenario that is helpful in minor languages. Also, we prefer the phone recognition task because it is much more sensitive to an acoustic model quality than a large vocabulary continuous speech recognition (LVCSR) task. In recent years, many DNN published papers reported results on TIMIT. However, the reported phone error rates (PERs) were often much higher than a PER of a simple feed-forward (FF) DNN. That was the main motivation of this paper: To provide a baseline DNNs with open-source scripts to easily replicate the baseline results for future papers with lowest possible PERs. According to our knowledge, the best-achieved PER of this survey is better than the best-published PER to date.
Open AccessA Survey of Recent DNN Architectures on the TIMIT Phone Recognition Task
In this survey paper, we have evaluated several recent deep neural network (DNN) architectures on a TIMIT phone recognition task. We chose the TIMIT corpus due to its popularity and broad availability in the community. It also simulates a low-resource scenario that is helpful in minor languages. Also, we prefer the phone recognition task because it is much more sensitive to an acoustic model quality than a large vocabulary continuous speech recognition (LVCSR) task. In recent years, many DNN published papers reported results on TIMIT. However, the reported phone error rates (PERs) were often much higher than a PER of a simple feed-forward (FF) DNN. That was the main motivation of this paper: To provide a baseline DNNs with open-source scripts to easily replicate the baseline results for future papers with lowest possible PERs. According to our knowledge, the best-achieved PER of this survey is better than the best-published PER to date.
- Open Access
Recurrent DNNs and its Ensembles on the TIMIT Phone Recognition Task
In this paper, we have investigated recurrent deep neural networks (DNNs) in combination with regularization techniques as dropout, zoneout, and regularization post-layer. As a benchmark, we chose the TIMIT phone recognition task due to its popularity and broad availability in the community. It also simulates a low-resource scenario that is helpful in minor languages. Also, we prefer the phone recognition task because it is much more sensitive to an acoustic model quality than a large vocabulary continuous speech recognition task. In recent years, recurrent DNNs pushed the error rates in automatic speech recognition down. But, there was no clear winner in proposed architectures. The dropout was used as the regularization technique in most cases, but combination with other regularization techniques together with model ensembles was omitted. However, just an ensemble of recurrent DNNs performed best and achieved an average phone error rate from 10 experiments 14.84 % (minimum 14.69 %) on core test set that is slightly lower then the best-published PER to date, according to our knowledge. Finally, in contrast of the most papers, we published the open-source scripts to easily replicate the results and to help continue the development.
Open AccessRecurrent DNNs and its Ensembles on the TIMIT Phone Recognition Task
In this paper, we have investigated recurrent deep neural networks (DNNs) in combination with regularization techniques as dropout, zoneout, and regularization post-layer. As a benchmark, we chose the TIMIT phone recognition task due to its popularity and broad availability in the community. It also simulates a low-resource scenario that is helpful in minor languages. Also, we prefer the phone recognition task because it is much more sensitive to an acoustic model quality than a large vocabulary continuous speech recognition task. In recent years, recurrent DNNs pushed the error rates in automatic speech recognition down. But, there was no clear winner in proposed architectures. The dropout was used as the regularization technique in most cases, but combination with other regularization techniques together with model ensembles was omitted. However, just an ensemble of recurrent DNNs performed best and achieved an average phone error rate from 10 experiments 14.84 % (minimum 14.69 %) on core test set that is slightly lower then the best-published PER to date, according to our knowledge. Finally, in contrast of the most papers, we published the open-source scripts to easily replicate the results and to help continue the development.
- 27 September 2017
- Published by Springer Nature
- p. 204-214
A Regularization Post Layer: An Additional Way How to Make Deep Neural Networks Robust
- 27 September 2017
- Published by Springer Nature
- p. 204-214
A Regularization Post Layer: An Additional Way How to Make Deep Neural Networks Robust