Hoon Chung
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Hoon Chung has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Ho-Young Jung
Department of Artificial Intelligence, Kyungpook National University, Daegu 41566, Republic of Korea
2 shared publications - Jeon Gue Park
Electronics and Telecommunications Research Institute, Daejeon 34129, Korea
2 shared publications - June-Woo Kim
GIST InnoCORE AI-Nano Convergence Institute for Early Detection of Neurodegenerative Disease, Gwangju Institute of Science and Technology, Gwangju, Korea
2 shared publications - Sung Joo Lee
Electronics and Telecommunications Research Institute, Daejeon 34129, Korea
2 shared publications - Yun-Keun Lee
Electronics and Telecommunications Research Institute (ETRI)
2 shared publications - Byung Ok Kang
Integrated Intelligence Research Section Electronics and Telecommunications Research Institute Daejeon Republic of Korea
1 shared publication
Latest Publications
- Open Access
- 5 August 2023
- Published by MDPI AG in Mathematics
- Vol. 11 (15) , 3418
Spectral Salt-and-Pepper Patch Masking for Self-Supervised Speech Representation Learning
Recent advanced systems in the speech recognition domain use large Transformer neural networks that have been pretrained on massive speech data. General methods in the deep learning area have been frequently shared across various domains, and the Transformer model can also be used effectively across speech and image. In this paper, we introduce a novel masking method for self-supervised speech representation learning with salt-and-pepper (S&P) mask which is commonly used in computer vision. The proposed scheme includes consecutive quadrilateral-shaped S&P patches randomly contaminating the input speech spectrum. Furthermore, we modify the standard S&P mask to make it appropriate for the speech domain. In order to validate the effect of the proposed spectral S&P patch masking for the self-supervised representation learning approach, we conduct the pretraining and downstream experiments with two languages, English and Korean. To this end, we pretrain the speech representation model using each dataset and evaluate the pretrained models for feature extraction and fine-tuning performance on varying downstream tasks, respectively. The experimental outcomes clearly illustrate that the proposed spectral S&P patch masking is effective for various downstream tasks when combined with the conventional masking methods.
- 5 August 2023
- Published by MDPI AG in Mathematics
- Vol. 11 (15) , 3418
Open AccessSpectral Salt-and-Pepper Patch Masking for Self-Supervised Speech Representation Learning
Recent advanced systems in the speech recognition domain use large Transformer neural networks that have been pretrained on massive speech data. General methods in the deep learning area have been frequently shared across various domains, and the Transformer model can also be used effectively across speech and image. In this paper, we introduce a novel masking method for self-supervised speech representation learning with salt-and-pepper (S&P) mask which is commonly used in computer vision. The proposed scheme includes consecutive quadrilateral-shaped S&P patches randomly contaminating the input speech spectrum. Furthermore, we modify the standard S&P mask to make it appropriate for the speech domain. In order to validate the effect of the proposed spectral S&P patch masking for the self-supervised representation learning approach, we conduct the pretraining and downstream experiments with two languages, English and Korean. To this end, we pretrain the speech representation model using each dataset and evaluate the pretrained models for feature extraction and fine-tuning performance on varying downstream tasks, respectively. The experimental outcomes clearly illustrate that the proposed spectral S&P patch masking is effective for various downstream tasks when combined with the conventional masking methods.
- Open Access
- 26 January 2023
- Published by MDPI AG in Mathematics
- Vol. 11 (3) , 622
Unsupervised Representation Learning with Task-Agnostic Feature Masking for Robust End-to-End Speech Recognition
Unsupervised learning-based approaches for training speech vector representations (SVR) have recently been widely applied. While pretrained SVR models excel in relatively clean automatic speech recognition (ASR) tasks, such as those recorded in laboratory environments, they are still insufficient for practical applications with various types of noise, intonation, and dialects. To cope with this problem, we present a novel unsupervised SVR learning method for practical end-to-end ASR models. Our approach involves designing a speech feature masking method to stabilize SVR model learning and improve the performance of the ASR model in a downstream task. By introducing a noise masking strategy into diverse combinations of the time and frequency regions of the spectrogram, the SVR model becomes a robust representation extractor for the ASR model in practical scenarios. In pretraining experiments, we train the SVR model using approximately 18,000 h of Korean speech datasets that included diverse speakers and were recorded in environments with various amounts of noise. The weights of the pretrained SVR extractor are then frozen, and the extracted speech representations are used for ASR model training in a downstream task. The experimental results show that the ASR model using our proposed SVR extractor significantly outperforms conventional methods.
- 26 January 2023
- Published by MDPI AG in Mathematics
- Vol. 11 (3) , 622
Open AccessUnsupervised Representation Learning with Task-Agnostic Feature Masking for Robust End-to-End Speech Recognition
Unsupervised learning-based approaches for training speech vector representations (SVR) have recently been widely applied. While pretrained SVR models excel in relatively clean automatic speech recognition (ASR) tasks, such as those recorded in laboratory environments, they are still insufficient for practical applications with various types of noise, intonation, and dialects. To cope with this problem, we present a novel unsupervised SVR learning method for practical end-to-end ASR models. Our approach involves designing a speech feature masking method to stabilize SVR model learning and improve the performance of the ASR model in a downstream task. By introducing a noise masking strategy into diverse combinations of the time and frequency regions of the spectrogram, the SVR model becomes a robust representation extractor for the ASR model in practical scenarios. In pretraining experiments, we train the SVR model using approximately 18,000 h of Korean speech datasets that included diverse speakers and were recorded in environments with various amounts of noise. The weights of the pretrained SVR extractor are then frozen, and the extracted speech representations are used for ASR model training in a downstream task. The experimental results show that the ASR model using our proposed SVR extractor significantly outperforms conventional methods.
- Open Access
- 20 May 2020
- Published by MDPI AG in Applied Sciences
- Vol. 10 (10) , 3542
Semi-Supervised Speech Recognition Acoustic Model Training Using Policy Gradient
In this paper, we propose a policy gradient-based semi-supervised speech recognition acoustic model training. In practice, self-training and teacher/student learning are one of the widely used semi-supervised training methods due to their scalability and effectiveness. These methods are based on generating pseudo labels for unlabeled samples using a pre-trained model and selecting reliable samples using confidence measure. However, there are some considerations in this approach. The generated pseudo labels can be biased depending on which pre-trained model is used, and the training process can be complicated because the confidence measure is usually carried out in post-processing using external knowledge. Therefore, to address these issues, we propose a policy gradient method-based approach. Policy gradient is a reinforcement learning algorithm to find an optimal behavior strategy for an agent to obtain optimal rewards. The policy gradient-based approach provides a framework for exploring unlabeled data as well as exploiting labeled data, and it also provides a way to incorporate external knowledge in the same training cycle. The proposed approach was evaluated on an in-house non-native Korean recognition domain. The experimental results show that the method is effective in semi-supervised acoustic model training.
- 20 May 2020
- Published by MDPI AG in Applied Sciences
- Vol. 10 (10) , 3542
Open AccessSemi-Supervised Speech Recognition Acoustic Model Training Using Policy Gradient
In this paper, we propose a policy gradient-based semi-supervised speech recognition acoustic model training. In practice, self-training and teacher/student learning are one of the widely used semi-supervised training methods due to their scalability and effectiveness. These methods are based on generating pseudo labels for unlabeled samples using a pre-trained model and selecting reliable samples using confidence measure. However, there are some considerations in this approach. The generated pseudo labels can be biased depending on which pre-trained model is used, and the training process can be complicated because the confidence measure is usually carried out in post-processing using external knowledge. Therefore, to address these issues, we propose a policy gradient method-based approach. Policy gradient is a reinforcement learning algorithm to find an optimal behavior strategy for an agent to obtain optimal rewards. The policy gradient-based approach provides a framework for exploring unlabeled data as well as exploiting labeled data, and it also provides a way to incorporate external knowledge in the same training cycle. The proposed approach was evaluated on an in-house non-native Korean recognition domain. The experimental results show that the method is effective in semi-supervised acoustic model training.
- Open Access
- 17 January 2019
- Published by Wiley in ETRI Journal
- Vol. 41 (2) , 235-241
Rank‐weighted reconstruction feature for a robust deep neural network‐based acoustic model
- 17 January 2019
- Published by Wiley in ETRI Journal
- Vol. 41 (2) , 235-241
Open AccessRank‐weighted reconstruction feature for a robust deep neural network‐based acoustic model
- 12 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 316-319
I-vector based utterance verification for large-vocabulary speech recognition system
This paper proposes a new Utterance Verification (UV) algorithm based on i-vector. Phone segments are extracted and concatenated from the training data, which are used to train the Universal Background Model (UBM) and the Total Variability (TV) matrix, and then, i-vector is extracted from the enrollment and evaluation data using UBM and TV matrix. We compare two Confidence Measures (CMs), cosine distance scoring and Support Vector Machine (SVM). To compensate the channel effect, we use two channel compensation methods, Linear Discriminant Analysis (LDA) and Within-Class Covariance Normalization (WCCN). The decision is made by the word-level CM by combining the phone-level CMs. Experiments are conducted in the Korean isolated word recognition domain. Experimental results show that SVM is superior to cosine distance scoring. Best performance is achieved when SVM is used without any channel compensation method.
- 12 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 316-319
I-vector based utterance verification for large-vocabulary speech recognition system
This paper proposes a new Utterance Verification (UV) algorithm based on i-vector. Phone segments are extracted and concatenated from the training data, which are used to train the Universal Background Model (UBM) and the Total Variability (TV) matrix, and then, i-vector is extracted from the enrollment and evaluation data using UBM and TV matrix. We compare two Confidence Measures (CMs), cosine distance scoring and Support Vector Machine (SVM). To compensate the channel effect, we use two channel compensation methods, Linear Discriminant Analysis (LDA) and Within-Class Covariance Normalization (WCCN). The decision is made by the word-level CM by combining the phone-level CMs. Experiments are conducted in the Korean isolated word recognition domain. Experimental results show that SVM is superior to cosine distance scoring. Best performance is achieved when SVM is used without any channel compensation method.
- Open Access
- 1 October 2014
- Published by Wiley in ETRI Journal
- Vol. 36 (5) , 714-720
Weighted Finite State Transducer-Based Endpoint Detection Using Probabilistic Decision Logic
- 1 October 2014
- Published by Wiley in ETRI Journal
- Vol. 36 (5) , 714-720
Open AccessWeighted Finite State Transducer-Based Endpoint Detection Using Probabilistic Decision Logic
- Open Access
- 1 June 2014
- Published by Wiley in ETRI Journal
- Vol. 36 (3) , 514-517
Intra- and Inter-frame Features for Automatic Speech Recognition
- 1 June 2014
- Published by Wiley in ETRI Journal
- Vol. 36 (3) , 514-517
Open AccessIntra- and Inter-frame Features for Automatic Speech Recognition
- 25 August 2013
- Published by International Speech Communication Association
Endpoint detection using weighted finite state transducer
- 25 August 2013
- Published by International Speech Communication Association
Endpoint detection using weighted finite state transducer