• Phonetics and Speech Sciences
• /
• v.11 no.2
• /
• pp.57-64
• /
• 2019

In this paper, we construct a Korean text-to-speech system using the Merlin toolkit which is an open source system for speech synthesis. In the text-to-speech system, the HMM-based statistical parametric speech synthesis method is widely used, but it is known that the quality of synthesized speech is degraded due to limitations of the acoustic modeling scheme that includes context factors. In this paper, we propose an acoustic modeling architecture that uses deep neural network technique, which shows excellent performance in various fields. Fully connected deep feedforward neural network (DNN), recurrent neural network (RNN), gated recurrent unit (GRU), long short-term memory (LSTM), bidirectional LSTM (BLSTM) are included in the architecture. Experimental results have shown that the performance is improved by including sequence modeling in the architecture, and the architecture with LSTM or BLSTM shows the best performance. It has been also found that inclusion of delta and delta-delta components in the acoustic feature parameters is advantageous for performance improvement.

• Smart Media Journal
• /
• v.11 no.10
• /
• pp.65-75
• /
• 2022

Deep learning is used as a creative tool that could overcome the limitations of existing analysis models and generate various types of results such as text, image, and music. In this paper, we propose a method necessary to preprocess audio data using the Niko's MIDI Pack sound source file as a data set and to generate music using Bi-LSTM. Based on the generated root note, the hidden layers are composed of multi-layers to create a new note suitable for the musical composition, and an attention mechanism is applied to the output gate of the decoder to apply the weight of the factors that affect the data input from the encoder. Setting variables such as loss function and optimization method are applied as parameters for improving the LSTM model. The proposed model is a multi-channel Bi-LSTM with attention that applies notes pitch generated from separating treble clef and bass clef, length of notes, rests, length of rests, and chords to improve the efficiency and prediction of MIDI deep learning process. The results of the learning generate a sound that matches the development of music scale distinct from noise, and we are aiming to contribute to generating a harmonistic stable music.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.10
• /
• pp.1490-1499
• /
• 2022

Arrhythmia is a condition in which the heart has an irregular rhythm or abnormal heart rate, early diagnosis and management is very important because it can cause stroke, cardiac arrest, or even death. In this paper, we propose arrhythmia classification using hybrid combination model of CNN-BLSTM. For this purpose, the QRS features are detected from noise removed signal through pre-processing and a single bit segment was extracted. In this case, the GAN oversampling technique is applied to solve the data imbalance problem. It consisted of CNN layers to extract the patterns of the arrhythmia precisely, used them as the input of the BLSTM. The weights were learned through deep learning and the learning model was evaluated by the validation data. To evaluate the performance of the proposed method, classification accuracy, precision, recall, and F1-score were compared by using the MIT-BIH arrhythmia database. The achieved scores indicate 99.30%, 98.70%, 97.50%, 98.06% in terms of the accuracy, precision, recall, F1 score, respectively.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.5
• /
• pp.315-321
• /
• 2023

Long-span bridges are flexible structures with low natural frequencies and damping ratios, making them susceptible to vibrational serviceability problems. However, the current design guideline of South Korea assumes a uniform threshold of wind speed or vibrational amplitude to assess the occurrence of harmful vibrations, potentially overlooking the complex vibrational patterns observed in long-span bridges. In this study, we propose a pointwise vortex-induced vibration (VIV) detection method using a deep-learning-based signalsegmentation model. Departing from conventional supervised methods of data acquisition and manual labeling, we synthesize training data by generating sinusoidal waves with an envelope to accurately represent VIV. A Fourier synchrosqueezed transform is leveraged to extract time-frequency features, which serve as input data for training a bidirectional long short-term memory model. The effectiveness of the model trained on synthetic VIV data is demonstrated through a comparison with its counterpart trained on manually labeled real datasets from an actual cable-supported bridge.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.8
• /
• pp.1-7
• /
• 2022

With the development of artificial intelligence, efforts to incorporate neuroscience mining with AI have increased. Neuroscience mining, also known as NSM, expands on this concept by combining computational neuroscience and business analytics. Using fNIRS (functional near-infrared spectroscopy)-based experiment dataset, we have investigated the potential of NSM in the context of the BPSC (business problem-solving creativity) prediction. Although BPSC is regarded as an essential business differentiator and a difficult cognitive resource to imitate, measuring it is a challenging task. In the context of NSM, appropriate methods for assessing and predicting BPSC are still in their infancy. In this sense, we propose a novel NSM method that systematically combines CNN, BiLSTM, and attention network for the sake of enhancing the BPSC prediction performance significantly. We utilized a dataset containing over 150 thousand fNIRS-measured data points to evaluate the validity of our proposed NSM method. Empirical evidence demonstrates that the proposed NSM method reveals the most robust performance when compared to benchmarking methods.