|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | 1D-CNN:Speech Emotion Recognition System Using a Stacked Network with Dilated CNN Features显示文摘Emotion recognition from speech data is an active and emerging area of research that plays an important role in numerous applications,such as robotics,virtual reality,behavior assessments,and emergency call centers.Recently,researchers have developed many techniques in this field in order to ensure an improvement in the accuracy by utilizing several deep learning approaches,but the recognition rate is still not convincing.Our main aim is to develop a new technique that increases the recognition rate with reasonable cost computations.In this paper,we suggested a new technique,which is a one-dimensional dilated convolutional neural network(1D-DCNN)for speech emotion recognition(SER)that utilizes the hierarchical features learning blocks(HFLBs)with a bi-directional gated recurrent unit(BiGRU).We designed a one-dimensional CNN network to enhance the speech signals,which uses a spectral analysis,and to extract the hidden patterns from the speech signals that are fed into a stacked one-dimensional dilated network that are called HFLBs.Each HFLB contains one dilated convolution layer(DCL),one batch normalization(BN),and one leaky_relu(Relu)layer in order to extract the emotional features using a hieratical correlation strategy.Furthermore,the learned emotional features are feed into a BiGRU in order to adjust the global weights and to recognize the temporal cues.The final state of the deep BiGRU is passed from a softmax classifier in order to produce the probabilities of the emotions.The proposed model was evaluated over three benchmarked datasets that included the IEMOCAP,EMO-DB,and RAVDESS,which achieved 72.75%,91.14%,and 78.01%accuracy,respectively. | Mustaqeem Soonil Kwon | 2021 | Computers, Materials & Continua2021，，6： | 2 |
| 2 | Robust speaker identification based on selective use of feature vectors显示文摘 | Soonil Kwon Shrikanth Narayanan | 2007 | Pattern Recognition Letters2007，28，1： | 1 |
| 3 | Unsupervised speaker indexing using generic models显示文摘 | Soonil kwon Shrikanth Narayanan | 2005 | IEEE Transactions on Speech and Audio Processing2005，13，5： | 1 |
| 4 | Unsupervised Speaker Indexing Using Generic Models 显示文摘 | Kwon Soonil Narayanan | 2005 | IEEE Transactions on Speech and Audio Processing2005，13，5： | 1 |
| 5 | Unsupervised speaker indexing using generic models显示文摘 | Kwon Soonil Narayanan | 2005 | IEEE Transactions on Speech and Audio Processing2005，13，5： | 1 |
| 6 | Photoluminescence of Nominally Undoped Heavy n-Type ZnO Nanowires显示文摘We report the identification of a donor band and the correlation between n-type conductivity and the green emission in ZnO nanowires.Temperature-dependent photoluminescence is used to investigate nominally undoped ZnO nanowires with high n-type conductivity.Within the whole temperature range,a dominant free-to-bound transition with a donor band of about 150meV below the conduction band minimum is observed.The nanowires show very strong green emission,which is quenched with activation energy of about 220 meV.The correlation between the high n-type conductivity and the strong green emission is discussed in detail,and we suggest that they may have different origins. | 唐海平 何海平 刘超 KWON Bong-Jun 叶志镇 LEE Soonil PARK Ji-Yong | 2011 | Chinese Physics Letters2011，28，2： | 0 |
| 7 | Preventing condensation of objective lens in noncontact wide-angle viewing systems during vitrectomy显示文摘AIM: To assess the optimal conditions for preventing condensation of objective lens during vitrectomy with noncontact wide-angle viewing systems(WAVSs). METHODS: We explored the effectiveness of the coating with ophthalmic viscoelastic device(OVDs) on the corneal surface and the soaking the objective lens in warm-saline for preventing condensation of objective lens. First, to find the optimal soaking time to keep the objective lens warm, we measured the temperature of objective lens every minute after soaking in warm saline. Second, to find optimal distance between cornea and objective lens, which provide as wide a view as possible and less condensation at the same time, we measured the condensation time with different distances. With the obtained optimal soaking time and distance, we explored the effect of coating cornea with OVDs and soaking objective lens in warm saline on condensation time.RESULTS: One and 5 min of soaking in warm saline was most effective for keeping the lens warm enough(45.1℃±2.1℃ for 1 min and 46.4℃±1.0℃ for 5 min, P=0.109). The mean condensation times for the control group at 1, 3, and 5 mm from corneal surface to objective lens were 1±0.4, 4±1.4, 190±26.1 s, respectively, thus 5 mm was most optimal distance for vitrectomy with WAVSs. For the OVD coating group, the mean condensation times were 1.5±0.3, 13±1.4, and 200±23.9 s at 1, 3, and 5 mm distance and borderline significant compared with control group(P=0.068, 0.051, and 0.063, respectively). With the 1-minute warm saline soaking group, the mean condensation time were extended to 188±34.4, 416±65.7, and 600±121.3 s at 1, 3, and 5 mm distance and statistically significant compared with control(P=0.043, 0.041 and 0.043, respectively).CONCLUSION: OVD coating on corneal surface shows no difference on condensation time with control group. However, soaking the objective lens in warm saline revealed statistically significant extension of condensation time compared to control group. Therefore, keeping the objective lens warm with soaking in warm saline is a simple but effective to prevent condensation of objective lens during vitrectomy. The thermodynamics between objective lens and cornea during vitrectomy warrants further investigation. | Jung Pil Lee Jinsoo Kim Inwon Park Ho Ra Soonil Kwon | 2018 | International Journal of Ophthalmology(English edition)2018，11，11： | 0 |
| 8 | TC-Net:A Modest&Lightweight Emotion Recognition System Using Temporal Convolution Network显示文摘Speech signals play an essential role in communication and provide an efficient way to exchange information between humans and machines.Speech Emotion Recognition(SER)is one of the critical sources for human evaluation,which is applicable in many real-world applications such as healthcare,call centers,robotics,safety,and virtual reality.This work developed a novel TCN-based emotion recognition system using speech signals through a spatial-temporal convolution network to recognize the speaker’s emotional state.The authors designed a Temporal Convolutional Network(TCN)core block to recognize long-term dependencies in speech signals and then feed these temporal cues to a dense network to fuse the spatial features and recognize global information for final classification.The proposed network extracts valid sequential cues automatically from speech signals,which performed better than state-of-the-art(SOTA)and traditional machine learning algorithms.Results of the proposed method show a high recognition rate compared with SOTAmethods.The final unweighted accuracy of 80.84%,and 92.31%,for interactive emotional dyadic motion captures(IEMOCAP)and berlin emotional dataset(EMO-DB),indicate the robustness and efficiency of the designed model. | Muhammad Ishaq Mustaqeem Khan Soonil Kwon | 2023 | Computer Systems Science & Engineering2023，46，9： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员