Emotion Recognition from Chinese Speech for Smart Affective Services Using a Combination of SVM and DBN.

Accurate emotion recognition from speech is important for applications like smart health care, smart entertainment, and other smart services. High accuracy emotion recognition from Chinese speech is challenging due to the complexities of the Chinese language. In this paper, we explore how to improve the accuracy of speech emotion recognition, including speech signal feature extraction and emotion classification methods. Five types of features are extracted from a speech sample: mel frequency cepstrum coefficient (MFCC), pitch, formant, short-term zero-crossing rate and short-term energy. By comparing statistical features with deep features extracted by a Deep Belief Network (DBN), we attempt to find the best features to identify the emotion status for speech. We propose a novel classification method that combines DBN and SVM (support vector machine) instead of using only one of them. In addition, a conjugate gradient method is applied to train DBN in order to speed up the training process. Gender-dependent experiments are conducted using an emotional speech database created by the Chinese Academy of Sciences. The results show that DBN features can reflect emotion status better than artificial features, and our new classification approach achieves an accuracy of 95.8%, which is higher than using either DBN or SVM separately. Results also show that DBN can work very well for small training databases if it is properly designed.

DFML: Dynamic Federated Meta-Learning for Rare Disease Prediction.

Millions of patients suffer from rare diseases around the world. However, the samples of rare diseases are much smaller than those of common diseases. Hospitals are usually reluctant to share patient information for data fusion due to the sensitivity of medical data. These challenges make it difficult for traditional AI models to extract rare disease features for disease prediction. In this paper, we propose a Dynamic Federated Meta-Learning (DFML) approach to improve rare disease prediction. We design an Inaccuracy-Focused Meta-Learning (IFML) approach that dynamically adjusts the attention to different tasks according to the accuracy of base learners. Additionally, a dynamic weight-based fusion strategy is proposed to further improve federated learning, which dynamically selects clients based on the accuracy of each local model. Experiments on two public datasets show that our approach outperforms the original federated meta-learning algorithm in accuracy and speed with as few as five shots. The average prediction accuracy of the proposed model is improved by 13.28% compared with each hospital's local model.

Characterization of the first complete chloroplast genome sequence of Celtis sinensis (Cannabaceae) and its phylogenetic implications.

Celtis sinensis Pers. is a popular garden landscape tree in riparian areas and a valuable resistant tree in many extreme environments. Here, we determined the first complete chloroplast genome of C. sinensis using high-throughput sequencing technology. Our results showed the chloroplast genome of C.sinensis was 159,092 bp long and displayed a typical quadripartite structure consisting of a pair of inverted repeats with a length of 26,895 bp and separating by two single-copy regions (LSC, 86,085 bp and SSC, 19,217 bp). Besides, the chloroplast genome of C. sinensis totally contained 131 genes, including 87 protein coding genes, 36 transfer RNAs and eight ribosomal RNAs. Additionally, a maximum likelihood phylogenetic analysis based on the 19 chloroplast genomes demonstrated the monophyly of Cannabaceae and C. sinensis formed a sister clade to Celtis biondii.

The design and analysis of a new slipper-type hydraulic support.

To improve the safety and the stability of the support under mines and reduce the cost, we design a new slipper-type hydraulic support with energy-efficiency and high reliability. To study its dynamics, we build a reverse kinematics model. We analyze the motion and the force for each component of the new support with a simulation in Matlab/Simulink. The results show that it has appropriate structures with the required four-bar linkages. To compare the performance between the new slipper-type support and the traditional support, we design their mechanics models, deduce their mechanics relations and obtain the force curves for each component of both supports under the same loads. The results prove that the new slipper-type support has less demand on oil pressure for the hydraulic cylinder when working at middle and high positions and it has a larger supporting force and a higher supporting stability which would be more energy-efficient.