Search

With rapid development of voice control technology, making speech recognition more precisely in various IoT domains have been an intractable problem to be solved. Since there are various conversation scenes, understanding the context of a dialog scene is a key issue of voice control systems. However, the reality is available training data for dialog system are always insufficient. In this paper, we mainly solve the problem of data lacking in dialog systems by data augmentation technique. A Generative Adversarial Network(GAN)-based model is proposed and the data are augmented effectively. It can generate from text to text, enhance the original data with text retelling, and improve the robustness of parameter estimation of unknown data by using the sample data generated by GAN model. A new N-gram language model is used to evaluate multiple recognition candidates of speech recognition, and the candidate sentences with the highest evaluation scores are selected as the final result of speech recognition. Our data enhancement algorithm based on the Generative Model is verified by the experiments. In the result of model comparison test, the error rates of data set THCHS30 and AISHELL are 3.3% and 5.1% which are lower than that of the baseline system. [ABSTRACT FROM AUTHOR]

With the exponential growth of wireless users and their traffic demands, it is greatly increasing for the demand of the scarce spectrum resources in the communication networks. In order to enhance the performance of the wireless networks such as end-to-end delay, energy efficiency and throughput, the device-to-device (D2D) communication has been attracted more attention because the two devices in close proximity can communicate directly without traversing the central base station. However, most of users are very sensitive to the battery. Therefore, we aim to maximize the energy efficiency of wireless communication system in the context of underlaying device-to-device communication in this paper, We focus on the formulated power control and resource allocation problem which is non-convex in the fractional form. We reduce it from the power allocation of all users to the joint power and subchannel allocation of D2D users. Then, we tackle it by an iterative approximation algorithm leveraging to the properties of fractional programming. There are two studied cases for the subchannel allocation. One can be solved by the penalty function approach, and the other can be solved by the dual decomposition as well as sub-gradient method. Accordingly, we propose a dual-based algorithm in general. Numerical simulations demonstrate that the proposed algorithms outperform the conventional algorithm in terms of the energy efficiency. [ABSTRACT FROM AUTHOR]

An introduction is presented in which editor discusses various articles within the issue on topics including product model to reflect the recent purchasing patterns of customers, mobile application for a personal information disclosure attack on smartphones and Speech Recognition System (ASR).

Vehicular communications refer to a wide range of networks proposed for environments characterized by sparse connectivity, frequent network partitioning, intermittent connectivity, long propagation delays, asymmetric data rates, and high error rates. These environments may also be characterized by a potential non-available end-to-end path. To overcome these issues and improve the overall network performance, cooperation between network nodes must be severely considered. Nodes may cooperate by sharing their storage capacity, bandwidth, or even energy resources. However, nodes may be unwilling to cooperate due to a selfish behavior or to an intent to protect the integrity of their own resources. This selfish behavior significantly affects the functionality and performance of the network. This paper overviews the most recent advances related with cooperation on vehicular networks. It also studies the impact of different cooperation levels in the performance of Vehicular Delay-Tolerant Networks (VDTNs). It was shown that scenarios with a higher number of cooperative nodes present the best results in terms of bundle delivery delay. [ABSTRACT FROM AUTHOR]