ID 原文 译文
22075 VoIP 以语音流为传输媒介，具有传输数据量大和应用广泛的优点。 VoIP (Voice over Internet Protocol) is based on voice stream, which has the advantages of large data transmission and wide application.
22076 但 VoIP 系统也面临数据安全和隐私泄露的安全威胁。 But the VoIP system is confronted with the security threats of data security and privacy disclosure.
22077 针对编码标准 G.729 固定码本搜索的非遍历特性和具有一定冗余性的特点，该文提出基于 G.729语音编码非零脉冲位置信息的隐藏算法。 Thus, according to nonergodicity and redundancy of the fixed codebook search, an information hiding algorithm based on fixed codebook search process is proposed.
22078 该算法在固定码本搜索过程中，利用秘密信息控制码本的搜索过程，并在非零脉冲位置和秘密信息之间构建函数进行信息隐藏。 This information hiding algorithm is carried out by a functional relationship between the nonzero pulse positions and secret information.
22079 在搜索过程中利用最不重要脉冲替换思想并采用最小化失真准则控制由秘密信息的嵌入带来的音质失真。 The idea of least significant pulse replacement is used in the search process and a distortion minimization criterion is proposed to control the distortion of speech quality caused by the embedding of secret information.
22080 实验结果表明：算法隐藏容量可达 400 bit/s，算法具有良好的隐蔽性(PESQ 平均值约为 3.45)。 The experimental results show that the hiding capacity of the proposed algorithm is up to 400 bit/s and the average PESQ score is 3.45 which indicate that the algorithm has good imperceptibility.
22081 该文针对时变离群值环境下的在线学习问题，提出一种基于 M-estimator 与可变遗忘因子的在线贯序超限学习机算法(VFF-M-OSELM)。 To solve the online learning problem under the scenario of time-varying and containing outliers, this paper proposes an M-estimator and Variable Forgetting Factor based Online Sequential Extreme Learning Machine (VFF-M-OSELM).
22082 VFF-M-OSELM 以在线贯序超限学习机模型为基础，通过引入一种更加鲁棒的M-estimator 代价函数来替代传统的最小二乘代价函数，以提高模型对于离群值的在线处理能力和鲁棒性。 The VFF-M-OSELM is developed from the online sequential extreme learning machine algorithm and retains the same excellent sequential learning ability as it replaces the conventional Least-Squares (LS) cost function with a robust M-estimator based cost function to enhance the robustness of the learning model to outliers.
22083 同时VFF-M-OSELM 通过融合使用一种新的可变遗忘因子方法进一步增强了其在时变环境下的动态跟踪能力和自适应性。 Meanwhile, a new variable forgetting factor method is designed and incorporated in the VFF-M- OSELM to enhance further the dynamic tracking ability and adaptivity of the algorithm to time-varying system.
22084 仿真实例验证了所提算法的有效性。 The simulation results verify the effectiveness of the proposed algorithm.