阐述了一种连续语音信号的最大幅度的基音标注算法。
Describes an algorithm for marking pitch in the location of the highest amplitude of continuous speech signal.
所有问题连续编号,每个问题编号与文中划线部分和以方形标注的地方相对应。
The questions are numbered consecutively. Each question number corresponds to an underlined portion in the passage or to a box located in the passage.
阐述了一种连续语音信号的最高振幅位的基音标注算法,该算法是在声带振动信号中找到喉部振动关闭点,并进行参数调整,从而自动完成在连续语音信号中的基音位置标注。
An algorithm for marking pitch at the location of the largest amplitude of continuous speech signal is described, which is based on the finding of glottal closure instants in laryngograph signal.
利用连续的粒子群优化代价敏感主动学习的控制参数,该参数用于最大化未标注样本的信息度和最小化标注代价。
This optimization mechanism employs the continuous-valued PSO to optimize the control parameter to maximize the value of information of instance and minimize the cost of oracle.
利用连续的粒子群优化代价敏感主动学习的控制参数,该参数用于最大化未标注样本的信息度和最小化标注代价。
This optimization mechanism employs the continuous-valued PSO to optimize the control parameter to maximize the value of information of instance and minimize the cost of oracle.
应用推荐