-
1.发明授权公开(公告)号：US06324290B1
公开(公告)日：2001-11-27
申请号：US08949759
申请日：1997-10-14
IPC分类号： H04R300
CPC分类号： G01H1/00
摘要： Acoustics or vibrations at a plurality of predetermined positions are respectively detected in the form of analog values, and the detected analog values are respectively converted into digital values. The digital values are subjected to digital filtering by using the digital values and filter factors, and at least one of the filter factors is changed such that the difference between the total sum of the results of digital filtering and a targeted value becomes minimum. The rate of contribution to an evaluation point is computed by using the results of digital filtering within a predetermined time when the difference becomes minimum, and the computed rate of contribution is displayed.
摘要翻译： 分别以模拟值的形式检测多个预定位置处的声学或振动，并将检测到的模拟值分别转换为数字值。 通过使用数字值和滤波因子对数字值进行数字滤波，并且改变滤波器因子中的至少一个，使得数字滤波的结果和目标值的总和之间的差成为最小。 通过在差异变为最小时的预定时间内使用数字滤波的结果来计算对评价点的贡献率，并且显示计算出的贡献率。
-
2.发明授权公开(公告)号：US07680660B2
公开(公告)日：2010-03-16
申请号：US11060754
申请日：2005-02-18
申请人： Hirokazu Kudoh
发明人： Hirokazu Kudoh
IPC分类号： G10L15/00
CPC分类号： G10L15/04 , G10L2015/025
摘要： A voice analysis method and device are provided whereby processing can be performed in real time and an indefinitely large number of speakers can be coped with. A computer-executable method of voice analysis is for detecting boundaries of phonemes from input voice, and is characterized by repeating a step of specifying a time-point in an input voice signal, a step of extracting the voice signal contained in a time range of prescribed length from this time-point, and a step of decomposing the extracted voice signal into frequency component data; finding a plurality of frequency component data from the voice signal contained in time ranges of the prescribed length; finding a plurality of correlations using the frequency component data corresponding to the voice signal contained in mutually adjacent time ranges of the prescribed length; finding time ranges having a degree of change that is larger than the two adjacent degrees of change on either side thereof; and partitioning the input voice signal into a plurality of sections based on these time ranges.
摘要翻译： 提供了语音分析方法和装置，由此能够实时地执行处理，并且可以应付无限大数量的扬声器。 语音分析的计算机可执行方法是用于检测来自输入语音的音素的边界，其特征在于重复指定输入语音信号中的时间点的步骤，提取包含在时间范围内的语音信号的步骤 从该时间点起的规定长度，以及将所提取的语音信号分解为频率分量数据的步骤; 从包含在规定长度的时间范围内的语音信号中找出多个频率分量数据; 使用与包含在规定长度的相邻的时间范围内的语音信号相对应的频率分量数据来发现多个相关; 找到具有大于其任一侧上的两个相邻变化度的变化程度的时间范围; 并且基于这些时间范围将输入语音信号划分成多个部分。
-
公开(公告)号：US07477253B2
公开(公告)日：2009-01-13
申请号：US10441194
申请日：2003-05-20
CPC分类号： G06T13/00 , G06T2210/44 , Y10S345/956 , Y10S345/957
摘要： An animation image generating program is provided which allows animation images to be readily generated by CG without complicated setups, and more particularly the animation image generating program is suited to generate a plurality of types of face animation images by CG. The animation image generating program includes the steps of controlling selection of specific vertices of a standard model and a user model, providing control such that first target vertices are associated with second target vertices where the first target vertices are the selected specific vertices of the standard model and the second target vertices are the selected specific vertices of the user model, providing control by arithmetic means such that coordinates of the first target vertices approximate to those of the second target vertices, to generate fitting information, and animating the user model based on animation data of the standard model and on the fitting information.
摘要翻译： 提供了一种动画图像生成程序，其允许通过CG容易地生成动画图像，而不需要复杂的设置，更具体地，动画图像生成程序适合于通过CG生成多种类型的面部动画图像。 动画图像生成程序包括以下步骤：控制标准模型和用户模型的特定顶点的选择，提供控制，使得第一目标顶点与第二目标顶点相关联，其中第一目标顶点是标准模型的所选择的特定顶点 并且第二目标顶点是用户模型的所选择的特定顶点，通过算术装置提供控制，使得第一目标顶点的坐标近似于第二目标顶点的坐标，以生成拟合信息，以及基于动画来动画化用户模型 标准模型的数据和拟合信息。
-
4.发明申请公开(公告)号：US20050192805A1
公开(公告)日：2005-09-01
申请号：US11060754
申请日：2005-02-18
申请人： Hirokazu Kudoh
发明人： Hirokazu Kudoh
CPC分类号： G10L15/04 , G10L2015/025
摘要： A voice analysis method and device are provided whereby processing can be performed in real time and an indefinitely large number of speakers can be coped with. A computer-executable method of voice analysis is for detecting boundaries of phonemes from input voice, and is characterized by repeating a step of specifying a time-point in an input voice signal, a step of extracting the voice signal contained in a time range of prescribed length from this time-point, and a step of decomposing the extracted voice signal into frequency component data; finding a plurality of frequency component data from the voice signal contained in time ranges of the prescribed length; finding a plurality of correlations using the frequency component data corresponding to the voice signal contained in mutually adjacent time ranges of the prescribed length; finding time ranges having a degree of change that is larger than the two adjacent degrees of change on either side thereof; and partitioning the input voice signal into a plurality of sections based on these time ranges.
摘要翻译： 提供了语音分析方法和装置，由此能够实时地执行处理，并且可以应付无限大数量的扬声器。 语音分析的计算机可执行方法是用于检测来自输入语音的音素的边界，其特征在于重复指定输入语音信号中的时间点的步骤，提取包含在时间范围内的语音信号的步骤 从该时间点起的规定长度，以及将所提取的语音信号分解为频率分量数据的步骤; 从包含在规定长度的时间范围内的语音信号中找出多个频率分量数据; 使用与包含在规定长度的相邻的时间范围内的语音信号相对应的频率分量数据来发现多个相关; 找到具有大于其任一侧上的两个相邻变化度的变化程度的时间范围; 并且基于这些时间范围将输入语音信号划分成多个部分。
-
-
-
搜索结果
4 条记录 (耗时 0.016 秒)
