专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US07680657B2 Auto segmentation based partitioning and clustering approach to robust endpointing 失效
标题翻译：基于自动分割的分区和聚类方法进行鲁棒终点
公开(公告)号：US07680657B2
公开(公告)日：2010-03-16
申请号：US11504280
申请日：2006-08-15
申请人： Yu Shi , Frank Kao-ping Soong , Jian-Iai Zhou
发明人： Yu Shi , Frank Kao-ping Soong , Jian-Iai Zhou
IPC分类号： G10L15/04 , G10L15/20 , G10L15/00
CPC分类号： G10L25/87
摘要： Possible segmentations for an audio signal are scored based on distortions for feature vectors of the audio signal and the total number of segments in the segmentation. The scores are used to select a segmentation and the selected segmentation is used to identify a starting point and an ending point for a speech signal in the audio signal.
摘要翻译：基于音频信号的特征向量的失真和分段中的段的总数，对音频信号进行可能的分割。分数用于选择分割，并且所选择的分割用于识别音频信号中的语音信号的起始点和终点。

2. 发明授权

US07680664B2 Parsimonious modeling by non-uniform kernel allocation 有权
标题翻译：通过非均匀内核分配进行建模
公开(公告)号：US07680664B2
公开(公告)日：2010-03-16
申请号：US11505206
申请日：2006-08-16
申请人： Peng Liu , Jian-Iai Zhou , Frank Kao-ping Soong
发明人： Peng Liu , Jian-Iai Zhou , Frank Kao-ping Soong
IPC分类号： G10L15/14 , G10L15/28
CPC分类号： G06K9/6297 , G10L15/14 , G10L15/144
摘要： A multi-state pattern recognition model with non-uniform kernel allocation is formed by setting a number of states for a multi-state pattern recognition model and assigning different numbers of kernels to different states. The kernels are then trained using training data to form the multi-state pattern recognition model.
摘要翻译：通过为多状态模式识别模型设置多个状态并将不同数量的内核分配给不同的状态，形成具有不均匀内核分配的多状态模式识别模型。然后使用训练数据训练内核以形成多状态模式识别模型。

3. 发明授权

US08355917B2 Position-dependent phonetic models for reliable pronunciation identification 有权
标题翻译：位置依赖语音模型，用于可靠的发音识别
公开(公告)号：US08355917B2
公开(公告)日：2013-01-15
申请号：US13364144
申请日：2012-02-01
申请人： Peng Liu , Yu Shi , Frank Kao-ping Soong
发明人： Peng Liu , Yu Shi , Frank Kao-ping Soong
IPC分类号： G10L15/28
CPC分类号： G10L15/187 , G10L15/183
摘要： A representation of a speech signal is received and is decoded to identify a sequence of position-dependent phonetic tokens wherein each token comprises a phone and a position indicator that indicates the position of the phone within a syllable.
摘要翻译：语音信号的表示被接收并且被解码以识别位置依赖语音令牌的序列，其中每个令牌包括电话和指示电话在音节内的位置的位置指示符。

4. 发明申请

US20080172224A1 Position-dependent phonetic models for reliable pronunciation identification 有权
标题翻译：位置依赖语音模型，用于可靠的发音识别
公开(公告)号：US20080172224A1
公开(公告)日：2008-07-17
申请号：US11652451
申请日：2007-01-11
申请人： Peng Liu , Yu Shi , Frank Kao-ping Soong
发明人： Peng Liu , Yu Shi , Frank Kao-ping Soong
IPC分类号： G10L19/04
CPC分类号： G10L15/187 , G10L15/183
摘要： A representation of a speech signal is received and is decoded to identify a sequence of position-dependent phonetic tokens wherein each token comprises a phone and a position indicator that indicates the position of the phone within a syllable.
摘要翻译：语音信号的表示被接收并且被解码以识别位置依赖语音令牌的序列，其中每个令牌包括电话和位置指示符，其指示电话在音节内的位置。

5. 发明申请

US20120191456A1 POSITION-DEPENDENT PHONETIC MODELS FOR RELIABLE PRONUNCIATION IDENTIFICATION 有权
标题翻译：用于可靠发信标识的位置相关电话模型
公开(公告)号：US20120191456A1
公开(公告)日：2012-07-26
申请号：US13364144
申请日：2012-02-01
申请人： Peng Liu , Yu Shi , Frank Kao-ping Soong
发明人： Peng Liu , Yu Shi , Frank Kao-ping Soong
IPC分类号： G10L15/04
CPC分类号： G10L15/187 , G10L15/183
摘要： A representation of a speech signal is received and is decoded to identify a sequence of position-dependent phonetic tokens wherein each token comprises a phone and a position indicator that indicates the position of the phone within a syllable.
摘要翻译：语音信号的表示被接收并且被解码以识别位置依赖语音令牌的序列，其中每个令牌包括电话和位置指示符，其指示电话在音节内的位置。

6. 发明申请

US20080059169A1 Auto segmentation based partitioning and clustering approach to robust endpointing 失效
标题翻译：基于自动分割的分区和聚类方法进行鲁棒终点
公开(公告)号：US20080059169A1
公开(公告)日：2008-03-06
申请号：US11504280
申请日：2006-08-15
申请人： Yu Shi , Frank Kao-ping Soong , Jian-lai Zhou
发明人： Yu Shi , Frank Kao-ping Soong , Jian-lai Zhou
IPC分类号： G10L15/20
CPC分类号： G10L25/87
摘要： Possible segmentations for an audio signal are scored based on distortions for feature vectors of the audio signal and the total number of segments in the segmentation. The scores are used to select a segmentation and the selected segmentation is used to identify a starting point and an ending point for a speech signal in the audio signal.
摘要翻译：基于音频信号的特征向量的失真和分段中的段的总数，对音频信号进行可能的分割。分数用于选择分割，并且所选择的分割用于识别音频信号中的语音信号的起始点和终点。

7. 发明授权

US08077975B2 Handwriting symbol recognition accuracy using speech input 失效
标题翻译：使用语音输入的手写符号识别精度
公开(公告)号：US08077975B2
公开(公告)日：2011-12-13
申请号：US12037095
申请日：2008-02-26
申请人： Lei Ma , Yu Shi , Frank Kao-ping Soong
发明人： Lei Ma , Yu Shi , Frank Kao-ping Soong
IPC分类号： G06K9/00 , G06K9/62 , G06K9/72 , G06F3/00 , G10L15/26 , G09G5/00
CPC分类号： G10L15/24 , G06K9/00409 , G06K9/00422
摘要： Described is a bimodal data input technology by which handwriting recognition results are combined with speech recognition results to improve overall recognition accuracy. Handwriting data and speech data corresponding to mathematical symbols are received and processed (including being recognized) into respective graphs. A fusion mechanism uses the speech graph to enhance the handwriting graph, e.g., to better distinguish between similar handwritten symbols that are often misrecognized. The graphs include nodes representing symbols, and arcs between the nodes representing probability scores. When arcs in the first and second graphs are determined to match one another, such as aligned in time and associated with corresponding symbols, the probability score in the second graph for that arc is used to adjust the matching probability score in the first graph. Normalization and smoothing may be performed to correspond the graphs to one another and to control the influence of one graph on the other.
摘要翻译：描述了一种双模数据输入技术，通过该技术，手写识别结果与语音识别结果相结合，以提高整体识别精度。对应于数学符号的手写数据和语音数据被接收并处理（包括被识别）到各个图中。融合机制使用语音图来增强手写图，例如更好地区分经常被误识别的类似的手写符号。这些图包括表示符号的节点和表示概率分数的节点之间的弧。当第一和第二图中的弧被确定为彼此匹配时，例如在时间上对齐并与对应符号相关联时，该弧的第二图中的概率分数用于调整第一图中的匹配概率得分。可以执行归一化和平滑以将图彼此对应并且控制一个图的影响。

8. 发明授权

US08135590B2 Position-dependent phonetic models for reliable pronunciation identification 有权
标题翻译：位置依赖语音模型，用于可靠的发音识别
公开(公告)号：US08135590B2
公开(公告)日：2012-03-13
申请号：US11652451
申请日：2007-01-11
申请人： Peng Liu , Yu Shi , Frank Kao-ping Soong
发明人： Peng Liu , Yu Shi , Frank Kao-ping Soong
IPC分类号： G10L15/28
CPC分类号： G10L15/187 , G10L15/183
摘要： A representation of a speech signal is received and is decoded to identify a sequence of position-dependent phonetic tokens wherein each token comprises a phone and a position indicator that indicates the position of the phone within a syllable.
摘要翻译：语音信号的表示被接收并且被解码以识别位置依赖语音令牌的序列，其中每个令牌包括电话和指示电话在音节内的位置的位置指示符。

9. 发明申请

US20090214117A1 HANDWRITING SYMBOL RECOGNITION ACCURACY USING SPEECH INPUT 失效
标题翻译：使用语音输入的手写符号识别精度
公开(公告)号：US20090214117A1
公开(公告)日：2009-08-27
申请号：US12037095
申请日：2008-02-26
申请人： Lei Ma , Yu Shi , Frank Kao-ping Soong
发明人： Lei Ma , Yu Shi , Frank Kao-ping Soong
IPC分类号： G10L15/00
CPC分类号： G10L15/24 , G06K9/00409 , G06K9/00422
摘要： Described is a bimodal data input technology by which handwriting recognition results are combined with speech recognition results to improve overall recognition accuracy. Handwriting data and speech data corresponding to mathematical symbols are received and processed (including being recognized) into respective graphs. A fusion mechanism uses the speech graph to enhance the handwriting graph, e.g., to better distinguish between similar handwritten symbols that are often misrecognized. The graphs include nodes representing symbols, and arcs between the nodes representing probability scores. When arcs in the first and second graphs are determined to match one another, such as aligned in time and associated with corresponding symbols, the probability score in the second graph for that arc is used to adjust the matching probability score in the first graph. Normalization and smoothing may be performed to correspond the graphs to one another and to control the influence of one graph on the other.
摘要翻译：描述了一种双模数据输入技术，通过该技术，手写识别结果与语音识别结果相结合，以提高整体识别精度。对应于数学符号的手写数据和语音数据被接收并处理（包括被识别）到各个图中。融合机制使用语音图来增强手写图，例如更好地区分经常被误识别的类似的手写符号。这些图包括表示符号的节点和表示概率分数的节点之间的弧。当第一和第二图中的弧被确定为彼此匹配时，例如在时间上对准并与对应符号相关联时，该弧的第二图中的概率分数用于调整第一图中的匹配概率分数。可以执行归一化和平滑以将图彼此对应并且控制一个图的影响。

10. 发明申请

US20080201145A1 Unsupervised labeling of sentence level accent 有权
标题翻译：句子级别口音的无监督标签
公开(公告)号：US20080201145A1
公开(公告)日：2008-08-21
申请号：US11708442
申请日：2007-02-20
申请人： YiNing Chen , Frank Kao-ping Soong , Min Chu
发明人： YiNing Chen , Frank Kao-ping Soong , Min Chu
IPC分类号： G10L15/00
CPC分类号： G10L13/08
摘要： Methods are disclosed for automatic accent labeling without manually labeled data. The methods are designed to exploit accent distribution between function and content words.
摘要翻译：公开了无手动标记数据的自动重音标记的方法。这些方法旨在利用功能和内容单词之间的重音分配。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式