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. 发明申请

WO2003001507A1 HIDDEN MARKOV MODEL WITH FRAME CORRELATION 审中-公开
标题翻译：隐藏MARKOV模型与框架相关
公开(公告)号：WO2003001507A1
公开(公告)日：2003-01-03
申请号：PCT/CN2001/001037
申请日：2001-06-22
申请人： INTEL CORPORATION , INTEL CHINA LTD. , LI, Jinyu , JIA, Ying
发明人： LI, Jinyu , JIA, Ying
IPC分类号： G10L15/14
CPC分类号： G10L15/142 , G10L25/12
摘要： A method and system for incorporating frame correlation in a hidden Markov model is disclosed. The method includes calculating frame independent parts of a speech, inputting the frame independent parts to a Gaussian model, and computing frame probability. Auto-regressive coefficients are then computed from the frame probability.
摘要翻译：公开了一种在隐马尔可夫模型中并入帧相关的方法和系统。该方法包括计算语音的独立于帧的部分，将与帧无关的部分输入到高斯模型，以及计算帧概率。然后根据帧概率计算自回归系数。

2. 发明申请

WO2002091358A1 METHOD AND APPARATUS FOR REJECTION OF SPEECH RECOGNITION RESULTS IN ACCORDANCE WITH CONFIDENCE LEVEL 审中-公开
标题翻译：根据信心级别拒绝语音识别结果的方法和装置
公开(公告)号：WO2002091358A1
公开(公告)日：2002-11-14
申请号：PCT/CN2001/000685
申请日：2001-05-08
申请人： INTEL CORPORATION , INTEL CHINA LTD. , PI, Xiaobo , JIA, Ying
发明人： PI, Xiaobo , JIA, Ying
IPC分类号： G10L15/14
CPC分类号： G10L15/14
摘要： An automatic speech recognition system for continuous speech recognition of vocabulary words for an autoattendent system proving hand-free telephone calling and utilizing a vocabulary comprising numbers or names of people to be called using known techniques for automatic speech recognition models of word sequencing resulting in high confidence levels of recognition.
摘要翻译：一种自动语音识别系统，用于自动人事系统的词汇词的连续语音识别，证明免提电话呼叫并利用包括使用已知技术的人数或姓名的词汇，以使用字序列的自动语音识别模型导致高置信度认可水平

3. 发明申请

WO2002091355A1 HIGH-ORDER ENTROPY ERROR FUNCTIONS FOR NEURAL CLASSIFIERS 审中-公开
标题翻译：神经分类器的高阶入侵错误功能
公开(公告)号：WO2002091355A1
公开(公告)日：2002-11-14
申请号：PCT/CN2001/000686
申请日：2001-05-08
申请人： INTEL CORPORATION , INTEL CHINA LTD. , PI, Xiaobo , JIA, Ying
发明人： PI, Xiaobo , JIA, Ying
IPC分类号： G10L15/00
CPC分类号： G10L15/16 , G10L15/063 , G10L15/10
摘要： An automatic speech recognition system comprising a speech decoder to resolve phone and world level information, a vector generator to generate information vectors on which a confidence measure is based by a neural network classifier (ANN). An error signal is designed which is not subject to false saturation or over specialization. The error signal is integrated into an error function which is back propagated through the ANN.
摘要翻译：一种自动语音识别系统，包括用于解析电话和世界级信息的语音解码器，矢量生成器，用于生成由神经网络分类器（ANN）基于置信度量度的信息矢量。设计出不会产生假饱和或过度专业化的误差信号。误差信号被集成到通过ANN反向传播的误差函数中。

4. 发明申请

WO2007115428A1 MULTI-MODE ULTRASONIC SYSTEM 审中-公开
标题翻译：多模超声波系统
公开(公告)号：WO2007115428A1
公开(公告)日：2007-10-18
申请号：PCT/CN2006/000579
申请日：2006-03-31
申请人： INTEL CORPORATION , HE, Xiaoying, Janet , JIA, Ying
发明人： HE, Xiaoying, Janet , JIA, Ying
IPC分类号： G06F3/033
CPC分类号： G06F3/0433 , G06F3/03545 , G06F3/0488 , G06F3/04883
摘要： In one embodiment, an apparatus comprises a switching mechanism to switch an input line between a first input voltage and a second input voltage, apulse generator coupled to the switching mechanism to generate an electronic pulse train at a high frequency in response to the first input voltage and an electronic pulse train at a low frequency in response to the second input voltage, an oscillator circuit coupled to the pulse generator to receive the electronic pulse train, and an ultrasonic transmitter coupled to the oscillator circuit to produce an ultrasonic signal at a frequency that is a function of a frequency of the electronic pulse train.
摘要翻译：在一个实施例中，一种装置包括切换机构，用于在第一输入电压和第二输入电压之间切换输入线，耦合到开关机构的脉冲发生器响应于第一输入电压而产生高频的电子脉冲串以及响应于第二输入电压的低频的电子脉冲串，耦合到脉冲发生器以接收电子脉冲串的振荡器电路，以及耦合到振荡器电路的超声波发射器，以产生超声波信号，频率为是电子脉冲串的频率的函数。

5. 发明申请

WO2007059645A1 SIGNAL TRANSITION FEATURE BASED CODING FOR SERIAL LINK 审中-公开
标题翻译：基于信号转换特征的串行编码
公开(公告)号：WO2007059645A1
公开(公告)日：2007-05-31
申请号：PCT/CN2005/001975
申请日：2005-11-22
申请人： INTEL CORPORATION , HE, Xiaoying , LIANG, Luhong , JIA, Ying
发明人： HE, Xiaoying , LIANG, Luhong , JIA, Ying
IPC分类号： G06F5/00
CPC分类号： G11B20/10 , H04L25/4906
摘要： Signal transition feature based coding for serial link is described herein. According to one embodiment, in response to a data stream to be transmitted onto a serial communication link, one or more bits of the data stream are encoded according to a bit order determined based on a frequency of signal transitions of the data stream. As a result, a sequence of encoded data stream having a lower number of bit transitions with respect to the frequency of signal transitions of the data stream prior to the encoding is generated. Thereafter, the encoded data sequence is transmitted onto the serial communication link. Other methods and apparatuses are also described.
摘要翻译：本文描述了用于串行链路的基于信号转换特征的编码。根据一个实施例，响应于要发送到串行通信链路的数据流，数据流的一个或多个比特根据基于数据流的信号转换频率确定的比特顺序进行编码。结果，产生具有相对于编码之前的数据流的信号转换频率的较低数量位转换的编码数据流序列。此后，将编码数据序列发送到串行通信链路。还描述了其它方法和装置。

6. 发明申请

WO2006133327A1 ACOUSTIC SENSOR WITH COMBINED FREQUENCY RANGES 审中-公开
标题翻译：具有组合频率范围的声学传感器
公开(公告)号：WO2006133327A1
公开(公告)日：2006-12-14
申请号：PCT/US2006/022201
申请日：2006-06-06
申请人： INTEL CORPORATION , JIA, Ying , HE, Janet , MEINSCHEIN, Robert
发明人： JIA, Ying , HE, Janet , MEINSCHEIN, Robert
IPC分类号： G06F3/033
CPC分类号： G06F3/0433
摘要： A hybrid acoustic sensor can include a first acoustic sensor, a second acoustic sensor, and a mixer. The first acoustic sensor can generate a first signal to represent acoustic data in a first bandwidth around a first center frequency. The second acoustic sensor can be co-located with the first acoustic sensor, and can generate a second signal to represent the acoustic data in a second bandwidth around a second center frequency. A lower bound of the first bandwidth can be at a lower frequency than a lower bound of the second bandwidth, and a higher bound of the second bandwidth can be at a higher frequency than a higher bound of the first bandwidth. The mixer can combine the first signal and the second signal into a third signal to represent the acoustic data in a third bandwidth from the lower frequency to the higher frequency.
摘要翻译：混合声学传感器可以包括第一声学传感器，第二声学传感器和混合器。第一声传感器可以产生第一信号以在围绕第一中心频率的第一带宽中表示声学数据。第二声传感器可以与第一声传感器共定位，并且可以产生第二信号以在围绕第二中心频率的第二带宽中表示声学数据。第一带宽的下限可以处于比第二带宽的下限更低的频率，并且第二带宽的较高边界可以处于比第一带宽的较高边界更高的频率。混频器可以将第一信号和第二信号组合成第三信号，以在从较低频率到较高频率的第三带宽中表示声学数据。

7. 发明申请

WO2006131022A1 ULTRASONIC TRACKING 审中-公开
标题翻译：超声追踪
公开(公告)号：WO2006131022A1
公开(公告)日：2006-12-14
申请号：PCT/CN2005/000801
申请日：2005-06-07
申请人： INTEL CORPORATION , JIA, Ying
发明人： JIA, Ying
IPC分类号： G06F3/033
CPC分类号： G06F3/0433
摘要： In one embodiment, a tracking module (110) to select a plurality of pairs of digital ultrasonic signals to form two or more pairs of digital ultrasonic signals and estimate time difference of arrival (TDOA) for each of the two or more pairs of digital ultrasonic signals. The tracking module (110) may further determine an intersection of each pair of the TDOA estimated digital ultrasonic signals to form one or more intersections and a location of an ultrasonic signal generator (102) corresponding to at least one of the one or more intersections.
摘要翻译：在一个实施例中，跟踪模块（110）选择多对数字超声波信号以形成两对或更多对数字超声波信号，并估计两对或更多对数字超声波中的每一对的时间到达差（TDOA）信号。跟踪模块（110）还可以确定每对TDOA估计的数字超声波信号的交点以形成对应于一个或多个交叉点中的至少一个的一个或多个交叉点和超声信号发生器（102）的位置。

8. 发明申请

WO2006131021A1 ERROR DETECTION AND PREVENTION IN ACOUSTIC DATA 审中-公开
标题翻译：声音数据中的错误检测和预防
公开(公告)号：WO2006131021A1
公开(公告)日：2006-12-14
申请号：PCT/CN2005/000800
申请日：2005-06-07
申请人： INTEL CORPORATION , HU, Yongge , JIA, Ying
发明人： HU, Yongge , JIA, Ying
IPC分类号： G06F3/033
CPC分类号： G06F3/043 , G06F3/0383
摘要： A stream of raw acoustic data can be received at a client device. The client device can frame the stream of raw acoustic data at particular intervals with alignment information to create framed acoustic data, and buffer the framed acoustic data while waiting for a data request from a host device. In response to receiving the data request, the client device can provide the framed acoustic data to the host device.
摘要翻译：可以在客户端设备处接收原始声学数据流。客户端设备可以以具有对准信息的特定间隔来形成原始声学数据流，以产生成帧的声学数据，并且在等待来自主机设备的数据请求的同时缓冲成帧的声学数据。响应于接收到数据请求，客户端设备可以向框架的声音数据提供主机设备。

9. 发明申请

WO2018094676A1 WEARABLE ARTICLE HAVING ELASTIC BELT 审中-公开
公开(公告)号：WO2018094676A1
公开(公告)日：2018-05-31
申请号：PCT/CN2016/107188
申请日：2016-11-25
申请人： THE PROCTER & GAMBLE COMPANY , MORIMOTO, Koichi , TAMEISHI, Kazuaki , LEE, DonSub , YAMASHITA, Fumitake , JIA, Ying , LI, Shiqiao
发明人： MORIMOTO, Koichi , TAMEISHI, Kazuaki , LEE, DonSub , YAMASHITA, Fumitake , JIA, Ying , LI, Shiqiao
IPC分类号： A61F13/49
摘要： A wearable article (20) continuous in a longitudinal direction and a transverse direction comprises a center chassis (38) and a ring-like elastic belt (40) comprising a front belt (84) and a back belt (86). The front belt (84) and the back belt (86) each is formed by a laminate comprising a plurality of elastic bodies (96) running in the transverse direction, an inner sheet (94), an outer sheet (92), and an outer sheet fold over (93). At least one of the front belt (84) and the back belt (86) has: a) the longitudinal edge (38LE) of the center chassis (38) being distal than, preferably at least about 10mm distal than, the outer sheet fold over proximal edge (93PE); b) in the left and right side panels (82): b-1) at least one elastic body (96), preferably at least two elastic bodies (96), disposed between the outer sheet (92) and the outer sheet fold over (93) in regions where the inner sheet (94) does not overlap in the thickness direction; and b-2) the remaining elastic bodies (96) disposed between the outer sheet (92) and the inner sheet (94) in regions where the outer sheet fold over (93) does not overlap in the thickness direction.

10. 发明申请

WO2018094674A1 METHOD OF MAKING WEARABLE ARTICLE COMPRISING ELASTIC BELT 审中-公开
公开(公告)号：WO2018094674A1
公开(公告)日：2018-05-31
申请号：PCT/CN2016/107182
申请日：2016-11-25
申请人： THE PROCTER & GAMBLE COMPANY , TAMEISHI, Kazuaki , LEE, DonSub , YAMASHITA, Fumitake , JIA, Ying
发明人： TAMEISHI, Kazuaki , LEE, DonSub , YAMASHITA, Fumitake , JIA, Ying
IPC分类号： A61F13/15 , A61F13/49
摘要： Disclosed is a method of manufacturing a wearable article (20) assembled by a front belt (84), a back belt (86), and a central chassis (38), the central chassis (38) having a front waist region (26) and a back waist region (28) separated from each other by a crotch region (30), wherein at least one of the front belt (84) and the back belt (86) is formed by an elastic belt (40). The method mainly comprises making the elastic belt (40) and joining to the central chassis (38) by several steps.

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