公开(公告)号：WO2004086631A3

公开(公告)日：2005-06-23

申请号：PCT/US2004008950

申请日：2004-03-23

Applicant: QUALCOMM INC ,  GARUDADRI HARINATH ,  RAMCHANDRAN KANNAN

Inventor： GARUDADRI HARINATH ,  RAMCHANDRAN KANNAN

IPC: H03M13/29 ,  H04N7/26 ,  H04N7/66 ,  H04N7/68

CPC classification number: H03M13/6513 ,  H03M13/29 ,  H04N19/37

Abstract: According to one aspect of the present invention, a method and apparatus is provided in which input data (e.g., input video data) is encoded in accordance with a first coding standard (e.g., MPEG-4) to generate encoded data. The input data is also encoded based on a reconstruction of the input data to generate encoded side information associated with the input data. The encoded data are transmitted to a destination (e.g., a decoding subsystem) over a first channel and the encoded side information are transmitted to the destination over a second channel. The encoded data and the encoded side information are decoded and combined at the destination to generate output data.

Abstract translation: 根据本发明的一个方面，提供了一种方法和装置，其中根据第一编码标准（例如，MPEG-4）对输入数据（例如，输入视频数据）进行编码以产生编码数据。 输入数据也是基于输入数据的重构来编码的，以产生与输入数据相关联的编码侧信息。 经编码的数据通过第一信道发送到目的地（例如，解码子系统），经编码的侧信息通过第二信道发送到目的地。 编码数据和编码侧信息在目的地被解码和组合以产生输出数据。

公开(公告)号：WO02059871A3

公开(公告)日：2003-03-13

申请号：PCT/US0200890

申请日：2002-01-10

Applicant: QUALCOMM INC

Inventor： GARUDADRI HARINATH

IPC: G10L15/00 ,  G10L15/06 ,  G10L15/12 ,  G10L15/14 ,  H03M7/50 ,  G10L15/28 ,  H03M7/30

CPC classification number: G10L15/06

Abstract: A method and system that improves voice recognition by improving storage of voice recognition (VR) templates. The improved storage means that more VR models can be stored in memory. The more VR models that are stored in memory, the more robust the VR system and therefore the more accurate the VR system. Lossy compression techniques are used to compress VR models. In one embodiment, A-law compression and A-law expansion are used to compress and expand VR models. In another embodiment, Mu-law compression and Mu-law expansion are used to compress and expand VR models. VR models are compressed during a training process and they are expanded during voice recognition.

Abstract translation: 通过改进语音识别（VR）模板的存储来改善语音识别的方法和系统。 改进的存储意味着更多的VR型号可以存储在存储器中。 存储在存储器中的VR模型越多，VR系统越强大，因此VR系统越准确。 有损压缩技术用于压缩VR模型。 在一个实施例中，使用A律压缩和A律扩展来压缩和扩展VR模型。 在另一个实施例中，Mu法压缩和Mu法扩展用于压缩和扩展VR模型。 VR模型在训练过程中被压缩，并且在语音识别期间被扩展。

公开(公告)号：WO0250504A2

公开(公告)日：2002-06-27

申请号：PCT/US0147761

申请日：2001-12-13

Applicant: QUALCOMM INC

Inventor： GARUDADRI HARINATH ,  DEJACO ANDREW P ,  CHANG CHIENCHUNG

IPC: G10L15/28 ,  G01L15/00

CPC classification number: G10L15/30

Abstract: A novel and improved method and an accompanying apparatus provide for a distributed voice recognition (VR) capability in a remote device (201). Remote device (201) decides and controls what portions of the VR processing may take place at remote device (201) and what other portions may take place at a base station (202) in wireless communication with remote device (201).

Abstract translation: 一种新颖且改进的方法和伴随装置提供远程设备（201）中的分布式语音识别（VR）能力。 远程设备（201）决定并控制在远程设备（201）处可能发生的VR处理的哪些部分，以及在与远程设备（201）进行无线通信的基站（202）可能发生什么其他部分。

公开(公告)号：WO2015153150A3

公开(公告)日：2015-11-26

申请号：PCT/US2015021711

申请日：2015-03-20

Applicant: QUALCOMM INC

Inventor： ZHENG THOMAS JIAQIAN ,  SHAW JEFFREY CLINTON ,  GARUDADRI HARINATH

IPC: G06N3/04

CPC classification number: G06N3/049 ,  G06N3/0472

Abstract: A method of using spiking neural network delays to represent sequences includes assigning one or more symbol neurons to each symbol in a dictionary. The method also includes assigning a synapse from each symbol neuron in a group to a particular ngram neuron. A set of synapses associated with the group of symbol neurons comprises a bundle of synapses. In addition, the method includes assigning a delay to each synapse in the bundle. The method further includes representing a symbol sequence based on sequential spiking of symbol neurons and ngram neuron spikes in response to detecting inter event intervals.

Abstract translation: 使用尖峰神经网络延迟来表示序列的方法包括将一个或多个符号神经元分配给字典中的每个符号。 该方法还包括将来自组中每个符号神经元的突触分配给特定的ngram神经元。 与该组符号神经元相关联的一组突触包括一束突触。 另外，该方法包括为束中的每个突触分配延迟。 该方法还包括响应于检测到事件间间隔，基于符号神经元和ngram神经元尖峰的顺序尖峰来表示符号序列。

公开(公告)号：WO2012012189A3

公开(公告)日：2012-04-05

申请号：PCT/US2011042432

申请日：2011-06-29

Applicant: QUALCOMM INC ,  GARUDADRI HARINATH ,  BAHETI PAWAN KUMAR ,  MAJUMDAR SOMDEB

Inventor： GARUDADRI HARINATH ,  BAHETI PAWAN KUMAR ,  MAJUMDAR SOMDEB

IPC: A61B5/053 ,  H03M1/12 ,  H03M7/30

CPC classification number: A61B5/0531 ,  A61B5/0022 ,  A61B5/0024 ,  A61B5/02416 ,  A61B5/0402 ,  A61B5/0536 ,  A61B5/0816 ,  A61B5/14551 ,  A61B5/7232 ,  H03M7/30

Abstract: Certain aspects of the present disclosure relate to techniques for measuring body impedance based on baseband signal detection in analog domain. Proposed methods and apparatus are able to measure an impedance of human body based on sub-Nyquist sampling of signals. The proposed techniques can be particularly beneficial for reducing overall sensor power when an actuation signal generates electrical signals corresponding to vital signs in humans.

Abstract translation: 本公开的某些方面涉及基于模拟域中的基带信号检测来测量身体阻抗的技术。 提出的方法和装置能够基于子奈奎斯特信号采样来测量人体的阻抗。 当致动信号产生对应于人类生命体征的电信号时，所提出的技术可以特别有利于降低总传感器功率。

公开(公告)号：WO2010036894A3

公开(公告)日：2010-07-08

申请号：PCT/US2009058383

申请日：2009-09-25

Applicant: QUALCOMM INC ,  GARUDADRI HARINATH ,  BAHETI PAWAN KUMAR

Inventor： GARUDADRI HARINATH ,  BAHETI PAWAN KUMAR

IPC: H03M7/30 ,  A61B5/024 ,  A61B5/1455 ,  G06T11/00

CPC classification number: A61B5/0006 ,  A61B5/0024 ,  A61B5/021 ,  A61B5/14551 ,  A61B5/7232 ,  A61B2560/0209 ,  G06K9/00496 ,  H03M7/30

Abstract: Certain aspects of the present disclosure relate to a method for compressed sensing (CS). The CS is a signal processing concept wherein significantly fewer sensor measurements than that suggested by Shannon/Nyquist sampling theorem can be used to recover signals with arbitrarily fine resolution. In this disclosure, the CS framework is applied for sensor signal processing in order to support low power robust sensors and reliable communication in Body Area Networks (BANs) for healthcare and fitness applications.

Abstract translation: 本公开的某些方面涉及压缩感测（CS）的方法。 CS是一种信号处理概念，其中可以使用比由香农/奈奎斯特采样定理所建议的更少的传感器测量值来以任意精细的分辨率恢复信号。 在本公开中，CS框架被应用于传感器信号处理，以便支持用于医疗和健身应用的身体局域网（BAN）中的低功率鲁棒传感器和可靠的通信。

公开(公告)号：WO2007115127A2

公开(公告)日：2007-10-11

申请号：PCT/US2007065548

申请日：2007-03-29

Applicant: QUALCOMM INC ,  REZNIK YURIY ,  LUDWIN ALBERT SCOTT ,  CHUNG HYUKJUNE ,  GARUDADRI HARINATH ,  SRINIVASAMURTHY NAVEEN B ,  SAGETONG PHOOM

Inventor： REZNIK YURIY ,  LUDWIN ALBERT SCOTT ,  CHUNG HYUKJUNE ,  GARUDADRI HARINATH ,  SRINIVASAMURTHY NAVEEN B ,  SAGETONG PHOOM

IPC: G06F17/14

CPC classification number: G06F17/14 ,  G06F17/147 ,  H04N19/42 ,  H04N19/61

Abstract: Techniques for efficiently performing full and scaled transforms on data received via full and scaled interfaces, respectively, are described. A full transform is a transform that implements the complete mathematical description of the transform. A full transform operates on or provides full transform coefficients. A scaled transform is a transform that operates on or provides scaled transform coefficients, which are scaled versions of the full transform coefficients. The scaled transform may have lower computational complexity whereas the full transform may be simpler to use by applications. The full and scaled transforms may be for a 2D IDCT, which may be implemented in a separable manner with 1D IDCTs. The full and scaled transforms may also be for a 2D DCT, which may be implemented in a separable manner with 1D DCTs. The 1D IDCTs and 1D DCTs may be implemented in a computationally efficient manner.

Abstract translation: 描述了分别对通过完整和缩放界面接收的数据进行全面和缩放变换的技术。 全变换是实现变换的完整数学描述的变换。 完全变换在全变换系数上运行或提供全变换系数。 缩放变换是对或者提供缩放变换系数的变换，其是全变换系数的缩放版本。 缩放变换可能具有较低的计算复杂度，而全变换可能更易于由应用使用。 完整和缩放的变换可以用于2D IDCT，其可以用1D IDCT以可分离的方式实现。 完整和缩放变换也可以用于2D DCT，其可以用1D DCT以可分离的方式实现。 1D IDCT和1D DCT可以以计算有效的方式实现。

公开(公告)号：WO02059874A3

公开(公告)日：2002-12-19

申请号：PCT/US0200183

申请日：2002-01-02

Applicant: QUALCOMM INC

Inventor： GARUDADRI HARINATH

IPC: G10L15/30 ,  G10L15/10 ,  G10L15/14 ,  G10L15/26 ,  G10L15/28

CPC classification number: G10L15/30

Abstract: A method and system that improves voice recognition in a distributed voice recognition system. A distributed voice recognition system 50 includes a local VR engine 52 in a subscriber unit 54 and a server VR engine 56 on a server 58. When the local VR engine 52 does not recognize a speeh segment segment to the local VR engine 56 downloads information corresponding the speech segment to the local VR engine 52. The local VR engine 52 may combine its speech segement information with downloaded information to create resultant information for a speech segment. The local VR engine 52 may also apply a function to downloaded information to create resultant information for a speech segment. The local VR engine 52 may also apply a function to downloaded information to create resultant information. Resultant information then may be uploaded from the local VR engine 52 to the server VR engine 56.

Abstract translation: 一种改进分布式语音识别系统中语音识别的方法和系统。 分布式语音识别系统50包括在用户单元54中的本地VR引擎52和服务器58上的服务器VR引擎56.当本地VR引擎52不识别到本地VR引擎56的语音片段时，下载相应的信息 语音段到本地VR引擎52.本地VR引擎52可以将其语音分段信息与下载的信息组合，以产生用于语音段的结果信息。 本地VR引擎52还可以对下载的信息应用功能，以产生语音段的结果信息。 本地VR引擎52还可以应用下载信息的功能以产生结果信息。 然后可以将结果信息从本地VR引擎52上传到服务器VR引擎56。

公开(公告)号：WO2008086509A3

公开(公告)日：2008-11-27

申请号：PCT/US2008050822

申请日：2008-01-10

Applicant: QUALCOMM INC ,  GARUDADRI HARINATH ,  SUNDARRAMAN CHANDRU ,  VIJ GAJINDER ,  SRINIVASAMURTHY NAVEEN B ,  SAGETONG PHOOM

Inventor： GARUDADRI HARINATH ,  SUNDARRAMAN CHANDRU ,  VIJ GAJINDER ,  SRINIVASAMURTHY NAVEEN B ,  SAGETONG PHOOM

IPC: H04L12/801 ,  H04L29/06

CPC classification number: H04L65/4092 ,  H04L47/10 ,  H04L65/1059 ,  H04L65/602 ,  H04L65/80 ,  H04W72/08

Abstract: This disclosure describes techniques that can facilitate multimedia telephony. In one example, a method for communication of multimedia data comprises determining a first level of throughput associated with multimedia data communication from a first access terminal to a network, determining a second level of throughput associated with multimedia data communication from the network to a second access terminal based on feedback from the second access terminal to the first access terminal via the network, determining a budget associated with communication of a video unit of the multimedia data, and coding the video unit of the multimedia data based on the budget and the first and second levels of throughput.

Abstract translation: 本公开描述了可以促进多媒体电话的技术。 在一个示例中，用于通信多媒体数据的方法包括确定与从第一接入终端到网络的多媒体数据通信相关联的吞吐量的第一级别，确定与从网络到第二接入的多媒体数据通信相关联的吞吐量的第二级别 基于从所述第二接入终端经由所述网络到所述第一接入终端的反馈，确定与所述多媒体数据的视频单元的通信相关联的预算，以及基于所述预算对所述多媒体数据的视频单元进行编码，以及 第二级吞吐量。

公开(公告)号：WO2008039871A3

公开(公告)日：2008-07-31

申请号：PCT/US2007079595

申请日：2007-09-26

Applicant: QUALCOMM INC ,  CHENG MINGXIA ,  MAURO ANTHONY PATRICK II ,  CHOY EDDIE L T ,  GAO YUJIE ,  SAMPAT KUNTAL DILIPSINH ,  ZIVNEY MATTHEW BLAINE ,  GOVERDHAN SATISH ,  GUPTA SAMIR KUMAR ,  GARUDADRI HARINATH

Inventor： CHENG MINGXIA ,  MAURO ANTHONY PATRICK II ,  CHOY EDDIE L T ,  GAO YUJIE ,  SAMPAT KUNTAL DILIPSINH ,  ZIVNEY MATTHEW BLAINE ,  GOVERDHAN SATISH ,  GUPTA SAMIR KUMAR ,  GARUDADRI HARINATH

IPC: H04N7/24 ,  H04N19/89

CPC classification number: H04N21/2368 ,  G10L19/005 ,  H04N19/436 ,  H04N19/68 ,  H04N19/89 ,  H04N21/4307 ,  H04N21/4341 ,  H04N21/439 ,  H04N21/44204 ,  H04N21/4425 ,  H04N21/8106

Abstract: A method and system for resynchronizing an embedded multimedia system using bytes consumed in an audio decoder. The bytes consumed provides a mechanism to compensate for bit error handling and correction in a system that does not require re-transmission. The audio decoder keeps track of the bytes consumed and periodically reports the bytes consumed. A host microprocessor indexes the actual bytes consumed since bit errors may have been handled or corrected to a predetermined byte count to determine whether resynchronization is necessary.

Abstract translation: 一种使用音频解码器中消耗的字节重新同步嵌入式多媒体系统的方法和系统。 消耗的字节提供了一种补偿不需要重新传输的系统中的位错误处理和校正的机制。 音频解码器跟踪消耗的字节，并定期报告消耗的字节。 主机微处理器索引所消耗的实际字节，因为位错误可能被处理或校正到预定的字节数以确定是否需要重新同步。