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

US20130322645A1 DATA RECOGNITION AND SEPARATION ENGINE 审中-公开
标题翻译：数据识别和分离发动机
公开(公告)号：US20130322645A1
公开(公告)日：2013-12-05
申请号：US13915443
申请日：2013-06-11
申请人： CYPHER, LLC
发明人： Tyson LaVar Edwards
IPC分类号： H04R3/00
CPC分类号： H04R3/00 , G06F16/683 , G06F16/783 , G06K9/00744 , G10L25/51
摘要： Embodiments disclosed herein extend to methods, systems, and computer program products for analyzing digital data. A source of digital data is analyzed and separated into segments, each segment having an identifiable characteristic. The separated segments are copied into planes of a higher dimension. The separated segments are compared to determine a resemblance factor. A fingerprint is generated for segments having a resemblance factor above a particular threshold. Based upon the generated fingerprint, a data source may be filtered to block or to pass data corresponding to the generated fingerprint. The digital data may be audio data, video data, or other data.
摘要翻译：本文公开的实施例扩展到用于分析数字数据的方法，系统和计算机程序产品。数字数据的来源被分析并分成段，每段具有可识别的特征。分离的段被复制到更高维度的平面中。将分离的段进行比较以确定相似因子。生成具有高于特定阈值的相似因子的段的指纹。基于生成的指纹，可以对数据源进行滤波以阻止或传递与所生成的指纹相对应的数据。数字数据可以是音频数据，视频数据或其他数据。

2. 发明申请

US20160093313A1 NEURAL NETWORK VOICE ACTIVITY DETECTION EMPLOYING RUNNING RANGE NORMALIZATION 有权
标题翻译：神经网络语音活动检测运行范围正常化
公开(公告)号：US20160093313A1
公开(公告)日：2016-03-31
申请号：US14866824
申请日：2015-09-25
申请人： CYPHER, LLC
发明人： Earl Vickers
IPC分类号： G10L21/0264 , G10L21/0224 , G10L25/84 , G10L25/30 , G10L25/60
CPC分类号： G10L21/0264 , G10L21/0224 , G10L25/30 , G10L25/60 , G10L25/78 , G10L25/84 , G10L2015/0636
摘要： A “running range normalization” method includes computing running estimates of the range of values of features useful for voice activity detection (VAD) and normalizing the features by mapping them to a desired range. Running range normalization includes computation of running estimates of the minimum and maximum values of VAD features and normalizing the feature values by mapping the original range to a desired range. Smoothing coefficients are optionally selected to directionally bias a rate of change of at least one of the running estimates of the minimum and maximum values. The normalized VAD feature parameters are used to train a machine learning algorithm to detect voice activity and to use the trained machine learning algorithm to isolate or enhance the speech component of the audio data.
摘要翻译： “运行范围归一化”方法包括计算对语音活动检测（VAD）有用的特征值的范围的运行估计，并且通过将它们映射到期望的范围来对特征进行归一化。运行范围归一化包括计算VAD特征的最小值和最大值的运行估计值，并通过将原始范围映射到所需范围来对特征值进行归一化。任选地选择平滑系数来定向地偏置最小值和最大值的运行估计中的至少一个的变化率。归一化的VAD特征参数用于训练机器学习算法以检测语音活动，并使用经过训练的机器学习算法来隔离或增强音频数据的语音分量。

3. 发明申请

US20160133272A1 ADAPTIVE INTERCHANNEL DISCRIMINATIVE RESCALING FILTER 有权
标题翻译：自适应通道间分辨率滤波器
公开(公告)号：US20160133272A1
公开(公告)日：2016-05-12
申请号：US14938816
申请日：2015-11-11
申请人： Cypher, LLC
发明人： Erik Sherwood , Carl Grundstrom
IPC分类号： G10L21/0232 , G10L21/0264
CPC分类号： G10L21/0232 , G10L21/0208 , G10L25/84 , G10L2021/02165
摘要： A method for adjusting a degree of filtering applied to an audio signal includes modeling a probability density function (PDF) of a fast Fourier transform (FFT) coefficient of a primary channel and reference channel of the audio signal; maximizing at least one of PDFs to provide a discriminative relevance difference (DRD) between a noise magnitude estimate of the reference channel and a noise magnitude estimate of the primary channel. The method further includes emphasizing the primary channel when the spectral magnitude of the primary channel is stronger than the spectral magnitude of the reference channel; and deemphasizing the primary channel when the spectral magnitude of the reference channel is stronger than the spectral magnitude of the primary channel. The emphasizing and deemphasizing includes computing a multiplicative rescaling factor and applying the multiplicative rescaling factor to a gain computed in a prior stage of a speech enhancement filter chain when there is a prior stage, and directly applying a gain when there is no prior stage.
摘要翻译：用于调整应用于音频信号的滤波程度的方法包括对音频信号的主要信道和参考信道的快速傅立叶变换（FFT）系数的概率密度函数（PDF）进行建模; 使至少一个PDF最大化以在参考信道的噪声幅度估计和主信道的噪声幅度估计之间提供鉴别相关性差异（DRD）。该方法还包括当主信道的频谱幅度比参考信道的频谱幅度更强时强调主信道; 并且当参考信道的频谱幅度比主信道的频谱幅度更强时，对主信道进行去加重。强调和不强调包括计算乘法重定标因子，并且在存在先前阶段时将乘法重定标因子应用于在语音增强滤波器链的先前阶段计算的增益，并且当不存在前级时直接应用增益。

4. 发明申请

US20150373453A1 MULTI-AURAL MMSE ANALYSIS TECHNIQUES FOR CLARIFYING AUDIO SIGNALS 审中-公开
标题翻译：用于清除音频信号的多重MMSE分析技术
公开(公告)号：US20150373453A1
公开(公告)日：2015-12-24
申请号：US14308541
申请日：2014-06-18
申请人： Cypher, LLC
发明人： Fredrick D. Geiger , Bryant V. Bunderson , Carl Grundstrom
IPC分类号： H04R3/00 , H04R1/08
CPC分类号： H04R3/00 , G10L21/02 , G10L25/27 , G10L2021/02165 , H04R2410/05 , H04R2499/11
摘要： Techniques for processing audio signals include removing noise from the audio signals or otherwise clarifying the audio signals prior to outputting the audio signals. The disclosed techniques may employ minimum mean squared error (MMSE) analyses on audio signals received from a primary microphone and at least one reference microphone, and to techniques in which the MMSE analyses are used to reduce or eliminate noise from audio signals received by the primary microphone. Optionally, confidence intervals may be assigned to different frequency bands of an audio signal, with each confidence interval corresponding to a likelihood that its respective frequency band includes targeted audio, and each confidence interval representing a contribution of its respective frequency band in a reconstructed audio signal from which noise has been removed.
摘要翻译：用于处理音频信号的技术包括在输出音频信号之前从音频信号中去除噪声或以其它方式澄清音频信号。所公开的技术可以对从主麦克风和至少一个参考麦克风接收的音频信号采用最小均方误差（MMSE）分析，以及使用MMSE分析来减少或消除由主要麦克风接收的音频信号的噪声的技术麦克风。可选地，可以将置信区间分配给音频信号的不同频带，每个置信区间对应于其相应频带包括目标音频的似然性，并且每个置信区间表示重构音频信号中其各自频带的贡献从哪个噪音已被删除。

5. 发明申请

US20160134984A1 DETERMINING NOISE AND SOUND POWER LEVEL DIFFERENCES BETWEEN PRIMARY AND REFERENCE CHANNELS 审中-公开
标题翻译：确定主要和参考通道之间的噪音和声压级别差异
公开(公告)号：US20160134984A1
公开(公告)日：2016-05-12
申请号：US14938798
申请日：2015-11-11
申请人： CYPHER, LLC
发明人： Jan S. Erkelens
IPC分类号： H04R29/00 , G10L25/21 , G10L21/0232 , G10L25/12
CPC分类号： G10L21/0232 , G10L25/12 , G10L25/21 , G10L2021/02165 , H04R3/005 , H04R2410/05
摘要： A method for estimating a noise power level difference (NPLD) between a primary microphone and a reference microphone of an audio device includes obtaining primary and reference channels of an audio signal with primary and reference microphones of an audio device and estimating a noise magnitude of the reference channel of the audio signal to provide a noise variance estimate for one or more frequencies. A modelled probability density function (PDF) of a fast Fourier transform (FFT) coefficient of the primary channel of the audio signal is maximized to provide a NPLD between the noise variance estimate of the reference channel and a noise variance estimate of the primary channel. A modelled PDF of an FFT coefficient of the reference channel of the audio signal is maximized to provide a complex speech power level difference (SPLD) coefficient between the speech FFT coefficients of the primary and reference channel. A corrected noise magnitude of the reference channel is then calculated based on the noise variance estimate, the NPLD and the SPLD coefficient.
摘要翻译：用于估计音频设备的主麦克风和参考麦克风之间的噪声功率电平差（NPLD）的方法包括：利用音频设备的主要和参考麦克风来获得音频信号的主要参考信道和参考信道，并估计音频信号的噪声幅度音频信号的参考通道，以提供一个或多个频率的噪声方差估计。将音频信号的主要信道的快速傅里叶变换（FFT）系数的建模概率密度函数（PDF）最大化，以在参考信道的噪声方差估计和主信道的噪声方差估计之间提供NPLD。将音频信号的参考信道的FFT系数的建模PDF最大化，以在主信道和参考信道的语音FFT系数之间提供复合语音功率电平差（SPLD）系数。然后基于噪声方差估计，NPLD和SPLD系数来计算参考信道的校正噪声幅度。

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