-
公开(公告)号：US07080008B2
公开(公告)日：2006-07-18
申请号：US10843011
申请日：2004-05-11
申请人： Hao Jiang , Hong-Jiang Zhang
发明人： Hao Jiang , Hong-Jiang Zhang
IPC分类号： G10L11/06
摘要： A portion of an audio signal is separated into multiple frames from which one or more different features are extracted. These different features are used, in combination with a set of rules, to classify the portion of the audio signal into one of multiple different classifications (for example, speech, non-speech, music, environment sound, silence, etc.). In one embodiment, these different features include one or more of line spectrum pairs (LSPs), a noise frame ratio, periodicity of particular bands, spectrum flux features, and energy distribution in one or more of the bands. The line spectrum pairs are also optionally used to segment the audio signal, identifying audio classification changes as well as speaker changes when the audio signal is speech.
摘要翻译： 音频信号的一部分被分成多个帧，从中提取一个或多个不同的特征。 使用这些不同的特征，结合一组规则，将音频信号的一部分分类成多个不同分类（例如，语音，非语音，音乐，环境声音，静音等）之一。 在一个实施例中，这些不同的特征包括一个或多个频带中的一个或多个线路频谱对（LSP），噪声帧比，特定频带的周期性，频谱通量特征以及能量分布。 线谱对也可选地用于分割音频信号，识别音频分类改变以及当音频信号是语音时的扬声器变化。
-
2.发明授权Method and apparatus for compression and decompression of digital image data 失效用于数字图像数据的压缩和解压缩的方法和装置
公开(公告)号：US5347600A
公开(公告)日：1994-09-13
申请号：US781586
申请日：1991-10-23
申请人： Michael F. Barnsley , Alan D. Sloan
发明人： Michael F. Barnsley , Alan D. Sloan
摘要： Digital image data is automatically processed by dividing stored image data into domain blocks and range blocks. The range blocks are subjected to processes such as a shrinking process to obtain mapped range blocks. The range blocks or domain blocks may also be processed by processes such as affine transforms. Then, for each domain block, the mapped range block which is most similar to the domain block is determined, and the address of that range block and the processes the blocks were subjected to are combined as an identifier which is appended to a list of identifiers for other domain blocks. The list of identifiers for all domain blocks is called a fractal transform and constitutes a compressed representation of the input image. To decompress the fractal transform and recover the input image, an arbitrary input image is formed into range blocks and the range blocks processed in a manner specified by the identifiers to form a representation of the original input image.
摘要翻译： 通过将存储的图像数据分割成域块和范围块来自动处理数字图像数据。 范围块经受诸如收缩处理的处理以获得映射的范围块。 范围块或域块也可以由诸如仿射变换之类的过程来处理。 然后，对于每个域块，确定与域块最相似的映射范围块，并且该范围块的地址和块所经历的处理被组合为附加到标识符列表的标识符 对于其他域块。 所有域块的标识符列表称为分形变换，并构成输入图像的压缩表示。 为了解压缩分形变换并恢复输入图像，将任意输入图像形成为范围块，并且以由标识符指定的方式处理的范围块以形成原始输入图像的表示。
-
3.发明申请NOISE SPEED-UPS IN HIDDEN MARKOV MODELS WITH APPLICATIONS TO SPEECH RECOGNITION 审中-公开噪音速度型UPS用于语音识别应用公开(公告)号：US20160005399A1
公开(公告)日：2016-01-07
申请号：US14802760
申请日：2015-07-17
申请人： Kartik Audhkhasi , Osonde Osoba , Bart Kosko
发明人： Kartik Audhkhasi , Osonde Osoba , Bart Kosko
CPC分类号： G10L15/144 , G10L15/063 , G10L25/33 , G10L25/36
摘要： A learning computer system may estimate unknown parameters and states of a stochastic or uncertain system having a probability structure. The system may include a data processing system that may include a hardware processor that has a configuration that: receives data; generates random, chaotic, fuzzy, or other numerical perturbations of the data, one or more of the states, or the probability structure; estimates observed and hidden states of the stochastic or uncertain system using the data, the generated perturbations, previous states of the stochastic or uncertain system, or estimated states of the stochastic or uncertain system; and causes perturbations or independent noise to be injected into the data, the states, or the stochastic or uncertain system so as to speed up training or learning of the probability structure and of the system parameters or the states.
摘要翻译： 学习计算机系统可以估计具有概率结构的随机或不确定系统的未知参数和状态。 该系统可以包括数据处理系统，其可以包括具有以下配置的硬件处理器：接收数据; 产生数据的随机，混乱，模糊或其他数字扰动，一个或多个状态或概率结构; 使用数据，生成的扰动，随机或不确定系统的先前状态或随机或不确定系统的估计状态的随机或不确定系统的观测和隐藏状态的估计; 并引起扰动或独立噪声注入到数据，状态或随机或不确定系统中，以加速对概率结构和系统参数或状态的训练或学习。
-
4.发明申请Apparatus for predicting the spectral information of voice signals and a method therefor 审中-公开用于预测语音信号的频谱信息的装置及其方法公开(公告)号：US20070011001A1
公开(公告)日：2007-01-11
申请号：US11483890
申请日：2006-07-10
申请人： Hyun-Soo Kim
发明人： Hyun-Soo Kim
IPC分类号： G10L19/00
摘要： Disclosed is a method for predicting the spectral information of voice signals, including inputting the voice signals, performing morphological operations with the waveform image of the voice signals, extracting harmonic peaks as a result of the morphological operations, and predicting the spectral envelope information of the voice signals by interpolating the harmonic peaks.
摘要翻译： 公开了一种用于预测语音信号的频谱信息的方法，包括输入语音信号，用语音信号的波形图像进行形态学运算，作为形态学运算的结果提取谐波峰值，以及预测频谱包络信息 通过内插谐波峰值的语音信号。
-
公开(公告)号：US20050060152A1
公开(公告)日：2005-03-17
申请号：US10974298
申请日：2004-10-27
申请人： Hao Jiang , Hongjiang Zhang
发明人： Hao Jiang , Hongjiang Zhang
摘要： A portion of an audio signal is separated into multiple frames from which one or more different features are extracted. These different features are used, in combination with a set of rules, to classify the portion of the audio signal into one of multiple different classifications (for example, speech, non-speech, music, environment sound, silence, etc.). In one embodiment, these different features include one or more of line spectrum pairs (LSPs), a noise frame ratio, periodicity of particular bands, spectrum flux features, and energy distribution in one or more of the bands. The line spectrum pairs are also optionally used to segment the audio signal, identifying audio classification changes as well as speaker changes when the audio signal is speech.
-
公开(公告)号：US20060178877A1
公开(公告)日：2006-08-10
申请号：US11278250
申请日：2006-03-31
申请人： Hao Jiang , Hongjiang Zhang
发明人： Hao Jiang , Hongjiang Zhang
IPC分类号： G10L19/00
摘要： A portion of an audio signal is separated into multiple frames from which one or more different features are extracted. These different features are used, in combination with a set of rules, to classify the portion of the audio signal into one of multiple different classifications (for example, speech, non-speech, music, environment sound, silence, etc.). In one embodiment, these different features include one or more of line spectrum pairs (LSPs), a noise frame ratio, periodicity of particular bands, spectrum flux features, and energy distribution in one or more of the bands. The line spectrum pairs are also optionally used to segment the audio signal, identifying audio classification changes as well as speaker changes when the audio signal is speech.
摘要翻译： 音频信号的一部分被分成多个帧，从中提取一个或多个不同的特征。 使用这些不同的特征，结合一组规则，将音频信号的一部分分类成多个不同分类（例如，语音，非语音，音乐，环境声音，静音等）之一。 在一个实施例中，这些不同的特征包括一个或多个频带中的一个或多个线路频谱对（LSP），噪声帧比，特定频带的周期性，频谱通量特征以及能量分布。 线谱对也可选地用于分割音频信号，识别音频分类改变以及当音频信号是语音时的扬声器变化。
-
公开(公告)号：US06901362B1
公开(公告)日：2005-05-31
申请号：US09553166
申请日：2000-04-19
申请人： Hao Jiang , Hongjiang Zhang
发明人： Hao Jiang , Hongjiang Zhang
摘要： A portion of an audio signal is separated into multiple frames from which one or more different features are extracted. These different features are used, in combination with a set of rules, to classify the portion of the audio signal into one of multiple different classifications (for example, speech, non-speech, music, environment sound, silence, etc.). In one embodiment, these different features include one or more of line spectrum pairs (LSPs), a noise frame ratio, periodicity of particular bands, spectrum flux features, and energy distribution in one or more of the bands. The line spectrum pairs are also optionally used to segment the audio signal, identifying audio classification changes as well as speaker changes when the audio signal is speech.
摘要翻译： 音频信号的一部分被分成多个帧，从中提取一个或多个不同的特征。 使用这些不同的特征，结合一组规则，将音频信号的一部分分类成多个不同分类（例如，语音，非语音，音乐，环境声音，静音等）之一。 在一个实施例中，这些不同的特征包括一个或多个频带中的一个或多个线路频谱对（LSP），噪声帧比，特定频带的周期性，频谱通量特征以及能量分布。 线谱对也可选地用于分割音频信号，识别音频分类改变以及当音频信号是语音时的扬声器变化。
-
公开(公告)号：US20050075863A1
公开(公告)日：2005-04-07
申请号：US10998766
申请日：2004-11-29
申请人： Hao Jiang , Hongjiang Zhang
发明人： Hao Jiang , Hongjiang Zhang
摘要： A portion of an audio signal is separated into multiple frames from which one or more different features are extracted. These different features are used, in combination with a set of rules, to classify the portion of the audio signal into one of multiple different classifications (for example, speech, non-speech, music, environment sound, silence, etc.). In one embodiment, these different features include one or more of line spectrum pairs (LSPs), a noise frame ratio, periodicity of particular bands, spectrum flux features, and energy distribution in one or more of the bands. The line spectrum pairs are also optionally used to segment the audio signal, identifying audio classification changes as well as speaker changes when the audio signal is speech.
-
公开(公告)号：US5768437A
公开(公告)日：1998-06-16
申请号：US295637
申请日：1994-08-26
摘要： A method of fractal coding of data and apparatus therefor, which method comprises dividing data into domains, determining a set of transformations relating the domains to the data in such a manner as to minimize error between the data and an approximation to the data obtained by application of the transformation, and providing an expression of a series of quantized fractal coefficients characterizing the transformations. A transformation includes at least one part indicating a domain and at least another (functional) part indicating a value for a measure associatable with a specific domain or aspect thereof.
摘要翻译： PCT No.PCT / GB93 / 00422 Sec。 371日期1994年8月26日 102（e）日期1994年8月26日PCT 1993年3月1日PCT公布。 公开号WO93 / 17519 日期1993年9月2日一种数据及其装置的分形编码方法，该方法包括将数据划分成域，确定一组与数据相关的变换，以使数据之间的误差最小化， 通过应用变换获得的数据，并提供表征变换的一系列量化分形系数的表达式。 变换包括指示域的至少一个部分和指示与特定域或其方面相关联的度量的值的至少另一（功能）部分。
-
公开(公告)号：US20090240508A1
公开(公告)日：2009-09-24
申请号：US12363293
申请日：2009-01-30
申请人： Junichi Saito
发明人： Junichi Saito
IPC分类号： G10L19/00
摘要： A sampling rate conversion apparatus and a method thereof are provided which increase the sampling rate of a discrete audio signal sampled at a predetermined sampling rate by using a fractal interpolation function (FIF). An audio signal portion formed by a predetermined number of sampling data items is divided into a plurality of interpolation intervals. On the audio signal portion, mapping points are determined. The number of the mapping points is in accordance with the degree of increase in the sampling rate. For the respective interpolation intervals, mapping parameters for performing mapping using the FIF on the mapping points are calculated. In all of the interpolation intervals, the mapping using the FIF is performed on the mapping points with the use of the mapping parameters according to the respective interpolation intervals. Thereby, new sampling data items are generated.
摘要翻译： 提供了一种采样率转换装置及其方法，其通过使用分形插值函数（FIF）来增加以预定采样率采样的离散音频信号的采样率。 由预定数量的采样数据项形成的音频信号部分被分成多个插值间隔。 在音频信号部分，确定映射点。 映射点的数量符合采样率的增加程度。 对于各个内插间隔，计算使用映射点上的FIF执行映射的映射参数。 在所有内插间隔中，使用FIF的映射使用映射参数根据各自的内插间隔在映射点上进行。 由此，生成新的采样数据项。
-
-
-
-
-
-
-
-
-
搜索结果
18 条记录 (耗时 0.017 秒)