公开(公告)号：US20050261893A1

公开(公告)日：2005-11-24

申请号：US10344624

申请日：2002-06-11

申请人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji ,  Masayuki Nishiguchi

发明人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji ,  Masayuki Nishiguchi

IPC分类号： G10L19/00 ,  G10L19/02 ,  G10L19/035 ,  G10L19/038 ,  H03M7/30 ,  H04N19/00 ,  H04N19/124 ,  H04N19/154 ,  H04N19/33 ,  H04N19/46 ,  H04N19/463 ,  H04N19/70 ,  H04N19/94

CPC分类号： G10L19/032 ,  G10L19/02 ,  G10L19/06 ,  H03M7/30

摘要： In a quantization step information encoding unit, an average value of the quantization step information is found in an approximate shape extraction unit (20), first of all, from one set of a given number of unitary quantization units to another. In an approximate shape encoding unit (21), the approximate shape information is vector-quantized. In a residual signal computing unit (22), the residual signals between the quantization step information and the quantized approximate shape vector are computed. In a residual signal encoding unit (23), the residual signals are variable length encoded, and the so encoded residual signals and the vector quantized approximate shape information are output.

摘要翻译： 在量化步长信息编码单元中，在近似形状提取单元（20）中找到量化步长信息的平均值，首先，从一组给定数量的单位量化单元到另一组。 在近似形状编码单元（21）中，近似形状信息被矢量量化。 在残差信号计算单元（22）中，计算量化步长信息和量化近似形状矢量之间的残差信号。 在剩余信号编码单元（23）中，剩余信号是可变长度编码的，并且输出如此编码的残差信号和矢量量化的近似形状信息。

公开(公告)号：US07212973B2

公开(公告)日：2007-05-01

申请号：US10344624

申请日：2002-06-11

申请人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji ,  Masayuki Nishiguchi

发明人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji ,  Masayuki Nishiguchi

IPC分类号： G10L19/02 ,  G06K9/38

CPC分类号： G10L19/032 ,  G10L19/02 ,  G10L19/06 ,  H03M7/30

摘要： In a quantization step information encoding unit, an average value of the quantization step information is found in an approximate shape extraction unit (20), first of all, from one set of a given number of unitary quantization units to another. In an approximate shape encoding unit (21), the approximate shape information is vector-quantized. In a residual signal computing unit (22), the residual signals between the quantization step information and the quantized approximate shape vector are computed. In a residual signal encoding unit (23), the residual signals are variable length encoded, and the so encoded residual signals and the vector quantized approximate shape information are output.

摘要翻译： 在量化步长信息编码单元中，在近似形状提取单元（20）中找到量化步长信息的平均值，首先，从一组给定数量的单位量化单元到另一组。 在近似形状编码单元（21）中，近似形状信息被矢量量化。 在残差信号计算单元（22）中，计算量化步长信息和量化近似形状矢量之间的残差信号。 在剩余信号编码单元（23）中，剩余信号是可变长度编码的，并且输出如此编码的残差信号和矢量量化的近似形状信息。

公开(公告)号：US07627482B2

公开(公告)日：2009-12-01

申请号：US11951056

申请日：2007-12-05

申请人： Minoru Tsuji ,  Shiro Suzuki ,  Keisuke Toyama

发明人： Minoru Tsuji ,  Shiro Suzuki ,  Keisuke Toyama

IPC分类号： G10L19/00 ,  G10L21/00 ,  G10L21/04 ,  H03M7/00

CPC分类号： G10L19/008

摘要： A sound signal encoder for high efficiency encoding of sound signals from a plurality of channels is provided which includes a to-be-correlated object setter (52), to-be-correlated object selector (56) and a variable-length encoder (58). The to-be-correlated object setter (52) sets, on the basis of left-channel frequency information held in a left-channel frequency information holder (50) and right-channel frequency information held in a right-channel frequency information holder (51), index [i] indicating which ones of sine waves on the left channel are to be correlated with, namely, are to be subtracted from, sine waves on the right channel. The to-be-correlated object selector (56) selects a default value read from a storage unit (55) or index [i]-th amplitude information read from a left-channel amplitude information holder (53) as an object to be subtracted from the i-th amplitude information on the right channel according to the index [i]. The variable-length encoder (58) makes variable-length encoding of a difference resulted from subtraction of the left-channel amplitude information or default value as the to-be-correlated object from the amplitude information on the right channel.

摘要翻译： 提供了一种用于对来自多个通道的声音信号进行高效编码的声音信号编码器，其包括待相关对象设定器（52），待相关对象选择器（56）和可变长度编码器（58​​） ）。 要相关对象设定器（52）基于保持在左声道频率信息保持器（50）中的左声道频率信息和保持在右声道频率信息保持器（50）中的右声道频率信息， 51），指示左侧信道上哪个正弦波与哪个相关联的索引[i]，即从右声道上的正弦波中减去。 待相关对象选择器（56）从作为要减去的对象中选择从存储单元（55）读取的默认值或从左声道幅度信息保持器（53）读取的索引[i]个幅度信息 根据索引[i]从右声道上的第i幅度信息。 可变长度编码器（58​​）对从右声道的振幅信息中减去左声道幅度信息或默认值作为要相关对象产生的差的可变长度编码。

公开(公告)号：US07583804B2

公开(公告)日：2009-09-01

申请号：US10534175

申请日：2003-10-10

申请人： Shiro Suzuki ,  Minoru Tsuji ,  Keisuke Toyama

发明人： Shiro Suzuki ,  Minoru Tsuji ,  Keisuke Toyama

IPC分类号： H04R5/00

CPC分类号： G10L19/028

摘要： In an audio-information encoding apparatus, in order to encode an audio signal containing a white-noise component, an index iL indicating the energy level of the white-noise component and an index iR designating the start index of a random-number table are introduced into a code train. In an audio-information decoding apparatus (20), a white-noise generating unit (25) uses the indices iL and iR contained in the code train, thereby generating a white-noise signal Sw(t) on the time axis, which has the same level as the white noise, and an adder (26) adds the white-noise signal to an audio signal Sf(t) decoded on the time axis, outputting as an output audio signal So(t).

摘要翻译： 在音频信息编码装置中，为了对包含白噪声分量的音频信号进行编码，指示白噪声分量的能级的索引iL和指定随机数表的起始索引的索引iR是 引入代码列表。 在音频信息解码装置（20）中，白噪声产生单元（25）使用代码串中包含的索引iL和iR，从而在时间轴上产生白噪声信号Sw（t） 与白噪声相同，并且加法器26将白噪声信号与在时间轴上解码的音频信号Sf（t）相加，作为输出音频信号So（t）输出。

公开(公告)号：US20060153402A1

公开(公告)日：2006-07-13

申请号：US10534175

申请日：2003-10-10

申请人： Shiro Suzuki ,  Minoru Tsuji ,  Keisuke Toyama

发明人： Shiro Suzuki ,  Minoru Tsuji ,  Keisuke Toyama

IPC分类号： H03G5/00

CPC分类号： G10L19/028

摘要： In an audio-information encoding apparatus, in order to encode an audio signal containing a white-noise component, an index iL indicating the energy level of the white-noise component and an index iR designating the start index of a random-number table are introduced into a code train. In an audio-information decoding apparatus (20), a white-noise generating unit (25) uses the indices iL and iR contained in the code train, thereby generating a white-noise signal Sw(t) on the time axis, which has the same level as the white noise, and an adder (26) adds the white-noise signal to an audio signal Sf(t) decoded on the time axis, outputting as an output audio signal So(t).

摘要翻译： 在音频信息编码装置中，为了对包含白噪声分量的音频信号进行编码，指示白噪声分量的能级的索引iL和指定随机数表的起始索引的索引iR是 引入代码列表。 在音频信息解码装置（20）中，白噪声产生单元（25）使用代码串中包含的索引iL和iR，从而在时间轴上产生白噪声信号Sw（t） 与白噪声相同，并且加法器26将白噪声信号与在时间轴上解码的音频信号Sf（t）相加，作为输出音频信号So（t）输出。

公开(公告)号：US06995699B2

公开(公告)日：2006-02-07

申请号：US10432924

申请日：2002-09-26

申请人： Keisuke Toyama ,  Minoru Tsuji ,  Shiro Suzuki

发明人： Keisuke Toyama ,  Minoru Tsuji ,  Shiro Suzuki

IPC分类号： H03M7/30

CPC分类号： G06F7/38 ,  G06F2207/3832 ,  G10L19/00 ,  G10L19/02 ,  G10L2019/0004 ,  H03M1/121

摘要： In an encoding apparatus, when difference value between adjacent quantization units of quantization accuracy information where, e.g., distribution range is 0˜7, e.g., if difference value is 3 or more, 8 is subtracted, and if difference value is less than −4, 8 is added to thereby transform two difference values where difference therebetween is 8 into the same value. Thus, the distribution range of difference value becomes −4˜3, and size of code book (table) can be held down to the same size in the case where difference is not taken. In addition, high order 1 bit of difference value may be masked to carry out replacement into value consisting of only low order 3 bits, thus also making it possible to prevent increase in size of code book (table).

摘要翻译： 在编码装置中，当减去量化精度信息的相邻量化单位之间的差值（例如分布范围为0〜7），例如，如果差值为3以上，则减去8，如果差值小于-4 ，8，从而将两个差分值变换为相同的值。 因此，差分值的分布范围变为-4〜3，并且在不采用差异的情况下，可以将码本（表）的大小保持为相同的大小。 此外，可以对高阶1比特的差值进行掩码以将替换执行为仅由低3位组成的值，从而也可以防止代码簿（表）的大小增加。

公开(公告)号：US06850179B2

公开(公告)日：2005-02-01

申请号：US10344842

申请日：2002-06-14

申请人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji

发明人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji

IPC分类号： H03M7/00

CPC分类号： G10L19/032 ,  G10L19/0204 ,  G10L19/0208

摘要： Information is encoded into a code with a smaller number of bits. An encoding apparatus encodes an acoustic time series signal such that the acoustic time series signal is split into band signals in predetermined bands and gain adjustment is made on the band signals, at a gain control position by a gain control amount, for each of as many positions as indicated by a gain control number. Although the gain control number can take any one of values from 0 to 7, it has a high probability of taking a particular value (0, for example). The gain control number is encoded such that a code with a small number of bits is assigned to the gain control number equal to a value having a high-occurrence probability. The encoding is also applied to a voice recording apparatus.

摘要翻译： 信息被编码成具有较少位数的代码。 编码装置对声学时间序列信号进行编码，使得声学时间序列信号被分割成预定频带的频带信号，并且在增益控制位置对增益控制位置对增益控制位置进行增益调整，对于每个频带信号 位置，如增益控制编号所示。 虽然增益控制号码可以取0到7之间的任一值，但它具有获取特定值的可能性很高（例如0）。 对增益控制编号进行编码，使得具有较少位数的代码被分配给等于具有高出现概率的值的增益控制数。 该编码也被应用于语音记录装置。

公开(公告)号：US07933417B2

公开(公告)日：2011-04-26

申请号：US11324633

申请日：2006-01-03

申请人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji

发明人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji

IPC分类号： H04R5/00 ,  G10L19/00 ,  G10L19/14 ,  H03M7/00 ,  H03M7/34 ,  H04B14/04

CPC分类号： H04B1/667 ,  G10L19/0208

摘要： The present invention relates to an encoding device for saving the number of bits of codes. In step S11, the differential value between a normalization coefficient Bi to be encoded and a normalization coefficient Bi-1 for an encoding unit Ai-1 in a band adjacent to the lower side of an encoding unit Ai corresponding to the normalization coefficient Bi is computed. In step S12, reference is made to a table in which a differential value having a high frequency of occurrence is associated with a code having a small number of bits, and a code corresponding to the computed differential value is read. In step S13, it is determined whether or not all normalization coefficients B have been encoded. If it is determined that all normalization coefficients B have been encoded, in step S14, the code read in step S12 is output. The present invention is applicable to an audio recorder.

摘要翻译： 本发明涉及一种用于节省代码比特数的编码装置。 在步骤S11中，计算要编码的归一化系数Bi与与归一化系数Bi相对应的编码单元Ai的下侧的频带中的编码单位Ai-1的归一化系数Bi-1的差分值 。 在步骤S12中，参考具有高出现频率的差分值与具有少量位的代码相关联的表，并且读取与所计算出的差分值相对应的代码。 在步骤S13中，确定所有归一化系数B是否已被编码。 如果确定所有归一化系数B已被编码，则在步骤S14中，输出在步骤S12中读取的代码。 本发明可应用于音频记录器。

公开(公告)号：US07366661B2

公开(公告)日：2008-04-29

申请号：US10203733

申请日：2001-12-14

申请人： Minoru Tsuji ,  Shiro Suzuki ,  Keisuke Toyama

发明人： Minoru Tsuji ,  Shiro Suzuki ,  Keisuke Toyama

IPC分类号： G10L11/00 ,  G10L13/00 ,  G10L19/00

CPC分类号： G10L19/093 ,  G10L19/022

摘要： The present invention relates to an information extraction apparatus capable of analyzing an acoustic signal with accuracy and high efficiency.An amplitude analysis section 32 determines whether or not an attack or release is contained on the basis of an amplitude value for each small region of an input time-series signal. When it is determined that there is an attack or release, an analysis region setting section 33 sets the portion from an attack position to a release position as an analysis region. A frequency analysis section 34 analyzes the input time-series signal by generalized harmonic analysis and outputs extracted waveform information. An extracted waveform synthesis section 35 synthesizes the extracted waveform information and outputs the information to a time-series compensation section 36. The time-series compensation section 36 compensates the signal of the synthesized result with a signal outside the analysis region and outputs an extracted waveform time-series signal to a subtraction unit 37. The subtraction unit 37 generates a residual time-series signal from the input time-series signal and the extracted waveform time-series signal. The present invention can be applied to various audio apparatuses, voice recognition apparatuses, voice synthesis apparatuses, etc., for processing an acoustic signal.

摘要翻译： 信息提取装置技术领域本发明涉及能够精确，高效地分析声音信号的信息提取装置。 振幅分析部分32基于输入时间序列信号的每个小区域的幅度值来确定是否包含攻击或释放。 当确定存在攻击或释放时，分析区域设置部33将来自攻击位置的部分设置为释放位置作为分析区域。 频率分析部分34通过广义谐波分析分析输入的时间序列信号并输出​​提取的波形信息。 提取的波形合成部分35合成提取的波形信息并将其输出到时间序列补偿部分36.时间序列补偿部分36用分析区域外的信号补偿合成结果的信号，并输出提取的波形 时间序列信号发送到减法单元37.减法单元37根据输入时间序列信号和提取的波形时间序列信号产生残留时间序列信号。 本发明可以应用于用于处理声信号的各种音频装置，语音识别装置，语音合成装置等。

公开(公告)号：US20060115092A1

公开(公告)日：2006-06-01

申请号：US11324633

申请日：2006-01-03

申请人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji

发明人： Keisuke Toyama ,  Shiro Suzuki ,  Minoru Tsuji

IPC分类号： H04R5/00 ,  G10L21/00 ,  G10L19/00

CPC分类号： H04B1/667 ,  G10L19/0208

摘要： The present invention relates to an encoding device for saving the number of bits of codes. In step S11, the differential value between a normalization coefficient Bi to be encoded and a normalization coefficient Bi-1 for an encoding unit Ai-1 in a band adjacent to the lower side of an encoding unit Ai corresponding to the normalization coefficient Bi is computed. In step S12, reference is made to a table in which a differential value having a high frequency of occurrence is associated with a code having a small number of bits, and a code corresponding to the computed differential value is read. In step S13, it is determined whether or not all normalization coefficients B have been encoded. If it is determined that all normalization coefficients B have been encoded, in step S14, the code read in step S12 is output. The present invention is applicable to an audio recorder.

摘要翻译： 本发明涉及一种用于节省代码比特数的编码装置。 在步骤S11中，要编码的归一化系数B i i i与编码单元A i-1的归一化系数B i-1 <1的差分值< 计算与归一化系数B＆lt; I＆gt;对应的编码单元A的下侧的频带中的/ SUB。 在步骤S12中，参考具有高出现频率的差分值与具有少量位的代码相关联的表，并且读取与所计算出的差分值相对应的代码。 在步骤S13中，确定所有归一化系数B是否已被编码。 如果确定所有归一化系数B已被编码，则在步骤S14中，输出在步骤S12中读取的代码。 本发明可应用于音频记录器。