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. 发明专利

JP2013157960A Information processing device and method for processing acoustic signal of the same 审中-公开
标题翻译：信息处理装置及其处理声音信号的方法
公开(公告)号：JP2013157960A
公开(公告)日：2013-08-15
申请号：JP2012019326
申请日：2012-01-31
申请人： Toshiba Corp , 株式会社東芝
发明人： AMADA KO , YAMAMOTO TOSHIFUMI , MISEKI KIMIO
IPC分类号： H04R3/04
摘要： PROBLEM TO BE SOLVED: To provide an information processing device and an acoustic signal processing method that can correct the characteristic of an earphone with low cost and simple work.SOLUTION: An information processing device comprises: measuring means for measuring the frequency characteristic of an earphone connected to an output terminal from an audio signal for measurement which is output from the output terminal for outputting the audio signal and which is collected by a microphone; correction filter design means for designing a correction filter to correct the frequency characteristic of the audio output from the earphone to a predetermined target frequency characteristic regarding one sound range of a high sound range higher than a crossover zone, which is a frequency zone switching from a low sound range to the high sound range, or the low sound range lower than the crossover zone, on the basis of the frequency characteristic of the earphone measured by the measuring means and the predetermined target frequency characteristic; and adjusting means for adjusting volume balance between the high sound range and the low sound range on the basis of designation from outside.
摘要翻译：要解决的问题：提供可以以低成本和简单的工作来校正耳机的特性的信息处理装置和声音信号处理方法。解决方案：信息处理装置包括：用于测量耳机的频率特性的测量装置从用于输出音频信号的输出端子输出并由麦克风收集的用于测量的音频信号连接到输出端子; 校正滤波器设计装置，用于设计校正滤波器，以将来自耳机的音频输出的频率特性校正为关于高于跨越区域的高声音范围的一个声音范围的预定目标频率特性，该交替区域是从基于由测量装置测量的耳机的频率特性和预定的目标频率特性，低声音范围到高声音范围，或低于交叉区域的低声音范围; 以及用于根据来自外部的指定来调节高声音范围和低声音范围之间的音量平衡的调节装置。

2. 发明专利

JP2012003277A Decoding device and spectrum shaping method 有权
标题翻译：解码器件和光谱形成方法
公开(公告)号：JP2012003277A
公开(公告)日：2012-01-05
申请号：JP2011172221
申请日：2011-08-05
申请人： Toshiba Corp , 株式会社東芝
发明人： MISEKI KIMIO
IPC分类号： G10L19/14
摘要： PROBLEM TO BE SOLVED: To provide a decoding device and a spectrum shaping method in which muffled sound and fluctuation of sound caused by excessive spectrum enhancement are reduced, and subjective quality is improved, thereby facilitating high quality spectrum shaping.SOLUTION: A composition filter coefficient generating section 10 generates a composition filter coefficient used in a composition filter section 50 and a post filter section 60 from a coding data for a composition filter. A sound source generating section 20 generates a sound source signal from a coding data for a sound source signal. A sound source analysis section 30 acquires a flattening parameter for flattening a spectrum of the sound source signal based on the sound source signal generated by the sound source generating section 20. A sound source spectrum flattening section 40 uses the flattening parameter to perform the spectrum shaping so as to flatten excessive slope and roughness of the spectrum of the sound source signal given by the sound source generating section 20.
摘要翻译：解决的问题：提供一种解码装置和频谱整形方法，其中减少了由过度频谱增强引起的消声和声音波动，并提高了主观质量，从而促进了高质量的频谱整形。解决方案：组合滤波器系数产生部分10从组合滤波器的编码数据产生在组合滤波器部分50和后滤波器部分60中使用的合成滤波器系数。声源生成部20从声源信号的编码数据生成声源信号。声源分析部30基于由声源生成部20生成的声源信号，获取平坦化声源信号的频谱的平坦化参数。声源谱平坦化部40使用平坦化参数进行频谱整形以便平坦化由声源生成部分20给出的声源信号的频谱的过大的斜率和粗糙度。（C）2012，JPO和INPIT

3. 发明专利

JP2011237826A Decoding device and method for spectrum shaping 有权
标题翻译：解码设备和用于光谱形成的方法
公开(公告)号：JP2011237826A
公开(公告)日：2011-11-24
申请号：JP2011172222
申请日：2011-08-05
申请人： Toshiba Corp , 株式会社東芝
发明人： MISEKI KIMIO
IPC分类号： G10L19/14 , G10L19/12
摘要： PROBLEM TO BE SOLVED: To provide a decoding device and a method for easily shaping spectrum allowing high-quality spectrum shaping that improves subjective quality by reducing a furry feeling and an unstable feeling in a sound due to excessively emphasized spectrum.SOLUTION: A composite filter coefficient generating section 10 generates a composite filter coefficient used in a composite filter section 50 and a post filter section 60 from coded data for the composite filter. A sound source generating section 20 generates a sound source signal from coded data for the sound source signal. A sound source analyzing section 30 calculates a flattening parameter to flatten the spectrum of the sound source signal based on the generated sound source signal in the sound source generating section 20. Then, a sound source spectrum flattening section 40 shapes the spectrum to flatten an excessive gradient or an unevenness of the sound source signal spectrum given from the sound source generating section 20 by employing the flattening parameter.
摘要翻译：解决方案：提供一种解码装置和方法，用于容易地整形频谱，从而通过减少由于过分强调的频谱引起的毛茸茸的感觉和不稳定的声音感觉而提高主观质量的高质量频谱成形。解决方案：复合滤波器系数产生部10根据复合滤波器的编码数据，生成复合滤波器部分50和后滤波器部分60中使用的复合滤波器系数。声源生成部20从声源信号的编码数据生成声源信号。声源分析部30基于声源生成部20中生成的声源信号，计算平坦化的声源信号的频谱的平坦化参数。然后，声源谱平坦化部40将频谱变平，通过使用平坦化参数从声源生成部20给出的声源信号频谱的梯度或不平坦度。版权所有（C）2012，JPO＆INPIT

4. 发明专利

JP2011234047A Audio signal correcting device, method of the same, and software recording medium 有权
标题翻译：音频信号校正装置，其方法和软件记录介质
公开(公告)号：JP2011234047A
公开(公告)日：2011-11-17
申请号：JP2010101387
申请日：2010-04-26
申请人： Toshiba Corp , 株式会社東芝
发明人： MISEKI KIMIO , CHIBA NORIKATSU , KANISHIMA YASUHIRO
IPC分类号： H04R3/04
CPC分类号： H04R3/04
摘要： PROBLEM TO BE SOLVED: To provide an acoustic correcting apparatus by which a sense of incongruity due to occlusion caused by headphones or earphones is eliminated.SOLUTION: The acoustic signal processing device includes an input means and a correcting means. The input means receives input specifying information that specifies a first frequency related to resonance at the time of occlusion by wearing earphones or headphones. In the correcting means, on the basis of the specifying information or the first frequency, correction is performed on the acoustic signal such that a predetermined second frequency is enhanced as a frequency related to resonance at the time of release from the occlusion. In the correction by the correcting means, the second frequency is enhanced in a range that is not less than 2 dB and not greater than 12 dB.
摘要翻译：要解决的问题：提供一种消除由耳机或耳机引起的由于遮挡引起的不协调感的声学校正装置。解决方案：声信号处理装置包括输入装置和校正装置。输入装置接收输入指定信息，该信息指定通过佩戴耳机或耳机在闭塞时与谐振有关的第一频率。在校正装置中，基于指定信息或第一频率，对声信号执行校正，使得预定的第二频率作为与从闭塞释放时的共振相关的频率增强。在校正装置的校正中，第二频率在不小于2dB且不大于12dB的范围内增强。版权所有（C）2012，JPO＆INPIT

5. 发明专利

JP2011176830A Acoustic processor and acoustic processing method 审中-公开
标题翻译：声学处理器和声音处理方法
公开(公告)号：JP2011176830A
公开(公告)日：2011-09-08
申请号：JP2011044244
申请日：2011-03-01
申请人： Toshiba Corp , 株式会社東芝
发明人： CHIBA NORIKATSU , MISEKI KIMIO
IPC分类号： H04R3/04 , G10K15/00 , H04R1/10
摘要： PROBLEM TO BE SOLVED: To provide an acoustic processor and an acoustic processing method capable of suppressing a resonance phenomenon induced by sealing an ear canal and capable of easily improving the quality of listening sound.
SOLUTION: The acoustic processor includes: a generation means for generating test signals for which an amplitude value at one of frequency positions included in a specified frequency band is corrected; a selection receiving means for acoustically outputting the test signals at the frequency positions for which the amplitude value is corrected comparably and receiving the selection of one test signal from the acoustically outputted test signals; and a sound quality correction means for correcting sound quality by using a setting related to the correction at the frequency position corresponding to the test signal received by the selection receiving means.
COPYRIGHT: (C)2011,JPO&INPIT
摘要翻译：解决的问题：提供能够抑制通过密封耳道而引起的共振现象并且能够容易地提高听力音质的声学处理器和声学处理方法。解决方案：声学处理器包括：产生装置，用于产生校正在指定频带中包括的一个频率位置处的振幅值的测试信号; 选择接收装置，用于在相对于振幅值被校正的频率位置声学输出测试信号，并从声输出的测试信号接收一个测试信号的选择; 以及声音质量校正装置，用于通过使用与由选择接收装置接收的测试信号相对应的频率位置处的校正相关的设置来校正声音质量。版权所有（C）2011，JPO＆INPIT

6. 发明专利

JP2008164823A Audio data processor 审中-公开
标题翻译：音频数据处理器
公开(公告)号：JP2008164823A
公开(公告)日：2008-07-17
申请号：JP2006352916
申请日：2006-12-27
申请人： Toshiba Corp , 株式会社東芝
发明人： OSADA SHIYOUKO , TAKEUCHI HIROKAZU , MISEKI KIMIO
IPC分类号： G10L19/02 , G10L11/00 , G10L19/00 , H03M7/40
CPC分类号： G10L19/008
摘要： PROBLEM TO BE SOLVED: To provide an audio data processor shortening the time needed to analyze features of audio data. SOLUTION: The audio data processor is provided with: an M/S stereo decision unit which decides whether M/S stereo is applied to each scale band by dequantized quantized frequency range audio data, and extracts and outputs frequency-range audio data of an S channel in a scale factor band for the scale factor band to which the M/S stereo is applied or generates and outputs frequency range audio data of an S channel for a scale factor band to which the M/S stereo is not applied by using frequency range audio data of L and R channels in the scale factor band; and a feature analyzing unit which analyzes features of audio encoded data by using frequency range audio data of the S channel. COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：要解决的问题：提供音频数据处理器，缩短分析音频数据特征所需的时间。解决方案：音频数据处理器具有：M / S立体声判定单元，其通过反量化量化频率范围音频数据判定是否将M / S立体声应用于每个音阶，并提取并输出频率范围音频数据在施加或产生M / S立体声的比例因子频带的比例因子频带中的S频道的频率范围音频数据输出到不应用M / S立体声的比例因子频带的S频道的频率范围音频数据通过使用比例因子频带中的L和R声道的频率范围音频数据; 以及特征分析单元，其通过使用S信道的频率范围音频数据来分析音频编码数据的特征。版权所有（C）2008，JPO＆INPIT

7. 发明专利

JP2008059383A Content presentation device and computer program 审中-公开
标题翻译：内容介绍设备和计算机程序
公开(公告)号：JP2008059383A
公开(公告)日：2008-03-13
申请号：JP2006236855
申请日：2006-08-31
申请人： Toshiba Corp , 株式会社東芝
发明人： SUGIURA CHIKASHI , ISAKA TAKEHIKO , MISEKI KIMIO
IPC分类号： G06F17/30
CPC分类号： G06F17/30029 , G06F17/30053 , Y10S707/99943
摘要： PROBLEM TO BE SOLVED: To successively calculate a statistical quantity for retrieving a content desired to be added to a play list on the basis of determination whether or not the content can be added to the play list and to automatically prepare a play list a user really desires and quickly on the basis of the statistical quantity.
SOLUTION: The content presentation device includes: a candidate content retrieval statistical quantity calculation part 5 for calculating a candidate content retrieval statistical quantity; a candidate content retrieval part 6 for retrieving a candidate content by using the candidate content retrieval statistical quantity; a candidate content display part 7a for displaying the candidate content; a part 7b for selecting contents to be deleted for selecting an undesired content; and a candidate content retrieval statistical quantity calculation part 5 for calculating the desired content retrieval statistical quantity by using the candidate content retrieval statistical quantity, the undesired content, and a temporarily desired content with an arbitrary content which is not selected as the undesired content as the temporarily desired content.
COPYRIGHT: (C)2008,JPO&INPIT
摘要翻译：要解决的问题：基于确定是否可以将内容添加到播放列表并且自动准备播放列表来连续地计算用于检索期望添加到播放列表的内容的统计量用户在统计数量的基础上真正需要并快速。解决方案：内容呈现装置包括：候选内容检索统计量计算部分5，用于计算候选内容检索统计量; 候选内容检索部分6，用于通过使用候选内容检索统计量来检索候选内容; 候选内容显示部分7a，用于显示候选内容; 用于选择要删除的内容以选择不期望的内容的部分7b; 以及候选内容检索统计量计算部分5，用于通过使用候选内容检索统计量，不期望内容和具有任意内容的临时期望内容来计算所需内容检索统计量，该任意内容不被选为不期望内容作为暂时期望的内容。版权所有（C）2008，JPO＆INPIT

8. 发明专利

JP2004219506A Method and apparatus for code book creation and communication terminal device 审中-公开
标题翻译：编码和通信终端设备的方法和装置
公开(公告)号：JP2004219506A
公开(公告)日：2004-08-05
申请号：JP2003003947
申请日：2003-01-10
申请人： Toshiba Corp , 株式会社東芝
发明人： ISAKA TAKEHIKO , MISEKI KIMIO
IPC分类号： G10L19/00
摘要： PROBLEM TO BE SOLVED: To provide a method and an apparatus for code book creation which enable a code book to be created without requiring a large storage area nor much calculations.
SOLUTION: A speech for registration is split and feature variable vectors are generated from split speeches for registration; and a representative vector generating device repeats a process (step 21d) of reading in the specified number of feature variable vectors, a process (step 21e) of creating a small code book for them (step 21i), and a process (step 21i) of storing vectors in it into a code book, thus creating the code book.
COPYRIGHT: (C)2004,JPO&NCIPI
摘要翻译：要解决的问题：提供一种能够在不需要大的存储区域和大量计算的情况下创建代码簿的代码簿创建的方法和装置。解决方案：用于注册的语音被分割，并且特征变量向量从用于注册的分割语音产生; 并且代表矢量生成装置重复读取特定数量的特征变量向量的处理（步骤21d），为它们创建小码本的处理（步骤21e）（步骤21i）和处理（步骤21i）将其中的向量存储到码本中，从而创建代码簿。版权所有（C）2004，JPO＆NCIPI

9. 发明专利

JP2002084194A GAIN FORM VECTOR QUALIFYING METHOD AND DEVICE FOR SOUND CODING USING THE SAME 审中-公开
公开(公告)号：JP2002084194A
公开(公告)日：2002-03-22
申请号：JP2001215842
申请日：2001-07-16
申请人： TOSHIBA CORP
发明人： OSHIKIRI MASAHIRO , AKAMINE MASAMI , MISEKI KIMIO , KANBA SUSUMU , AMADA KO
IPC分类号： G10L19/00 , H03M7/30
摘要： PROBLEM TO BE SOLVED: To provide a method for quantizing a gain form vector, where a form vector and a gain vector are searched with less calculation amount, without degrading the coding efficiency. SOLUTION: A first target vector is used to search and decide a first form vector in a first-stage form vector code book (S1). The first form vector and a temporary gain are used to generate a second target vector (S2). The second target vector is used to search and decide a second form vector in a second- stage form vector code book (S3). Based on the first and second form vectors, a gain vector is decided (S4).

10. 发明专利

JP2001117573A METHOD AND DEVICE TO EMPHASIZE VOICE SPECTRUM AND VOICE DECODING DEVICE 失效
公开(公告)号：JP2001117573A
公开(公告)日：2001-04-27
申请号：JP29850599
申请日：1999-10-20
申请人： TOSHIBA CORP
发明人： OSHIKIRI MASAHIRO , MISEKI KIMIO
IPC分类号： G10L13/00 , G10L19/06 , G10L19/12 , H03M7/30
摘要： PROBLEM TO BE SOLVED: To enable ideal voice spectra to be obtained without generating an inappropriate spectrum inclination, without sliding projected part of the spectra nor emphasizing recessed part. SOLUTION: An LPC coefficient computing section 102 computes LPC coefficients which are amplitude spectrum approximate shapes of voice signals. Then, a projected part frequency/recessed part frequency determining section 103 obtains projected and recessed part frequencies of the amplitude spectrum approximate shapes, a projected part band/recessed part band determining section 104 determines a projected part band and a recessed part band, which respectively include the projected and recessed part frequencies, from these frequencies. A filter constituting section 106 has a filter having the characteristics in which the amplitude spectra of the frequency components included in the projected part band are emphasized and the amplitude spectra of the frequency components included in the recessed part band are attenuated. By multiplying the characteristics of the filter by a multiplier 109, voice signals are filtered to conduct spectrum emphasis.

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