公开(公告)号：JP2008275881A

公开(公告)日：2008-11-13

申请号：JP2007119194

申请日：2007-04-27

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： FUKUDA TAKASHI ,  ICHIKAWA OSAMU ,  NISHIMURA MASAFUMI

IPC: G10L15/20 ,  G10L15/28 ,  H04R3/00

CPC classification number: G10L15/20 ,  G10L15/02 ,  G10L21/02 ,  G10L21/0272 ,  G10L2021/02161

Abstract: PROBLEM TO BE SOLVED: To extract only voice of a target person under noise environment, without requiring a large scale microphone array and a reference signal of noise.
SOLUTION: An object sound extraction method is disclosed in which a practical speech recognition performance is actualized only by performing gain adjustment between spectrum subtraction (SS) processing and flooring processing, as processing for two channel input speech which is obtained from the microphones 1 and 2 etc. As the gain adjustment, a CSP (Cross-power Spectrum Phase) coefficient, which is cross-correlation between two channel signals, can be utilized. In an indoor environment including a vehicle where audio background sound etc., a recognition rate of a voice command in a car navigation system is improved, then, usability of a speaker such as a driver is improved.
COPYRIGHT: (C)2009,JPO&INPIT

Abstract translation: 要解决的问题：在噪声环境下仅提取目标人员的声音，而不需要大规模的麦克风阵列和噪声的参考信号。 解决方案：公开了一种对象声音提取方法，其中仅通过在频谱减法（SS）处理和地板处理之间进行增益调整来实现实际语音识别性能，作为从麦克风获得的两声道输入语音的处理 1和2等。作为增益调整，可以使用在两个信道信号之间互相关的CSP（跨功率谱相位）系数。 在包括音响背景音等的车辆的室内环境中，提高了汽车导航系统中的语音命令的识别率，因此提高了诸如驾驶员的扬声器的可用性。 版权所有（C）2009，JPO＆INPIT

公开(公告)号：JP2008058900A

公开(公告)日：2008-03-13

申请号：JP2006238873

申请日：2006-09-04

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ICHIKAWA OSAMU ,  FUKUDA TAKASHI ,  NISHIMURA MASAFUMI

IPC: G10L15/20 ,  G10L21/02

CPC classification number: G10L2021/02082

Abstract: PROBLEM TO BE SOLVED: To solve the problem wherein although the performance of a voice recognition device deteriorates significantly in the circumstances in which there exists long reverberation, which is generally known, and most of the conventional reverberation removal methods require a large amount of calculation is not large, or for those where the amount of calculation is not large, some kind of previous knowledge (reverberation time of a room, etc.) is required.
SOLUTION: The coefficient determination in the conventional techniques, in which the multiple value of the coefficient of power spectrum of the past frame is subtracted from the power spectrum of the current frame is calculated at low cost, without having to use the information that incurs calculation cost, such as acoustic model or multi-channel input. As a specific method, a voice power track that properly follows the frame of large power and follows the frame of small power late is obtained, and the interval of which the voice power differs significantly from the voice power of the current frame that is smoothed in the time direction is deduced as being an utterance terminal reverberation interval, and the filter coefficient is decided, in such a manner as to minimize the weighted total sum of the residual voice power in the interval and the subtracted power in the utterance interval (not including the reverberation interval).
COPYRIGHT: (C)2008,JPO&INPIT

Abstract translation: 要解决的问题为了解决这样一个问题，即在通常已知的存在长混响的情况下语音识别装置的性能显着恶化，并且大多数传统的混响消除方法需要大量的 计算量不大，或对于计算量不大的情况，需要某种以前的知识（房间的混响时间等）。 解决方案：以低成本计算过去帧的功率谱系数的多个值从当前帧的功率谱中减去的常规技术中的系数确定，而不必使用该信息 导致计算成本，如声学模型或多通道输入。 作为具体的方法，获得了正确跟随大功率帧并且跟随小功率帧的语音功率轨迹，并且语音功率的间隔与当前平滑化的帧的语音功率显着不同 将时间方向推定为发声终端混响间隔，并且以使得间隔中的剩余语音功率的加权总和和话音间隔中的减法功率（不包括）的方式来决定滤波器系数 混响间隔）。 版权所有（C）2008，JPO＆INPIT

公开(公告)号：JP2005338402A

公开(公告)日：2005-12-08

申请号：JP2004156571

申请日：2004-05-26

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： NISHIMURA MASAFUMI ,  TAKIGUCHI TETSUYA ,  ICHIKAWA OSAMU

IPC: H04R5/027 ,  G10L15/20 ,  G10L17/00 ,  G10L21/02 ,  H04R3/00 ,  H04R5/00

CPC classification number: G10L21/028

Abstract: PROBLEM TO BE SOLVED: To provide a method of specifying speakers of individual voices from recorded voices of a plurality of speakers with simple device constitution, and a system using the same method. SOLUTION: The system is equipped with microphones 10 which are provided by the speakers, a speech processing section 20 which imparts unique characteristics to speech signals of two channels recorded by the microphones 10 through mutually different speech processes and mixes the signals by the channels, and an analysis section 40 which takes analyses corresponding to the unique characteristics imparted to the speech signals by the microphones 10 through the processes of the speech process section 20 to specify speakers by utterance sections of the speech signals. COPYRIGHT: (C)2006,JPO&NCIPI

Abstract translation: 要解决的问题：提供一种用简单的装置构成从多个扬声器的记录的声音指定各个声音的扬声器的方法，以及使用相同方法的系统。 解决方案：该系统配备有由扬声器提供的麦克风10，语音处理部分20，其通过相互不同的语音处理向麦克风10记录的两个声道的语音信号赋予独特的特性，并将信号混合在一起 频道，以及分析部分40，通过语音处理部分20的处理，通过麦克风10对语音信号赋予的独特特征进行分析，以通过语音信号的话语部分指定扬声器。 版权所有（C）2006，JPO＆NCIPI

公开(公告)号：JP2004279241A

公开(公告)日：2004-10-07

申请号：JP2003071735

申请日：2003-03-17

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ICHIKAWA OSAMU ,  NISHIMURA MASAFUMI

IPC: H04R1/34 ,  G01S3/80 ,  G01S3/801 ,  G01S3/802 ,  G01S5/18 ,  G10K11/26

CPC classification number: G01S3/801 ,  G01S3/802

Abstract: PROBLEM TO BE SOLVED: To make it possible to estimate the sound source position that has been difficult for the conventional system using a small number of microphones, so as to improve the conventional estimating accuracy of the sound source position.
SOLUTION: It is made possible to estimate the sound source position by forming a reflection surface RS as a enveloping surface of spheroid using the location of a collection means and the sound source location as the focal points, by generating main reflected waves with the amount of delay corresponding to the sound source location, and by inspecting the amount of delay between the direct wave and the reflected wave so as to acquire the sound source location and to make estimable the location.
COPYRIGHT: (C)2005,JPO&NCIPI

公开(公告)号：JP2011113044A

公开(公告)日：2011-06-09

申请号：JP2009271890

申请日：2009-11-30

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ICHIKAWA OSAMU ,  FUKUDA TAKASHI ,  NISHIMURA MASAFUMI

IPC: G10L15/28 ,  G10L15/20 ,  G10L21/02

CPC classification number: G10L25/78 ,  G10L15/20 ,  G10L21/028 ,  G10L2021/02166

Abstract: PROBLEM TO BE SOLVED: To provide technology for extracting objective voice by efficiently suppressing mixing of other voice than objective voice, in a plurality pieces of voice which come from different directions.
SOLUTION: The objective voice is extracted by performing at least either gain adjustment processing and segmentation processing of an utterance section, on a voice signal obtained by each of first and second voice input units which are arranged with a predetermined distance apart, by using a weighted Cross-Power Spectrum Phase (CSP) coefficient which becomes a small value in a frequency band which is likely to be influenced by other voice than the objective voice.
COPYRIGHT: (C)2011,JPO&INPIT

Abstract translation: 要解决的问题：通过有效地抑制来自不同方向的多个声音中的其他声音的混合而不是客观语音来提取目标声音的技术。 解决方案：通过对由发声部分进行的增益调整处理和分段处理，对由第一和第二语音输入单元中的每一个以预定距离间隔排列而获得的语音信号进行至少一个提取，目标声音由 使用加权的跨功率谱相位（CSP）系数，该系数在可能受到客观声音的其他声音影响的频带中变成小值。 版权所有（C）2011，JPO＆INPIT

公开(公告)号：JP2002152452A

公开(公告)日：2002-05-24

申请号：JP2000306542

申请日：2000-10-05

Applicant: IBM

Inventor： ICHIKAWA OSAMU

IPC: G06F13/00 ,  H04M11/00 ,  H04N1/00 ,  H04N1/32 ,  H04N1/34

Abstract: PROBLEM TO BE SOLVED: To allow a user who reads a Web page by Internet connection equipment to which any printer is not connected, for example, a portable telephone 19 to smoothly obtain the hard copy of the Web page. SOLUTION: An icon 52 for requesting facsimile transmission is displayed in a Web page. When a user clicks the icon 52 for requesting the facsimile transmission, a facsimile server 12 is informed of the URL of the Web page, and the picture of a display part 51 is switched to the Web page of the facsimile server 12. Then, a user inputs a membership number for charging and the facsimile number of the destination of facsimile transmission on this switched picture. The facsimile server 12 performs access to the communicated URL, and generates data for facsimile output from the Web page, and transmits the data to facsimile equipment 36 at the destination of facsimile transmission.

公开(公告)号：JP2010021854A

公开(公告)日：2010-01-28

申请号：JP2008181514

申请日：2008-07-11

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ICHIKAWA OSAMU ,  NISHIMURA MASAFUMI ,  NAGANO TORU

IPC: H04R3/00 ,  A63F13/00 ,  G01S5/30 ,  G10L11/00 ,  H04R1/40

CPC classification number: G01S5/30 ,  H04R3/005

Abstract: PROBLEM TO BE SOLVED: To provide a method and system for detecting a position of a user of a home television game machine. SOLUTION: A speaker 506 mounted in a remote controller is used to reproduce a signal of a predetermined reproduced sound, the reproduced sound is observed respectively by two microphones properly provided in the vicinity of a television screen, CSP (while mutual correlation) coefficients of a signal of an observation sound respectively observed and the signal of the reproduced sound are calculated, and distances between the speaker inside the remote controller and the microphones are calculated, thereby acquiring longitudinal and lateral absolute positions of the remote controller with respect to a microphone array. An interference sound of an environmental sound or noise is canceled by the correlation calculation. COPYRIGHT: (C)2010,JPO&INPIT

Abstract translation: 要解决的问题：提供一种用于检测家庭电视游戏机的用户的位置的方法和系统。 解决方案：安装在遥控器中的扬声器506用于再现预定再现声音的信号，再现的声音分别由在电视机屏幕附近适当提供的两个麦克风CSS（相互相关） 分别观察观测声音的信号的系数和再生声音的信号，并且计算遥控器内的扬声器与麦克风之间的距离，从而获得遥控器相对于扬声器的纵向和横向的绝对位置 麦克风阵列 通过相关计算来消除环境声音或噪声的干扰声。 版权所有（C）2010，JPO＆INPIT

公开(公告)号：JP2009058708A

公开(公告)日：2009-03-19

申请号：JP2007225195

申请日：2007-08-31

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： ICHIKAWA OSAMU ,  FUKUDA TAKASHI ,  NISHIMURA MASAFUMI

IPC: G10L15/02 ,  G10L11/00

CPC classification number: G10L15/20 ,  G10L15/02 ,  G10L25/24

Abstract: PROBLEM TO BE SOLVED: To provide a voice processing technique attaining stable voice recognition even in noise. SOLUTION: A high-order term and a low-order term of cepstrum of an observation voice are cut to design a filter directly from the observation voice itself. The filter is thereby made a filter with weight at a harmonic structure part in a section of a voiced sound, and a filter close to flat in a section of voiceless sound without the harmonic structure. Since this change is continuous, stable processing can be performed without distinguishing the voiced sound section from the voiceless sound section. COPYRIGHT: (C)2009,JPO&INPIT

Abstract translation: 要解决的问题：提供甚至在噪声中实现稳定语音识别的语音处理技术。 解决方案：切割观察语音的倒谱的高阶项和低阶项，直接从观察声音本身设计滤波器。 由此，过滤器在声音的一部分中的谐波结构部分处具有重量的滤波器，并且在没有谐波结构的无声声音部分中接近平坦的滤波器。 由于该变化是连续的，因此可以进行稳定的处理，而不区分浊音部分与无声音部分。 版权所有（C）2009，JPO＆INPIT

公开(公告)号：JP2013186258A

公开(公告)日：2013-09-19

申请号：JP2012050603

申请日：2012-03-07

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Machines Corporation

Inventor： ICHIKAWA OSAMU

IPC: G10L15/20 ,  G10L15/02 ,  G10L15/14 ,  G10L21/0232

CPC classification number: G10L21/0208 ,  G10L15/20

Abstract: PROBLEM TO BE SOLVED: To provide a novel method for noise reduction applied to a speech recognition front-end.SOLUTION: An output of a front-end 3000 is optimized by giving, as a weight to the output for each band, a confidence index representing the remarkableness of the harmonic structure of observation speech. In a first method, when clean speech is estimated by executing MMSE estimation on a model that gives a probability distribution of noise-removed speech generated from observation speech, the posterior probability of the MMSE estimation is weighted using the confidence index as a weight. In a second method, linear interpolation is executed, for each band, between an observed value of observation speech and an estimated value of clean speech, with the confidence index serving as a weight. The first method and the second method can be combined.

Abstract translation: 要解决的问题：提供一种应用于语音识别前端的降噪新方法。解决方案：前端3000的输出通过给出每个频带的输出的权重来优化置信指数 代表了观察语音谐波结构的重要性。 在第一种方法中，当通过对给出从观察语音产生的噪声去除语音的概率分布的模型执行MMSE估计来估计清洁语音时，使用置信度指数作为权重来加权MMSE估计的后验概率。 在第二种方法中，对于每个频带，在观测语音的观测值和干净语音的估计值之间执行线性内插，其中置信指数用作权重。 可以组合第一种方法和第二种方法。

公开(公告)号：JP2009210617A

公开(公告)日：2009-09-17

申请号：JP2008050537

申请日：2008-02-29

Applicant: Internatl Business Mach Corp ,  インターナショナル・ビジネス・マシーンズ・コーポレーションInternational Business Maschines Corporation

Inventor： FUKUDA TAKASHI ,  ICHIKAWA OSAMU ,  NISHIMURA MASAFUMI

IPC: G10L11/02 ,  G10L11/00 ,  G10L15/04

CPC classification number: G10L25/93

Abstract: PROBLEM TO BE SOLVED: To provide a highly accurate voice activity detection method in a low S/N environment.
SOLUTION: The voice activity is performed by extracting a long-term spectrum variation component and a harmonic structure as feature vectors from a speech signal and increasing difference in feature vectors between speech and non-speech included in the speech signal by using the long-term spectrum variation component feature, or a long-term spectrum variation component extraction and a harmonic structure feature extraction. A correct rate and an accuracy rate of the voice activity detection is improved over conventional methods by using a long-term spectrum variation component having a window length over an average phoneme duration of an utterance in the speech signal. The voice activity detection system and method provides speech processing, automatic speech recognition, and speech output capable of very accurate voice activity detection.
COPYRIGHT: (C)2009,JPO&INPIT

Abstract translation: 要解决的问题：在低S / N环境中提供高精度的语音活动检测方法。 解决方案：通过从语音信号中提取长期频谱变化分量和谐波结构作为特征向量并且通过使用语音信号增加语音信号中包括的语音和非语音之间的特征向量的差异来执行语音活动 长期光谱变化分量特征，或长期光谱变化分量提取和谐波结构特征提取。 通过使用具有在语音信号中的话语的平均音素持续时间上的窗口长度的长期频谱变化分量，语音活动检测的正确率和准确率比常规方法得到改进。 语音活动检测系统和方法提供能够进行非常精确的语音活动检测的语音处理，自动语音识别和语音输出。 版权所有（C）2009，JPO＆INPIT