公开(公告)号：HK1064159A1

公开(公告)日：2005-01-21

申请号：HK04105975

申请日：2004-08-10

Applicant: YAMAHA CORP

Inventor： FUJISHIMA TAKUYA ,  YAMAURA ATSUSHI ,  NOGUCHI AKANE ,  TAKAHASHI TAKUYA

IPC: G06F12/00 ,  G06F13/00 ,  G06F13/14 ,  G06F17/00 ,  G06F17/30 ,  H04N7/173 ,  H04N21/2381 ,  H04N21/258 ,  H04N21/8547 ,  H04N21/8549 ,  G06F

公开(公告)号：DE602008003987D1

公开(公告)日：2011-02-03

申请号：DE602008003987

申请日：2008-02-26

Applicant: NAT INST OF ADVANCED IND SCIEN ,  YAMAHA CORP

Inventor： GOTO MASATAKA ,  FUJISHIMA TAKUYA ,  ARIMOTO KEITA

IPC: G10H1/00 ,  G10H1/36

公开(公告)号：AT515021T

公开(公告)日：2011-07-15

申请号：AT05800146

申请日：2005-10-27

Applicant: YAMAHA CORP

Inventor： FUJISHIMA TAKUYA ,  BONADA JORDI

IPC: G10L21/04 ,  G10H1/00 ,  G10L25/90

Abstract: A pitch shifting apparatus detects peak spectra P1 and P2 from amplitude spectra of inputs sound. The pitch shifting apparatus compresses or expands an amplitude spectrum distribution AM1 in a first frequency region A1 including a first frequency f1 of the peak spectrum P1 using a pitch shift ratio which keeps its shape to obtain an amplitude spectrum distribution AM10 for a pitch-shifted first frequency region A10. The pitch shifting apparatus similarly compresses or expands an amplitude spectrum distribution AM2 adjacent to the peak spectrum P2 to obtain an amplitude spectrum distribution AM20. The pitch shifting apparatus performs pitch shifting by compressing or expanding amplitude spectra in an intermediate frequency region A3 between the peak spectra P1 and P2 at a given pitch shift ratio in response to the each amplitude spectrum.

公开(公告)号：AT492873T

公开(公告)日：2011-01-15

申请号：AT08101972

申请日：2008-02-26

Applicant: NAT INST OF ADVANCED IND SCIEN ,  YAMAHA CORP

Inventor： GOTO MASATAKA ,  FUJISHIMA TAKUYA ,  ARIMOTO KEITA

IPC: G10H1/00 ,  G10H1/36

Abstract: A sound analysis apparatus employs tone models which are associated with various fundamental frequencies and each of which simulates a harmonic structure of a performance sound generated by a musical instrument, then defines a weighted mixture of the tone models to simulate frequency components of the performance sound, further sequentially updates and optimizes weight values of the respective tone models so that a frequency distribution of the weighted mixture of the tone models corresponds to a distribution of the frequency components of the performance sound, and estimates the fundamental frequency of the performance sound based on the optimized weight values.

公开(公告)号：EP1806740A4

公开(公告)日：2009-12-16

申请号：EP05800146

申请日：2005-10-27

Applicant: YAMAHA CORP

Inventor： FUJISHIMA TAKUYA ,  BONADA JORDI

IPC: G10L21/04 ,  G10H1/00

CPC classification number: G10H7/002 ,  G10H2210/331 ,  G10H2250/235 ,  G10H2250/621 ,  G10L21/003 ,  G10L21/013 ,  G10L21/04

公开(公告)号：JP2013047938A

公开(公告)日：2013-03-07

申请号：JP2012158412

申请日：2012-07-17

Applicant: Yamaha Corp ,  ヤマハ株式会社

Inventor： SUMI KOHEI ,  FUJISHIMA TAKUYA ,  KIKUCHI TAKESHI

IPC: G06F17/30 ,  G10G1/00

CPC classification number: G10H7/00 ,  G06F17/30743 ,  G10H1/0008 ,  G10H2210/066 ,  G10H2240/141 ,  G10H2240/311 ,  G10H2250/015

Abstract: PROBLEM TO BE SOLVED: To realize music analysis (especially, music search) robust against an error in a designated note sequence.SOLUTION: A feature extraction unit 22 generates a time series (feature vector series) X of feature vectors xm from a designated note sequence corresponding to indication from a user. A probabilistic model (a weight λk for a feature function fk()) generated by machine learning using time series of feature vectors xm of a plurality of reference music pieces is applied to the feature vector series X of the designated note sequence, and thereby an evaluation index value SC[n] depending on the probability P[ym=Ln] that the designated note sequence is a note sequence in a reference music piece is computed for each reference music piece.

Abstract translation: 要解决的问题：实现针对指定笔记序列中的错误的音乐分析（特别是音乐搜索）。 解决方案：特征提取单元22从对应于来自用户的指示的指定音符序列生成特征向量xm的时间序列（特征向量序列）X。 通过使用多个参考音乐片段的特征向量xm的时间序列的机器学习产生的概率模型（对于特征函数fk（）的权重λk）被应用于指定音符序列的特征向量序列X， 针对每个参考音乐片段计算指定音符序列是参考乐曲中的音符序列的概率P [ym = Ln]的评估指标值SC [n]。 版权所有（C）2013，JPO＆INPIT

公开(公告)号：JP2012234168A

公开(公告)日：2012-11-29

申请号：JP2012094414

申请日：2012-04-18

Applicant: Yamaha Corp ,  ヤマハ株式会社

Inventor： WATANABE DAICHI ,  FUJISHIMA TAKUYA

IPC: G10H1/00 ,  G10G1/00

CPC classification number: G06Q30/02 ,  G06F17/30743 ,  G06F17/30758 ,  G06F17/30864 ,  G10H1/383 ,  G10H1/386 ,  G10H1/42 ,  G10H2210/361 ,  G10H2240/141

Abstract: PROBLEM TO BE SOLVED: To search for musical data related to an automatic accompaniment including a phrase configured of a musical tone pattern of which a degree of similarity to the pattern of an intended musical tone on a part specified by a user satisfies predetermined conditions.SOLUTION: A user inputs, in a rhythm input device 10, a rhythm pattern using an operation form of an operator corresponding to any part of a plurality of parts constituting automatic accompaniment data. An input rhythm pattern storage section 212 causes RAM to store an input rhythm pattern on the basis of a clock signal output by a bar line clock output section 211 and an input trigger data. A part specification section 213 specifies an object part from MIDI information according to the operator. A rhythm pattern search section 214 searches an automatic accompaniment DB 221 for automatic accompaniment data including a rhythm pattern that has a highest degree of similarity to the input rhythm pattern, on the specified part.

Abstract translation: 要解决的问题：搜索与自动伴奏有关的音乐数据，其中包括由用户指定的部分上的预期音调的图案的相似度满足预定的乐音模式构成的乐句 条件。

解决方案：用户在节奏输入装置10中使用与构成自动伴奏数据的多个部分的任何部分相对应的操作者的操作形式来输入节奏模式。 输入节奏模式存储部212基于由条形线时钟输出部211输出的时钟信号和输入触发数据，使RAM存储输入节奏模式。 部分规格部分213根据操作者从MIDI信息指定对象部分。 节奏模式搜索部分214在指定的部分上搜索自动伴奏DB 221用于包括与输入节奏模式具有最高相似程度的节奏模式的自动伴奏数据。 版权所有（C）2013，JPO＆INPIT

公开(公告)号：JP2010113278A

公开(公告)日：2010-05-20

申请号：JP2008287608

申请日：2008-11-10

Applicant: Yamaha Corp ,  ヤマハ株式会社

Inventor： FUJISHIMA TAKUYA ,  DE BOER MAARTEN ,  DE JONG FOKKE

IPC: G10H7/02 ,  G10L19/00 ,  G11B27/034

Abstract: PROBLEM TO BE SOLVED: To prevent unnecessary sound from being generated between fragments and make the seam inconspicuous when connecting a plurality of musical pieces in units of fragments to generate a new musical piece. SOLUTION: The music processing device includes a storage means for storing music sound data showing the waveform of each fragment obtained by dividing a musical piece. The device is configured to: sequentially read out the relevant musical sound data from the storage means according to reproduction instruction showing a plurality of fragments to be reproduced and the reproduction timing of each fragment; apply an envelope for fading out in a fragment represented by the relevant musical sound data, to the musical sound data; on the other hand, apply an envelope for fading in before starting the fragment represented by the musical sound data when the musical sound data is output following another musical sound data; and outputting the musical sound data with the envelopes applied thereto with cross-fading. COPYRIGHT: (C)2010,JPO&INPIT

Abstract translation: 要解决的问题：为了防止在碎片之间产生不必要的声音，并且当以片段为单位连接多个音乐作品时使得接缝不显眼，以产生新的音乐作品。 解决方案：音乐处理装置包括存储装置，用于存储表示通过划分乐曲而获得的每个片段的波形的音乐声音数据。 该装置被配置为：根据表示要再现的多个片段的再现指示和每个片段的再现定时，从存储装置依次读出相关音乐数据; 将由相关音乐数据表示的片段中的信封应用于音乐声音数据; 另一方面，当音乐声音数据按照另一音乐声音数据输出时，在开始由音乐声音数据表示的片段之前应用包围衰落; 并且以交叉衰落的方式输出应用于其上的信封的音乐声音数据。 版权所有（C）2010，JPO＆INPIT

公开(公告)号：JP2010008448A

公开(公告)日：2010-01-14

申请号：JP2008164057

申请日：2008-06-24

Applicant: Yamaha Corp ,  ヤマハ株式会社

Inventor： STREICH SEBASTIAN ,  FUJISHIMA TAKUYA

IPC: G10G3/04 ,  G10G1/00 ,  G10L11/00

CPC classification number: G10H1/0008 ,  G10H1/383 ,  G10H2210/066 ,  G10H2210/601 ,  G10H2250/031

Abstract: PROBLEM TO BE SOLVED: To precisely evaluate a degree of consonance or dissonance between a plurality of sounds. SOLUTION: A mask generation section 30 generates an evaluating mask M indicating, per each frequency, a degree of dissonance between a target sound VA and a sound at the frequency by setting a dissonance function Fd indicating relationship between a frequency difference from a peak p and a degree of dissonance with a component of the peak, for each of a plurality of peaks p in a spectral sequence RA of the target sound VA. An index calculation section 60 collates a spectral sequence RB of an evaluated sound VB with the evaluating mask M to calculate a consonance index value D indicating a degree of consonance or dissonance between the target sound VA and the evaluated sound VB. COPYRIGHT: (C)2010,JPO&INPIT

Abstract translation: 要解决的问题：精确评估多个声音之间的一致程度或不一致。 解决方案：掩模生成部分30通过设置指示与频率差异的频率差异之间的关系的不一致函数Fd来产生每个频率指示目标声音VA和频率之间的声音之间的不平衡度的评估屏蔽M 对于目标声音VA的频谱序列RA中的多个峰值p中的每一个，峰值p和与峰值的分量的不一致程度。 索引计算部分60将评估声音VB的频谱序列RB与评估掩码M进行比较，以计算表示目标声音VA和评估声音VB之间的协调程度或不和音的协调指数值D. 版权所有（C）2010，JPO＆INPIT

公开(公告)号：JP2008216413A

公开(公告)日：2008-09-18

申请号：JP2007051092

申请日：2007-03-01

Applicant: Yamaha Corp ,  ヤマハ株式会社

Inventor： FUJISHIMA TAKUYA ,  STREICH SEBASTIAN ,  ONG BEE SUAN ,  SUGII KIYOHISA ,  KOJIMA NAOAKI ,  DE BOER MAARTEN

IPC: G10H1/00

Abstract: PROBLEM TO BE SOLVED: To enable a user to visually grasp a range of sounds matching his or her image and play using sounds within the range. SOLUTION: The player displays on a display device an image representing as points in a two-dimensional space respective sound data stored in a database storing feature quantity data indicative of degrees of two kinds of features of sounds that a plurality of sound data represent corresponding to the sound data. When an operation to specify one point in the two-dimensional space is detected, the player reads out sound data corresponding to a point whose distance from the specified point is less than a predetermined threshold, and performs output control over a sound signal according to the sound data. COPYRIGHT: (C)2008,JPO&INPIT

Abstract translation: 要解决的问题：为了使用户能够目视地掌握与他或她的图像相匹配的声音范围，并且使用在该范围内的声音进行播放。 解决方案：播放器在显示设备上显示表示二维空间中的点的图像，存储在数据库中的各个声音数据存储表示多个声音数据的两种声音特征的度数的特征量数据 表示对应于声音数据。 当检测到指定二维空间中的一个点的操作时，播放器读出与从指定点的距离小于预定阈值的点相对应的声音数据，并且根据声音信号执行对声音信号的输出控制 声音数据。 版权所有（C）2008，JPO＆INPIT