Masuzo Yanagida
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Masuzo Yanagida has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Diego Humberto Milone
Research Institute for Signals, Systems and Computational Intelligence, sinc(i), FICH-UNL, CONICET, Ciudad Universitaria UNL, Santa Fe, Argentina
3 shared publications - Leandro Ezequiel Di Persia
Lab. Señales y Dinánicas no Lineales IBB -UNER FICH -UNL -CONICET Santa Fe ( 3000) Santa Fe Argentina
3 shared publications - Masanobu Miura
Dept. of Media Informatics, Ryukoku Univ., 1-5, Seta, Oe-cho, Yokotani, Otsu 5202194
2 shared publications Norio Emura
Faculty of Engineering, Doshisha University, 1-3, Tataramiyakodani, 610-0321 Kyo-tanabe, Kyoto, Japan
2 shared publications- Hugo Leonardo Rufiner
Instituto de Investigación en Señales, Sistemas e Inteligencia Computacional, sinc(i), FICH-UNL/CONICET, Santa Fe, Argentina
1 shared publication Ichiro Nakayama
1 shared publication
Latest Publications
- 22 December 2009
- Published by Springer Nature in Journal of Signal Processing Systems
- Vol. 63 (3) , 333-344
Correlated Postfiltering and Mutual Information in Pseudoanechoic Model Based Blind Source Separation
- 22 December 2009
- Published by Springer Nature in Journal of Signal Processing Systems
- Vol. 63 (3) , 333-344
Correlated Postfiltering and Mutual Information in Pseudoanechoic Model Based Blind Source Separation
- 13 January 2009
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Audio, Speech, and Language Processing
- Vol. 17 (2) , 299-311
Indeterminacy Free Frequency-Domain Blind Separation of Reverberant Audio Sources
Blind separation of convolutive mixtures is a very complicated task that has applications in many fields of speech and audio processing, such as hearing aids and man-machine interfaces. One of the proposed solutions is the frequency-domain independent component analysis. The main disadvantage of this method is the presence of permutation ambiguities among consecutive frequency bins. Moreover, this problem is worst when reverberation time increases. Presented in this paper is a new frequency-domain method, that uses a simplified mixing model, where the impulse responses from one source to each microphone are expressed as scaled and delayed versions of one of these impulse responses. This assumption, based on the similitude among waveforms of the impulse responses, is valid for a small spacing of the microphones. Under this model, separation is performed without any permutation or amplitude ambiguity among consecutive frequency bins. This new method is aimed mainly to obtain separation, with a small reduction of reverberation. Nevertheless, as the reverberation is included in the model, the new method is capable of performing separation for a wide range of reverberant conditions, with very high speed. The separation quality is evaluated using a perceptually designed objective measure. Also, an automatic speech recognition system is used to test the advantages of the algorithm in a real application. Very good results are obtained for both, artificial and real mixtures. The results are significantly better than those by other standard blind source separation algorithms.
- 13 January 2009
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Audio, Speech, and Language Processing
- Vol. 17 (2) , 299-311
Indeterminacy Free Frequency-Domain Blind Separation of Reverberant Audio Sources
Blind separation of convolutive mixtures is a very complicated task that has applications in many fields of speech and audio processing, such as hearing aids and man-machine interfaces. One of the proposed solutions is the frequency-domain independent component analysis. The main disadvantage of this method is the presence of permutation ambiguities among consecutive frequency bins. Moreover, this problem is worst when reverberation time increases. Presented in this paper is a new frequency-domain method, that uses a simplified mixing model, where the impulse responses from one source to each microphone are expressed as scaled and delayed versions of one of these impulse responses. This assumption, based on the similitude among waveforms of the impulse responses, is valid for a small spacing of the microphones. Under this model, separation is performed without any permutation or amplitude ambiguity among consecutive frequency bins. This new method is aimed mainly to obtain separation, with a small reduction of reverberation. Nevertheless, as the reverberation is included in the model, the new method is capable of performing separation for a wide range of reverberant conditions, with very high speed. The separation quality is evaluated using a perceptually designed objective measure. Also, an automatic speech recognition system is used to test the advantages of the algorithm in a real application. Very good results are obtained for both, artificial and real mixtures. The results are significantly better than those by other standard blind source separation algorithms.
- 1 May 2008
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 123 (5_Suppleme) , 3660
Feature exaggeration in scale performance on the piano
This paper proposes a set of parameters for describing features of scale‐playing on the piano. The parameter set consists of 15 parameters, among which 12 are three sets of four parameters pi1 through pi4 where i ={1, 2, 3} distinguishes three basic features; onset time, velocity, and duration. Each of these basic features is modeled as the sum of a global curve and the deviation from it, where the spline interpolation is employed using locally averaged points, or representative points, as the points to be passed. The local average is calculated for each sequence of notes played without finger crossing. The suffix j in pij distinguishes the standard deviations (j=0), the rms deviation from the spline curve (j=1), the range of the curve (j=2), the rms difference between successive notes (j=3), and the rms of the spline curve from the metronomic line (j=4). All parameters are made controllable with slider bars from 0% to 200% for synthesizing suppressed performance or exaggerated performance, where 100% represents the original performance. Proposed parameter set is expected to be useful in self‐training of piano, as it can indicate the features and undesirable habits of the player by setting values above 100% in exaggerated form.
- 1 May 2008
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 123 (5_Suppleme) , 3660
Feature exaggeration in scale performance on the piano
This paper proposes a set of parameters for describing features of scale‐playing on the piano. The parameter set consists of 15 parameters, among which 12 are three sets of four parameters pi1 through pi4 where i ={1, 2, 3} distinguishes three basic features; onset time, velocity, and duration. Each of these basic features is modeled as the sum of a global curve and the deviation from it, where the spline interpolation is employed using locally averaged points, or representative points, as the points to be passed. The local average is calculated for each sequence of notes played without finger crossing. The suffix j in pij distinguishes the standard deviations (j=0), the rms deviation from the spline curve (j=1), the range of the curve (j=2), the rms difference between successive notes (j=3), and the rms of the spline curve from the metronomic line (j=4). All parameters are made controllable with slider bars from 0% to 200% for synthesizing suppressed performance or exaggerated performance, where 100% represents the original performance. Proposed parameter set is expected to be useful in self‐training of piano, as it can indicate the features and undesirable habits of the player by setting values above 100% in exaggerated form.
- 1 January 2008
- Published by Acoustical Society of Japan in Acoustical Science and Technology
- Vol. 29 (1) , 51-57
A modular system generating Jazz-style arrangement for a given set of a melody and its chord name sequence
- 1 January 2008
- Published by Acoustical Society of Japan in Acoustical Science and Technology
- Vol. 29 (1) , 51-57
A modular system generating Jazz-style arrangement for a given set of a melody and its chord name sequence
- 1 March 2007
- Published by Elsevier in Signal Processing
- Vol. 87 (8) , 1951-1965
Objective quality evaluation in blind source separation for speech recognition in a real room
- 1 March 2007
- Published by Elsevier in Signal Processing
- Vol. 87 (8) , 1951-1965
Objective quality evaluation in blind source separation for speech recognition in a real room
- Open Access
- 23 March 2004
- Published by International Speech Communication Association
- p. 151-154
Variability of vibrato - a comparative study between Japanese traditional singing and bel canto -
Several styles of vibrato are investigated, comparing those of Japanese traditional singing with western bel canto from viewpoints of stationarity, depth, rate, build-up time, and synchronization between ¼ and power. Analyzed as Japanese traditional singing are Noh, Kyogen, Heikyoku(Biwa), Shomyo, Kabuki and Nagauta. Singers of Japanese traditional singing include living national treasures.
- 23 March 2004
- Published by International Speech Communication Association
- p. 151-154
Open AccessVariability of vibrato - a comparative study between Japanese traditional singing and bel canto -
Several styles of vibrato are investigated, comparing those of Japanese traditional singing with western bel canto from viewpoints of stationarity, depth, rate, build-up time, and synchronization between ¼ and power. Analyzed as Japanese traditional singing are Noh, Kyogen, Heikyoku(Biwa), Shomyo, Kabuki and Nagauta. Singers of Japanese traditional singing include living national treasures.