Presentation | 2017-03-21 On Data Extension For Signal Data Shu Eguchi, Masaru Tanaka, Jun Fujiki, Takio Kurita, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Efficient learning using limited data in machine learning is also important now even in the era of big data. In the case of image identification, data expansion such as parallel movement, specular inversion, coloring or the like is performed based on the original sample to increase variations and increase the number of data. In this research, we extend data to musical instrument sound data. Specifically, the instrument sound is converted into an image graphically represented by the reproduction time and the sample frequency, and the data is expanded by coloring the image. By coloring, it is possible to obtain as much information as simply being expanded to 3 ch than the original signal data. Then, learning is performed using GoogLeNet implemented by Chainer which is a deep learning tool using this extended data as input data. With this system, learning to recognize a musical instrument from instrumental sounds created with three kinds of instruments such as flute, oboe and fagott, it is possible to expand data similar to image identification even for signal processing confirmed. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | instrument sound / musical chord / signal processing / image recognition / chainer / ata extension |
Paper # | BioX2016-73,PRMU2016-236 |
Date of Issue | 2017-03-13 (BioX, PRMU) |
Conference Information | |
Committee | PRMU / BioX |
---|---|
Conference Date | 2017/3/20(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Eisaku Maeda(NTT) / Masakatsu Nishigaki(Shizuoka Univ.) |
Vice Chair | Seiichi Uchida(Kyushu Univ.) / Hironobu Fujiyoshi(Chubu Univ.) / Akira Otsuka(AIST) / Hiroshi Takano(Toyama Pref. Univ.) |
Secretary | Seiichi Uchida(Kyoto Univ.) / Hironobu Fujiyoshi(NTT) / Akira Otsuka(NEC) / Hiroshi Takano(AIST) |
Assistant | Masaki Oonishi(AIST) / Takuya Funatomi(NAIST) / Masatsugu Ichino(Univ. of Electro-Comm.) / Naoyuki Takada(Secom) / Takahiro Aoki(Fujitsu Labs.) |
Paper Information | |
Registration To | Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Biometrics |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | On Data Extension For Signal Data |
Sub Title (in English) | |
Keyword(1) | instrument sound |
Keyword(2) | musical chord |
Keyword(3) | signal processing |
Keyword(4) | image recognition |
Keyword(5) | chainer |
Keyword(6) | ata extension |
1st Author's Name | Shu Eguchi |
1st Author's Affiliation | Fukuoka University(Fukuoka Univ.) |
2nd Author's Name | Masaru Tanaka |
2nd Author's Affiliation | Fukuoka University(Fukuoka Univ.) |
3rd Author's Name | Jun Fujiki |
3rd Author's Affiliation | Fukuoka University(Fukuoka Univ.) |
4th Author's Name | Takio Kurita |
4th Author's Affiliation | Hiroshima University(Hiroshima Univ.) |
Date | 2017-03-21 |
Paper # | BioX2016-73,PRMU2016-236 |
Volume (vol) | vol.116 |
Number (no) | BioX-527,PRMU-528 |
Page | pp.pp.237-242(BioX), pp.237-242(PRMU), |
#Pages | 6 |
Date of Issue | 2017-03-13 (BioX, PRMU) |