| Project/Area Number |
25330436
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Entertainment and game informatics 1
|
|---|
| Research Institution | Hiroshima City University |
|---|
Principal Investigator |
Hayashi Akira 広島市立大学, 情報科学研究科, 名誉教授 (60240909)
|
|---|
| Project Period (FY) |
2013-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
Fiscal Year 2015: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2014: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2013: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
|
|---|
| Keywords | 音楽情報検索 / 楽曲セグメンテーション / コスト関数 / ２パス・アルゴリズム / ダイナミック・プログラミング / 音楽情報処理 / 楽曲の階層性 / 楽曲のセグメンテーション / ダイナミックプログラミング / ２パスアルゴリズム / 楽曲のせグメンテーション / 多声音楽 / マルコフ確率場 / 階層隠れマルコフモデル / ビタービアルゴリズム / 作曲家推定 |
|---|
| Outline of Final Research Achievements |
There are three major approaches to music segmentation: novelty-based, homogeneity-based, and repetition-based approaches. However, little work has been done so far to combine all of the three approaches.
We have developed a combined cost function-based approach in which the cost function has cost terms corresponding to each of the three approaches. We show in the experiments using RWC-POP and the Beatles datasets that the accuracy of segmentation is increased by combining the approaches.
|
