| Presentation | 2014-11-18 An Infinite Superimposed Discrete All-pole Model based on Gamma Processes for Music Signal Analysis Kazuyoshi YOSHII, Katsutoshi ITOYAMA, Masataka GOTO, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper presents a probabilistic model that is used for decomposing music audio signals into the three elements of sound: pitches (fundamental frequencies: F0s), timbres (spectral envelopes), and volumes. Conventional composite autoregressive models cannot deal with log-frequency spectrograms that match the characteristics of human auditory perception. When pithed sounds are analyzed, it is inappropriate to focus on all the frequency bands not limited to harmonic partials for estimating the spectral envelopes. To solve these problems, we propose a non-parametric Bayesian model based on gamma processes called an infinite superimposed discrete all-pole model by incorporating the idea of discrete all-pole modeling into a composite autoregressive model. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Music signal analysis / NMF / source separation / gamma process / nonparametric Bayes |
| Paper # | IBISML2014-60 |
| Date of Issue |
| Conference Information | |
| Committee | IBISML |
|---|---|
| Conference Date | 2014/11/10(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Information-Based Induction Sciences and Machine Learning (IBISML) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | An Infinite Superimposed Discrete All-pole Model based on Gamma Processes for Music Signal Analysis |
| Sub Title (in English) | |
| Keyword(1) | Music signal analysis |
| Keyword(2) | NMF |
| Keyword(3) | source separation |
| Keyword(4) | gamma process |
| Keyword(5) | nonparametric Bayes |
| 1st Author's Name | Kazuyoshi YOSHII |
| 1st Author's Affiliation | Kyoto University() |
| 2nd Author's Name | Katsutoshi ITOYAMA |
| 2nd Author's Affiliation | Kyoto University |
| 3rd Author's Name | Masataka GOTO |
| 3rd Author's Affiliation | National Institute of Advanced Industrial Science and Technology (AIST) |
| Date | 2014-11-18 |
| Paper # | IBISML2014-60 |
| Volume (vol) | vol.114 |
| Number (no) | 306 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |