2018/12/18 by Anish Acharya, Acharya, Anish
Computer Science · #Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Information Retrieval (cs.IR) #Machine Learning (cs.LG) #Music Technology and Sound Studies #Music and Audio Processing #Sound (cs.SD) #Speech and Audio Processing #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1901.02053
openalex publication_date 2018/12/18 · openalex created_date 2022/08/01 · openalex updated_date 2026/07/28
This work proposes a novel feature selection algorithm to classify Songs into\ndifferent groups. Classification of musical content is often a non-trivial job\nand still relatively less explored area. The main idea conveyed in this article\nis to come up with a new feature selection scheme that does the classification\njob elegantly and with high accuracy but with simpler but wisely chosen small\nnumber of features thus being less prone to over-fitting. This uses a very\nbasic general idea about the structure of the audio signal which is generally\nin the shape of a trapezium. So, using this general idea of the Musical\nCommunity we propose three frames to be considered and analyzed for feature\nextraction for each of the audio signal -- opening, stanzas and closing -- and\nit has been established with the help of a lot of experiments that this scheme\nleads to much efficient classification with less complex features in a low\ndimensional feature space thus is also a computationally less expensive method.\nStep by step analysis of feature extraction, feature ranking, dimensionality\nreduction using PCA has been carried in this article. Sequential Forward\nselection (SFS) algorithm is used to explore the most significant features both\nwith the raw Fisher Discriminant Ratio (FDR) and also with the significant\neigen-values after PCA. Also during classification extensive validation and\ncross validation has been done in a monte-carlo manner to ensure validity of\nthe claims.\n