Musical Instrument Audio Identification Based on Kernel Logistic Regression
Audio classification based on statistical learning has attracted widespread attention and been widely put into some commercial application, because of better theoretical foundation and simple implementation mechanisms. Based on exploration the theory of the classical logistic regression (LR) and kernel logistic regression (KLR), a novel approach for audio classifier is put forward with the help of KLR in this paper. It is used to handle music from the same type of musical instruments. Music signals are collected with violin, viola and cello, and all the signals are preprocessed to extract features. The processed samples are used in experiments, while the classification performances are compared with 3 different kernel functions. Simulation results show that KLR performs better than traditional LR on classification accuracy and has better non-linear processing ability. Furthermore, KLR model with RBF kernel function can have a better stability besides good prediction performances.
kernel logistic regression audio classification feature extraction
Zunxiong Liu Jinfeng Xu
School of Information Engineering East China Jiaotong University Nanchang, China
国际会议
The 10th International Conference on Intelligent Technologies(第十届智慧科技国际会议 InTech09)
桂林
英文
31-35
2009-12-12(万方平台首次上网日期,不代表论文的发表时间)