学习矢量量化LVQ改进算法在音频信号识别中的应用
he Application of Improved Learning Vector Quantization(LVQ) Algorithm in the Sound Signal Recognition
本文对学习矢量量化LVQ神经网络进行了描述，并利用其改进算法LVQ2和LVQ2.1分别进行了音频信号识别的实验。采用Mel频率倒谱系数MFCC作为音频信号的识别特征参数，利用MATLAB实现了音频信号识别过程。实验表明，使用LVQ2.1算法的效果较好，更适合对所选样本进行识别。
his paper describes the principle of the learning vector quantization neural network, brings forward the improved algorithms LVQ2 and LVQ2.1 and at last does some sound signal recognition experiments with these improved algorithms. The experiments choose Mel Frequency Cepstrum Coefficient (MFCC) of the sound signal as the feature parameter for recognition and realize the whole recognition process in the MATLAB environment. The results show that the LVQ2.1 algorithm is better and more suitable for the sound signal recognition.
宋培培、万遂人
电子技术应用
学习向量量化MFCCMATLAB音频信号识别
learning vector quantizationMFCCMATLABrecognition of the sound signal
宋培培,万遂人.学习矢量量化LVQ改进算法在音频信号识别中的应用[EB/OL].(2009-03-14)[2025-08-04].http://www.paper.edu.cn/releasepaper/content/200903-498.点此复制
评论