BriefGPT.xyz
Mar, 2017
音乐分类和回归任务的迁移学习
Transfer learning for music classification and regression tasks
HTML
PDF
Keunwoo Choi, György Fazekas, Mark Sandler, Kyunghyun Cho
TL;DR
本文介绍了一种基于转移学习的音乐分类和回归任务的方法,使用预训练的卷积网络提取出用于多个层次的特征向量进行音乐分类及回归。经实验证明,相较于传统的低、高层次的音乐特征和 MFCC 特征,使用卷积神经网络(convnet)作为特征提取方法的结果更好且具有通用性。
Abstract
In this paper, we present a
transfer learning
approach for
music classification
and regression tasks. We propose to use a pretrained convnet feature, a concatenated feature vector using activations of feature map
→