CNN-XGBoost混合模型在音频场景分类中的应用

Application of CNN-XGBoost Hybrid Model in Acoustic Scene Classification

在拥有海量数据和强大计算能力的人工智能时代,音频场景分类成为了场景理解的重要研究内容之一.针对音频场景分类建模困难和精确率不高的问题,本文提出一种基于卷积神经网络和极端梯度提升算法相结合的系统模型.首先,将预处理后的音频信号转换成梅尔声谱图,然后输入到卷积神经网络中完成抽象特征提取,最后利用极端梯度提升算法进行分类.为了评估模型的有效性,在城市音频场景UrbanSound8K数据集上进行分类性能测试,结果表明,该混合算法模型对音频场景的分类精确率可以达到89%,优于传统的神经网络算法模型,说明该混合模型对音频场景分类问题的有效性.

At the age of artificial intelligence w ith massive data and pow erful computing performance,acoustic scene classification has become one of the most important research contents in the field of scene understanding. To solve the problems of difficulty and low accuracy in audio scene classification modeling,this paper proposes a system model based on convolutional neural netw ork and extreme gradient boosting. Firstly,the preprocessed audio signals are transformed into M el spectrum,and then input to convolutional neural netw ork to extract abstract features. Finally,the extreme gradient boosting algorithm is used for classification. In order to evaluate the effectiveness of the model,the performance of the classification is tested on the UrbanSound8 K data set. The results show that the accuracy of the hybrid algorithm model for the classification of audio scenes can reach 89%,w hich is superior to the traditional neural netw ork algorithm model. The validity of the hybrid model for audio scene classification is verified.