基于神经网络的自监督学习方法音频分离器（Matlab代码实现）-Toy模板网

这篇具有很好参考价值的文章主要介绍了基于神经网络的自监督学习方法音频分离器（Matlab代码实现）。希望对大家有所帮助。如果存在错误或未考虑完全的地方，请大家不吝赐教，您也可以点击"举报违法"按钮提交疑问。

💥1 概述

📚2 运行结果

🎉3 参考文献

👨‍💻4 Matlab代码

💥1 概述

神经网络的输入是混合（男性+女性）音频的振幅谱。神经网络的输出目标是男性说话者理想的软掩模。损失函数是使输出和输入目标之间的均方误差最小化。在输出端，利用混合信号的输出幅度谱和相位将音频STFT转换回时域，并采用自监督学习方法。训练集是混合音频的前40秒中理想软掩模和二进制掩模的分离结果，网络输出是验证集为混合音频后的后20秒中男性说话者预测的理想软掩膜。

📚2 运行结果

基于神经网络的自监督学习方法音频分离器（Matlab代码实现）

主函数部分代码：

firstTrainingAudioFile = "f.mp3";

secondTrainingAudioFile = "m.mp3";

C=1;%用于强化背景音乐

firstsongTrain = C*audioread(firstTrainingAudioFile);

secondsongTrain = audioread(secondTrainingAudioFile);

L=500000;%采样值

firstsongTrain = firstsongTrain(L:2*L);

secondsongTrain = secondsongTrain(L:2*L);%训练集

firstValidationAudioFile ="f.mp3";

secondValidationAudioFile = "m.mp3";

firstsongValidate = C*audioread(firstValidationAudioFile);

secondsongValidate = audioread(secondValidationAudioFile);

L1=1000000;

firstsongValidate = firstsongValidate(3*L1:4*L1);

secondsongValidate = secondsongValidate(2.5*L1:3.5*L1);%验证级

% 将训练信号缩放到相同的功率。将验证信号缩放到相同的功率。

firstsongTrain =firstsongTrain/norm(firstsongTrain);%训练集,

secondsongTrain = secondsongTrain/norm(secondsongTrain);

firstsongValidate = firstsongValidate/norm(firstsongValidate);%验证集

secondsongValidate = secondsongValidate/norm(secondsongValidate);

mixTrain = firstsongTrain + secondsongTrain;

mixTrain = mixTrain / max(mixTrain);

mixValidate = firstsongValidate + secondsongValidate;

mixValidate = mixValidate / max(mixValidate);

WindowLength = 128;

FFTLength = 128;

OverlapLength = 128-1;

Fs = 44000;

win = hann(WindowLength,"periodic");

audiowrite('est_mix.wav',mixValidate,Fs);

P_mix0 = stft(mixTrain,'Window',win,'OverlapLength',OverlapLength,'FFTLength',FFTLength);

P_f = abs(stft(firstsongTrain,'Window',win,'OverlapLength',OverlapLength,'FFTLength',FFTLength));

P_s = abs(stft(secondsongTrain,'Window',win,'OverlapLength',OverlapLength,'FFTLength',FFTLength));

N = 1 + FFTLength/2;

🎉3 参考文献

[1]鲁玉军,周世豪,胡小勇.基于BP神经网络和小波神经网络的太阳辐射强度预测[J].软件工程,2023,26(01):5-8+4.DOI:10.19644/j.cnki.issn2096-1472.2023.001.002.文章来源地址https://www.toymoban.com/news/detail-402855.html