how to apply stft to audio signal?

7 次查看（过去 30 天）
studentmatlaber
studentmatlaber 2022-1-12
回答： Bjorn Gustavsson 2022-1-12
Hello everyone. This is my first time processing audio, so I'm a beginner at this. I want to apply stft to my audio signal. It needs to have a 256-point Hamming window length, 128-point overlap, and 256-point transform values. I repeat a study, it says there is a 256-point sample that corresponds to 16ms.
Is this code I wrote correct? I would be glad if you help.
[data,fs]=audioread('SA1.wav')
% data = 46797x1 double
% fs = 16000
d = seconds(16*10e-3);
stft(data,d,'Window',hamming(256),'OverlapLength',128,'FFTLength',256);

请先登录，再回答此问题。

回答（1 个）

Bjorn Gustavsson
Bjorn Gustavsson 2022-1-12
You might do this, but the call to seconds looks spooky. From audioread you get the sampling-frequency fs, that is the sampling-frequency you should send in to stft. If that matches the sampling-frequency from the study you're good. If that sampling-frequency doesn't match you will have to work - but making up a sampling-frequency or sampling-time array like you do cannot be anyting but an unnecessary risk, if you're right you might just as well have used fs if you're wrong you only set yourself up for more future pain and toil...
HTH

类别

在 Help Center 和 File Exchange 中查找有关 Measurements and Spatial Audio 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by