Tanky

@tiesiogdvd The spectrum of 3 minutes is already huge. In business, it is generally obtained in real time, such as 128ms data every 32ms. For long audio, the business can...

@johanez Our current maximum supported sample rate is 196,000, and we plan to support even higher sample rates in the future. If possible, you can downsample the audio before extracting...

@AlexHFan You can compile in embedded according to the mobile process [https://audioflux.top/installation.html#building-for-mobile](https://audioflux.top/installation.html#building-for-mobile)

@DrewdropLife It sounds like you're encountering a compatibility issue related to the use of Intel Math Kernel Library (MKL) and OpenMP when using various Python scientific computing libraries like PyTorch,...

@SteveTanggithub Hi, You can refer to the following code: ``` int main(int argc, char const *argv[]){ SpectrogramObj spec=NULL; int status=0; int num=0; int samplate=32000; float lowFre=0; float highFre=16000; float *mDataArr=NULL;...

@fire717 Thanks for your test feedback, it revealed some performance issues with our project. This is mainly reflected in the feature extraction of long audio, for which we released latest...

@Viyyy Please provide the system version

> > @Viyyy Please provide the system version > > Thanks for your reply, here is my system version: Windows 10 IoT Enterprise. We tested on win11/win10, but we don't...