Speech commands dataset


Is there a repository or a code base for the SpeechCommands dataset? I found the below link but it doesn’t have the code for adding background noise etc.


I believe the Dataset source code is here in torchaudio. It seems to exclude the ‘_background_noise_’ folder so you may need to make modification to include it.


Thanks for the reply. I think the addition of background noise is not that straightforward because the size of the noise signal is not equal to the audio signal size. The source code for its implementation in TensorFlow is in the below link. So I was wondering if something similar is there in PyTorch as well or not.