Noise inserted at end of audio files after uploaded

There seems to be an issue that when I upload small audio samples, a bit of noice/static gets appended to the end of the files:

What’s the effect of this on training/inference? Is there an ideal audio file length to prevent this from happening?

Im using the edge-impulse-uploader cli

Could you send me the wav file (jan@edgeimpulse.com)? Looks like something gets wrongly parsed there.