Open snkashis opened 6 years ago
Hi! There are shouldn't be any limitations. My suggestions are:
python capture.py
and pronounce something to mic for 5-10 secondsThis also happened to me. I solved it by changing the sample rate of the audio from 16k to 8k.
Ana
My file is already in 8k sr and I am still getting blank prediction. Is their a size/length limitation?
worked after I added 'data = data.astype('int8')
New to it so that might be the sample rate
I'm testing with an 18 second long file WAV and the predictions array is empty - is there a recommended minimum duration? Here is the avprobe output