arthurfortes/speech2text_keras

Dimension issues

Opened this issue · 2 comments

Hey ! First of all thanks for your work !

I have issues with some parts of your code :
if(len(samples)== 8000) :
all_wave.append(samples)
all_label.append(label)
I don't understand why do we check length of 8000... Isn't it the sample rate ? Because none of the samples (signal data ?) have a 8000 length... Maybe I'm wrong somewhere ?

Other problem is this one :
all_wave = np.array(all_wave).reshape(-1,8000,1)

error : "cannot reshape array of size 493 into shape (8000, 1)
(493 is the size of my "all_wave" list, aka the list of all my samples. )

Thanks for your attention, don't hesitate to tell me if I'm wrong.

Hi, I am also facing the same issue. Did you find anything about it?

I have the same issue for a randomly selected wav file from internet to do inference.