DmitryUlyanov/neural-style-audio-tf

what does it take to produce longer outputs?

ianni67 opened this issue · 3 comments

ianni67 commented

hello Dmitry,
a quick question. How do I produce longer output files with this approach? Should I necessary provide longer inputs or there is another way?
Thank you very much for sharing your results
Giancarlo

DmitryUlyanov commented

Hello, there's nothing that prevents using longer inputs, it probably just requires changing several lines.

lkoz commented

Hello, I am also wondering the same thing. Could you point me to what would need to be changed in order to get longer outputs ?

msm1089 commented

@ianni67 @lkoz If you use longer inputs, you can change the value in the 4th cell, below line:

S = np.log1p(np.abs(S[:,:430]))

So for 30s, use 430 x 3 = 1290 etc. Well, that will give you just under 30s actually.

@DmitryUlyanov What does the value there represent? Is there some formula that gives you the length of the output based on this value?

  • 👍1