I'd love if there was a way to batch convert, I have 87 tiny .wavs here

Hi,

Thanks for sharing this code. I was able to modify it to get images of 768x768 pictures for 13.374s of audio. My plan it is to train a bunch of images with dreambooth on Stable Diffusion 2.1 model, and then try to generate some new music

I am not sure if it will work, because of the patterns it generate.

I spend the whole day trying to do a code like yours (but I am not a programmer), to simply create an spectrogram from an audio, and then load this spectrogram to get the the original sound. The spectrogram has another structures at simply looking you can see some elements there, the high and low frecuencies. I am pretty surre in this way would work in Stable Diffusion.

So can you add this another way to do it?

Something like this:

audioin.zip

The site is sadly loading indefinitely for me on mobile (firefox)

Can i get something like api or sdk? Mass conversion in the web version is terribly inconvenient.
thanks in advance

when you convert a png file to a wave file you cant get image data back by wave file
so wav2png is lossy