Spectrogram to audio?

daddydean

Hi there...im wondering if someone has ever implemented an image of a spectrogram to audio conversion in Processing?
The only thing I found was: https://www.ee.columbia.edu/~dpwe/resources/Processing/LiveSpectrogram.pde
But I believe this is the other way around.
Any tips or leads would be greatly appreciated!

Chrisir

Not possible

daddydean

Why not? It seems it can be done with something called the "griffin-lim" algorithm. https://github.com/bkvogel/griffin_lim

kfrajer

But I believe this is the other way around.

So the code in your first post read the audio stream, does the fft and displays the data? However, from what I understand, you have a graph and you want to create the sound? Is the graph in frequency domain? Do you have a sample data set?

Please also provide details of what you want to do. There is no need for everybody to go and read somebody else's code to try to understand what you would like to do. Provide a brief description and the links will be also needed but they should be use as support for your post.

Kf