Replace Text with audio+Text

[arielkantorovich]

Hi, thank you for the great repo, I will want to try generate 3D from audio to succeed in this experiment I will want to take your project and just replace the guidance DDPM (Text to the image) with another one for example (Audio to the image) or (audio + text to image). I am not familiar with all your code so I get lost, I will be happy if you can focus me on the area of the code so that I can replace the model and check this feature because it can be awesome l generate 3D from audio.