[apps] [discord] [learn] [music] [prompts] [text] [text to image] [upscale] [video]
Looking to get started with AI art? A good place to start is one of the popular apps like DreamStudio , midjourney , Wombo , or NightCafe . You can get a quick sense of how you can use words and phrases to guide image generation. Read up on prompt engineering to improve your results. Then you may want to move on to using Google Colab notebooks linked below like Deforum. If you have a good nVidia GPU of your own then you can also use NMKD Stable Diffusion GUI or Visions of Chaos to run the most popular notebooks locally. If you want to train your own Ai models check out the Ai art model training page, for animations check Stable Diffusion animations .
There are a TON of shared Google Colab notebooks floating around for doing text to image with pre-trained GAN and diffusion models. I've been compiling the ones I come across and try out and find interesting. Please hit me up on twitter (@pharmapsychotic) if you know a cool notebook that I am missing! Stable Diffusion is most popular right now.
Check out the Upscaling Guide
To get good results with CLIP guided diffusion and VQGAN+CLIP you need to find the right words and phrases that will direct the neural network to the content and style you are looking for.
You can generate music with AI using OpenAI's Jukebox. You can prompt Jukebox with an artist and music genre or with a short audio clip in WAV format. It generates new music for you in phases of increasing quality (level_2, level_1, level_0) and takes about 8 hours on Colab.
There are quite a few Discord servers dedicated now to AI artists or discussing text to image techniques.
Has this guide been helpful to you? 🙂 You can follow me on twitter @pharmapsychotic and I'd love if you would