Type | |
Stats | 1,328 |
Reviews | (227) |
Published | Dec 20, 2022 |
Base Model | |
Hash | AutoV2 641C2CA3F2 |
Note: I am not the original creator or owner of this model. Credit goes to Seth Forsgren and Hayk Martiros https://www.riffusion.com/about
I put together a quick Riffusion tutorial explaining how to get started, with some samples, etc - The Ally's Riffusion Tutorial
This is a novel txt2audio model - and it's wild. Select the model, give it a prompt: "Chill Beats", for example, and it'll output Spectrogram images.
These images can be used in Automatic 1111's SD WebUI, in conjunction with the WebUI Riffusion extension (https://github.com/enlyth/sd-webui-riffusion) to generate music inside the interface.
"Clips" (Spectrogram images) can be chained together using the Seed Travel script, and merged into a single audio file from the extension Riffusion tab.
Here's an awesome example of what's possible - https://youtu.be/Je1CD-gZI6o
Checkpoint has been pickle scanned, is legit, but Safetensors version also available, just in case!
Please consider joining my Patreon! Advanced SD tutorials, settings explanations, adult-art, from a female content creator (me!) patreon.com/theally - Riffusion tutorial to go live soon!