This fork introduces a Radio Station feature where AI generates continuous radio music. The process involves two key components:
LLM: Generates the lyrics for the songs. ACE: Composes the music for the generated lyrics.
Due to the limitations of slower PCs, the demo video includes noticeable gaps (approximately 4 minutes) between the generated songs.
If your computer struggles to stream songs continuously, increasing the buffer size will result in a longer initial delay but fewer gaps between songs (until the buffer is depleted again).
By default the app attempts to load the model file gemma-3-12b-it-abliterated.q4_k_m.gguf from the same directory. However, you can also use alternative LLMs. Note that the quality of generated lyrics will vary depending on the LLM’s capabilities.
From a Dev and tinker perspective: this is neat!
From a personal perspective: wtf, this literally just accelerates a loss of art.
Thank you for attending my self-diverging opinions chat.
It’s scary and amazing at the same time, there are some jobs that are a drag (drag/dragself) like nobody cares that alarm clocks replaced door knockers or that street lights being turned on with a switch or timer replaced people who lit the lights every night, technology replacing those things is great but art feels different
I’m fairly certain in the future we’ll have art that applies to our every desire but right now it’s weird, I think this covers a lot of it (very very long article)
https://waitbutwhy.com/2015/01/artificial-intelligence-revolution-1.html
Until then we’re in a weird situation where AI is pumping out some genuine great stuff
A Sniper’s Dream EXTENDED FULL VERSION https://www.youtube.com/watch?v=R37IfQg0eTQ
But at the same time the idea of computers completely replacing human artists/everything is scary af because we’ve never had such a thing and have no plan/idea what happens next
Maybe this is something like process art, where it’s more about the process of its making, and not about the end product.
I’ll agree, though. AI is highly problematic for art, artists and probably all of society. And that is for a multitude of reasons. I’m a bit less worried about radio and pop music, since that doesn’t seem like art to me anyways. 😅
Nice idea. I recenty hacked together an AI radio station that mimics two radio hosts and queues music from Spotify. Somewhat similar to the Podcast scripts ( document-to-podcast, podcast-llm, podcastify, notebooklm ) just with the instructions to make up dialogue for a radio show and pick songs. Seems to work very well. Now I need it to generate 2h mp3 files and somehow connect that to Icecast, in order to stream it.
Also nice to see gemma-3-12b can do rhymes. I’ve always had issues with that and resorted to ChatGPT to write song lyrics, since the small local models couldn’t really do it properly. Is gemma3 a good pick for this? Any other local models I might want to try for song lyrics?
lmao
