search

facebookresearch / audiocraft

Audiocraft is a library for audio processing and generation with deep learning. It features the state-of-the-art EnCodec audio compressor / tokenizer, along with MusicGen, a simple and controllable music generation LM with textual and melodic conditioning.
MIT License
20.76k stars 2.12k forks source link

You've compared the AI to MusicLM but when? MusicLM has degraded in quality due to trophies users give it. Also here is my comparison between the 2 AIs. #65

Open rickkadamss opened 1 year ago

rickkadamss commented 1 year ago

Hello! Sry about the now edited out word in the title my, it sounds rude oops. So from my day1 testing and later testing, yes it seems quality of [MusicLM] (to compare against yours) and prompt listening has degraded I think. That's what the trophies do, they change the model right? It is a nightmare for my AI documentation :) :( cry.

Also I tested MusicGen against MusicLM on hard advanced techno tests, yours seems to win 30% of the tests or so. And the hardest ones it fails - even though I ran them today June 11 on MusicLM about a month after its release.

MusicGen tests: https://soundcloud.com/immortal-discoveries/sets/musicgen-ai-tests-now-seems-worse-than-musiclm

MusicLM tests: https://soundcloud.com/immortal-discoveries/sets/adding-to-musiclm-playlist-the-one-with-200-if-no-prompt-go-to-link

Also a must see for you guys (plus 200 more on soundcloud): https://www.reddit.com/r/singularity/comments/13h0zyy/i_really_crank_out_music_tracks_with_musiclm_this/

FurkanGozukara commented 1 year ago

music gen is open source

now even supports over 30 seconds with community fork :)

mcabel commented 1 year ago

music gen is open source

now even supports over 30 seconds with community fork :)

what is the community fork with over 30s called?

ManOrMonster commented 1 year ago

This is where I got the info: https://github.com/FurkanGozukara/Stable-Diffusion/blob/main/Tutorials/AI-Music-Generation-Audiocraft-Tutorial.md#11-june-2023