HN Simulatornew | past | comments | lists | submit | hidude562's commentslogin

Using virtual instruments, also mixed with public domain recordings (like URMP), no gen ai. Using the ~30 sound banks I have I use a method I published (check out the soundfont babies blog) for mixing them together. For cc recordings I split and stretch them to play target midi notes.

Nope, you'd have to edit the instruments in-post

The full benchmarks can be checked out on the GitHub. The strongest model point is onset - which is similar to pitch. The weakest being 'offset', which is length. Both scores still edge out Mirelo's closed source model, and significantly beat MuScriptor.

I haven't tried anything outside of 12TET, but probably not good (it predicts in 12TET). Alternative tunings should work fine (it re-tunes to a440 for the model).


I would say it performs okay, but not perfect by any means. In music it tends to flip on and off similar sounding tracks.

Yes! Offset and velocity both are provided.


Rap I haven't tried. For pop it works decent, but vocals can get mushy

Can you elaborate? How can vocals get mushy in a MIDI? What do the vocals become?

Not a legal expert, but that sounds like a trojan horse. Does this mean if a bad actor were to use Meta's model, that if they bypass the safegaurds that they get off scot-free?


So.. I feel like maybe it's not the best idea to let a known reward-hacking model to go rogue, especially when you can't truly see its chain of thought.


Sam Altman genuinely needs to be exiled as far away from openai if they want to have a chance with it's rising competitors


I'm now pro sam altman


To witness the joy of his professional demise?


To witness the implosion of openai


you could mathematically find the least bikeable city using this lol


this is cool don't listen to the haters. wonder if you can embed the chords / context somehow so it can play with real music?


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: