Generated music is only for 1 verse #388
Replies: 5 comments 1 reply
-
|
If you don’t have a language model (LM) for duration planning and you input a duration of -1, the result will be random. This is not a bug. The diffusion model (DiT) cannot determine the duration on its own unless you specify it manually. You need to use think mode. |
Beta Was this translation helpful? Give feedback.
-
i did it in think mode
now I've set 240 seconds for the music, given it a structure with verses and choruses, got a long intro, one verse, chorus, long play, chorus, and that's it... I can even show that I gave him to you and the file itself that turned out... How do you create a song with all the words in it??? c8962e51-2305-3afc-904d-3499dcfbcdf5.mp3 description: Words: |
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.
-
Beta Was this translation helpful? Give feedback.

Uh oh!
There was an error while loading. Please reload this page.
-
Describe the bug
I described the style in custom mode, inserted the text for the music below, there are 4 verses, but after generating I got only 1:20 of the music with a blackout at the end...
Why wasn't the whole song generated?
Audio Duration (seconds): -1
That is, it should have automatically selected the length of the song!
Why is there no mode for the song to be created entirely by default, as in other generators?
How do I choose the length of the song?? You can lose half a day on this until you find what you need, during this time the song itself can be lost, in the sense that the first version that was created suits me completely, but if I click generate again, then there will be another version, do you understand what the problem is, I hope?
Beta Was this translation helpful? Give feedback.
All reactions