small output
when I try the the example it show full output with a long song/music . but when I tried inference API it only show a small output that can't be a song
for example the prompt "tree" in the API it show : L1/8Q:1/4=90M:4/
and always the same. but in the demo it show output :
L1/8 M:4/4 K:none [K:D]!mf! D2 D2 E2 F2 | G2 A2 B2 c2 | d2 e2 d2 cB | A2 G2 F4 |!<(! D4 E4!<)! |!f! F4 G4 | !>(! A4 B4!>)! | c4 d4- | d4 z4 | z8 |!p! d4 c4 | B4 A4 | G4 F2 F>F | F2 E>E E2 D>D | D2 C>C C2 B,>B, | B,2 A,>A, A,2 G,>G, | G,2 F,>F, F,2 E,>E, | E,2 D,>D, D,2 C,>C, | C,2 B,,>B,, B,,2 A,,>A,, | A,,2 G,,>G,, G,,2 F,,>F,, | F,,2 E,,>E,, E,,2 D,,>D,, | D,,2 C,,>C,, C,,4 |]
Use the script in https://github.com/sander-wood/text-to-music instead
Incredible work! if you all are interested, we are going to do multimodal models including with ABC notation. would love your help @sander-wood and anyone else's. https://huggingface.co/Multi-Domain-Expert-Layers