Skip to main content
EURASIP Journal on Audio, Speech, and Music Processing

Table 4 MUSIB evaluation on Small-IrishFolk

From: MUSIB: musical score inpainting benchmark

Small IrishFolk Dataset (\(\approx\)2.4K samples)

Model

\(NLL \downarrow\)

\(pos_{F1} \uparrow\)

\(pAcc \uparrow\)

\(rAcc \uparrow\)

\(S_{div} \downarrow\)

\(H_{div} \downarrow\)

\(GS_{div} \downarrow\)

Anticipation-RNN

0.808

0.879

0.295

0.744

0.071

0.485

0.163

InpaintNet

0.655

0.864

0.419

0.690

0.055

0.489

0.392

SketchNet

0.779

0.937

0.386

0.888

0.562

0.527

0.176

VLI

0.415

0.851

0.371

0.880

0.102

0.474

0.307

  1. Values in boldface highlight the best performance among the four models on each metric
Back to article page