Add Caption v2 training tournament comparisons
Browse files
README.md
CHANGED
|
@@ -47,8 +47,7 @@ All renders use a 40-second maximum duration, 30 flow steps, flow CFG 1.7, and s
|
|
| 47 |
- **NextLat**: [MiniMaxMusicTraining/soad-mm3-nextlat-20260822](https://huggingface.co/MiniMaxMusicTraining/soad-mm3-nextlat-20260822)
|
| 48 |
- **NextLat + XM**: [MiniMaxMusicTraining/soad-mm3-nextlat-xm-20260822](https://huggingface.co/MiniMaxMusicTraining/soad-mm3-nextlat-xm-20260822)
|
| 49 |
|
| 50 |
-
|
| 51 |
-
<summary><strong>No trigger word / off-genre descriptors</strong></summary>
|
| 52 |
|
| 53 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 54 |
|
|
@@ -100,10 +99,7 @@ Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean elec
|
|
| 100 |
|
| 101 |
</details>
|
| 102 |
|
| 103 |
-
|
| 104 |
-
|
| 105 |
-
<details>
|
| 106 |
-
<summary><strong>Trigger word only</strong></summary>
|
| 107 |
|
| 108 |
Caption: `system of a down, soad style music`
|
| 109 |
|
|
@@ -155,10 +151,7 @@ Caption: `system of a down, soad style music`
|
|
| 155 |
|
| 156 |
</details>
|
| 157 |
|
| 158 |
-
|
| 159 |
-
|
| 160 |
-
<details>
|
| 161 |
-
<summary><strong>Matching descriptors, no trigger word</strong></summary>
|
| 162 |
|
| 163 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 164 |
|
|
@@ -210,10 +203,7 @@ Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with lay
|
|
| 210 |
|
| 211 |
</details>
|
| 212 |
|
| 213 |
-
|
| 214 |
-
|
| 215 |
-
<details>
|
| 216 |
-
<summary><strong>Trigger word + matching descriptors</strong></summary>
|
| 217 |
|
| 218 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 219 |
|
|
@@ -267,8 +257,6 @@ Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gent
|
|
| 267 |
|
| 268 |
</details>
|
| 269 |
|
| 270 |
-
</details>
|
| 271 |
-
|
| 272 |
<details>
|
| 273 |
<summary><strong>Full song without regularisation</strong> (576 renders)</summary>
|
| 274 |
|
|
@@ -285,8 +273,7 @@ All renders use a 40-second maximum duration, 30 flow steps, flow CFG 1.7, and s
|
|
| 285 |
|
| 286 |
> **Run note:** These four runs resumed their learned LoRA weights from checkpoint 500 with fresh optimizer, sampler, and RNG state after the original local trainer state was lost.
|
| 287 |
|
| 288 |
-
|
| 289 |
-
<summary><strong>No trigger word / off-genre descriptors</strong></summary>
|
| 290 |
|
| 291 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 292 |
|
|
@@ -338,10 +325,7 @@ Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean elec
|
|
| 338 |
|
| 339 |
</details>
|
| 340 |
|
| 341 |
-
|
| 342 |
-
|
| 343 |
-
<details>
|
| 344 |
-
<summary><strong>Trigger word only</strong></summary>
|
| 345 |
|
| 346 |
Caption: `system of a down, soad style music`
|
| 347 |
|
|
@@ -393,10 +377,7 @@ Caption: `system of a down, soad style music`
|
|
| 393 |
|
| 394 |
</details>
|
| 395 |
|
| 396 |
-
|
| 397 |
-
|
| 398 |
-
<details>
|
| 399 |
-
<summary><strong>Matching descriptors, no trigger word</strong></summary>
|
| 400 |
|
| 401 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 402 |
|
|
@@ -448,10 +429,7 @@ Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with lay
|
|
| 448 |
|
| 449 |
</details>
|
| 450 |
|
| 451 |
-
|
| 452 |
-
|
| 453 |
-
<details>
|
| 454 |
-
<summary><strong>Trigger word + matching descriptors</strong></summary>
|
| 455 |
|
| 456 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 457 |
|
|
@@ -505,8 +483,6 @@ Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gent
|
|
| 505 |
|
| 506 |
</details>
|
| 507 |
|
| 508 |
-
</details>
|
| 509 |
-
|
| 510 |
<details>
|
| 511 |
<summary><strong>Random windowing + regularisation</strong> (432 renders)</summary>
|
| 512 |
|
|
@@ -520,8 +496,7 @@ All renders use a 40-second maximum duration, 30 flow steps, flow CFG 1.7, and s
|
|
| 520 |
- **NextLat**: [MiniMaxMusicTraining/soad-mm3-nextlat-20260823-10k-adamw8e-5-randwin128-reg128-nocapdrop](https://huggingface.co/MiniMaxMusicTraining/soad-mm3-nextlat-20260823-10k-adamw8e-5-randwin128-reg128-nocapdrop)
|
| 521 |
- **NextLat + XM**: [MiniMaxMusicTraining/soad-mm3-nextlat-xm-20260823-10k-adamw8e-5-randwin128-reg128-nocapdrop](https://huggingface.co/MiniMaxMusicTraining/soad-mm3-nextlat-xm-20260823-10k-adamw8e-5-randwin128-reg128-nocapdrop)
|
| 522 |
|
| 523 |
-
|
| 524 |
-
<summary><strong>No trigger word / off-genre descriptors</strong></summary>
|
| 525 |
|
| 526 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 527 |
|
|
@@ -569,10 +544,7 @@ Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean elec
|
|
| 569 |
|
| 570 |
</details>
|
| 571 |
|
| 572 |
-
|
| 573 |
-
|
| 574 |
-
<details>
|
| 575 |
-
<summary><strong>Trigger word only</strong></summary>
|
| 576 |
|
| 577 |
Caption: `system of a down, soad style music`
|
| 578 |
|
|
@@ -620,10 +592,7 @@ Caption: `system of a down, soad style music`
|
|
| 620 |
|
| 621 |
</details>
|
| 622 |
|
| 623 |
-
|
| 624 |
-
|
| 625 |
-
<details>
|
| 626 |
-
<summary><strong>Matching descriptors, no trigger word</strong></summary>
|
| 627 |
|
| 628 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 629 |
|
|
@@ -671,10 +640,7 @@ Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with lay
|
|
| 671 |
|
| 672 |
</details>
|
| 673 |
|
| 674 |
-
|
| 675 |
-
|
| 676 |
-
<details>
|
| 677 |
-
<summary><strong>Trigger word + matching descriptors</strong></summary>
|
| 678 |
|
| 679 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 680 |
|
|
@@ -724,8 +690,6 @@ Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gent
|
|
| 724 |
|
| 725 |
</details>
|
| 726 |
|
| 727 |
-
</details>
|
| 728 |
-
|
| 729 |
<details>
|
| 730 |
<summary><strong>Caption v2: training-window comparison</strong> (400 renders)</summary>
|
| 731 |
|
|
@@ -742,8 +706,7 @@ All renders use a 40-second maximum duration, 30 flow steps, flow CFG 1.7, and s
|
|
| 742 |
|
| 743 |
> **Run note:** Checkpoint coverage differs by leg; empty table cells were not rendered.
|
| 744 |
|
| 745 |
-
|
| 746 |
-
<summary><strong>No trigger word / off-genre descriptors</strong></summary>
|
| 747 |
|
| 748 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 749 |
|
|
@@ -795,10 +758,7 @@ Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean elec
|
|
| 795 |
|
| 796 |
</details>
|
| 797 |
|
| 798 |
-
|
| 799 |
-
|
| 800 |
-
<details>
|
| 801 |
-
<summary><strong>Trigger word only</strong></summary>
|
| 802 |
|
| 803 |
Caption: `system of a down, soad style music`
|
| 804 |
|
|
@@ -850,10 +810,7 @@ Caption: `system of a down, soad style music`
|
|
| 850 |
|
| 851 |
</details>
|
| 852 |
|
| 853 |
-
|
| 854 |
-
|
| 855 |
-
<details>
|
| 856 |
-
<summary><strong>Matching descriptors, no trigger word</strong></summary>
|
| 857 |
|
| 858 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 859 |
|
|
@@ -905,10 +862,7 @@ Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with lay
|
|
| 905 |
|
| 906 |
</details>
|
| 907 |
|
| 908 |
-
|
| 909 |
-
|
| 910 |
-
<details>
|
| 911 |
-
<summary><strong>Trigger word + matching descriptors</strong></summary>
|
| 912 |
|
| 913 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 914 |
|
|
@@ -962,8 +916,6 @@ Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gent
|
|
| 962 |
|
| 963 |
</details>
|
| 964 |
|
| 965 |
-
</details>
|
| 966 |
-
|
| 967 |
<details>
|
| 968 |
<summary><strong>Caption v2: singer-tag comparison</strong> (480 renders)</summary>
|
| 969 |
|
|
@@ -980,8 +932,7 @@ All renders use a 40-second maximum duration, 30 flow steps, flow CFG 1.7, and s
|
|
| 980 |
|
| 981 |
> **Run note:** Checkpoint coverage differs by leg; empty table cells were not rendered.
|
| 982 |
|
| 983 |
-
|
| 984 |
-
<summary><strong>No trigger word / off-genre descriptors</strong></summary>
|
| 985 |
|
| 986 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 987 |
|
|
@@ -1033,10 +984,7 @@ Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean elec
|
|
| 1033 |
|
| 1034 |
</details>
|
| 1035 |
|
| 1036 |
-
|
| 1037 |
-
|
| 1038 |
-
<details>
|
| 1039 |
-
<summary><strong>Trigger word only</strong></summary>
|
| 1040 |
|
| 1041 |
Caption: `system of a down, soad style music`
|
| 1042 |
|
|
@@ -1088,10 +1036,7 @@ Caption: `system of a down, soad style music`
|
|
| 1088 |
|
| 1089 |
</details>
|
| 1090 |
|
| 1091 |
-
|
| 1092 |
-
|
| 1093 |
-
<details>
|
| 1094 |
-
<summary><strong>Matching descriptors, no trigger word</strong></summary>
|
| 1095 |
|
| 1096 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 1097 |
|
|
@@ -1143,10 +1088,7 @@ Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with lay
|
|
| 1143 |
|
| 1144 |
</details>
|
| 1145 |
|
| 1146 |
-
|
| 1147 |
-
|
| 1148 |
-
<details>
|
| 1149 |
-
<summary><strong>Trigger word + matching descriptors</strong></summary>
|
| 1150 |
|
| 1151 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 1152 |
|
|
@@ -1200,8 +1142,6 @@ Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gent
|
|
| 1200 |
|
| 1201 |
</details>
|
| 1202 |
|
| 1203 |
-
</details>
|
| 1204 |
-
|
| 1205 |
## Evaluation lyrics
|
| 1206 |
|
| 1207 |
The original comparison lyrics are shared by the first four tournaments. The latest singer-tag
|
|
|
|
| 47 |
- **NextLat**: [MiniMaxMusicTraining/soad-mm3-nextlat-20260822](https://huggingface.co/MiniMaxMusicTraining/soad-mm3-nextlat-20260822)
|
| 48 |
- **NextLat + XM**: [MiniMaxMusicTraining/soad-mm3-nextlat-xm-20260822](https://huggingface.co/MiniMaxMusicTraining/soad-mm3-nextlat-xm-20260822)
|
| 49 |
|
| 50 |
+
### No trigger word / off-genre descriptors
|
|
|
|
| 51 |
|
| 52 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 53 |
|
|
|
|
| 99 |
|
| 100 |
</details>
|
| 101 |
|
| 102 |
+
### Trigger word only
|
|
|
|
|
|
|
|
|
|
| 103 |
|
| 104 |
Caption: `system of a down, soad style music`
|
| 105 |
|
|
|
|
| 151 |
|
| 152 |
</details>
|
| 153 |
|
| 154 |
+
### Matching descriptors, no trigger word
|
|
|
|
|
|
|
|
|
|
| 155 |
|
| 156 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 157 |
|
|
|
|
| 203 |
|
| 204 |
</details>
|
| 205 |
|
| 206 |
+
### Trigger word + matching descriptors
|
|
|
|
|
|
|
|
|
|
| 207 |
|
| 208 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 209 |
|
|
|
|
| 257 |
|
| 258 |
</details>
|
| 259 |
|
|
|
|
|
|
|
| 260 |
<details>
|
| 261 |
<summary><strong>Full song without regularisation</strong> (576 renders)</summary>
|
| 262 |
|
|
|
|
| 273 |
|
| 274 |
> **Run note:** These four runs resumed their learned LoRA weights from checkpoint 500 with fresh optimizer, sampler, and RNG state after the original local trainer state was lost.
|
| 275 |
|
| 276 |
+
### No trigger word / off-genre descriptors
|
|
|
|
| 277 |
|
| 278 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 279 |
|
|
|
|
| 325 |
|
| 326 |
</details>
|
| 327 |
|
| 328 |
+
### Trigger word only
|
|
|
|
|
|
|
|
|
|
| 329 |
|
| 330 |
Caption: `system of a down, soad style music`
|
| 331 |
|
|
|
|
| 377 |
|
| 378 |
</details>
|
| 379 |
|
| 380 |
+
### Matching descriptors, no trigger word
|
|
|
|
|
|
|
|
|
|
| 381 |
|
| 382 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 383 |
|
|
|
|
| 429 |
|
| 430 |
</details>
|
| 431 |
|
| 432 |
+
### Trigger word + matching descriptors
|
|
|
|
|
|
|
|
|
|
| 433 |
|
| 434 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 435 |
|
|
|
|
| 483 |
|
| 484 |
</details>
|
| 485 |
|
|
|
|
|
|
|
| 486 |
<details>
|
| 487 |
<summary><strong>Random windowing + regularisation</strong> (432 renders)</summary>
|
| 488 |
|
|
|
|
| 496 |
- **NextLat**: [MiniMaxMusicTraining/soad-mm3-nextlat-20260823-10k-adamw8e-5-randwin128-reg128-nocapdrop](https://huggingface.co/MiniMaxMusicTraining/soad-mm3-nextlat-20260823-10k-adamw8e-5-randwin128-reg128-nocapdrop)
|
| 497 |
- **NextLat + XM**: [MiniMaxMusicTraining/soad-mm3-nextlat-xm-20260823-10k-adamw8e-5-randwin128-reg128-nocapdrop](https://huggingface.co/MiniMaxMusicTraining/soad-mm3-nextlat-xm-20260823-10k-adamw8e-5-randwin128-reg128-nocapdrop)
|
| 498 |
|
| 499 |
+
### No trigger word / off-genre descriptors
|
|
|
|
| 500 |
|
| 501 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 502 |
|
|
|
|
| 544 |
|
| 545 |
</details>
|
| 546 |
|
| 547 |
+
### Trigger word only
|
|
|
|
|
|
|
|
|
|
| 548 |
|
| 549 |
Caption: `system of a down, soad style music`
|
| 550 |
|
|
|
|
| 592 |
|
| 593 |
</details>
|
| 594 |
|
| 595 |
+
### Matching descriptors, no trigger word
|
|
|
|
|
|
|
|
|
|
| 596 |
|
| 597 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 598 |
|
|
|
|
| 640 |
|
| 641 |
</details>
|
| 642 |
|
| 643 |
+
### Trigger word + matching descriptors
|
|
|
|
|
|
|
|
|
|
| 644 |
|
| 645 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 646 |
|
|
|
|
| 690 |
|
| 691 |
</details>
|
| 692 |
|
|
|
|
|
|
|
| 693 |
<details>
|
| 694 |
<summary><strong>Caption v2: training-window comparison</strong> (400 renders)</summary>
|
| 695 |
|
|
|
|
| 706 |
|
| 707 |
> **Run note:** Checkpoint coverage differs by leg; empty table cells were not rendered.
|
| 708 |
|
| 709 |
+
### No trigger word / off-genre descriptors
|
|
|
|
| 710 |
|
| 711 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 712 |
|
|
|
|
| 758 |
|
| 759 |
</details>
|
| 760 |
|
| 761 |
+
### Trigger word only
|
|
|
|
|
|
|
|
|
|
| 762 |
|
| 763 |
Caption: `system of a down, soad style music`
|
| 764 |
|
|
|
|
| 810 |
|
| 811 |
</details>
|
| 812 |
|
| 813 |
+
### Matching descriptors, no trigger word
|
|
|
|
|
|
|
|
|
|
| 814 |
|
| 815 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 816 |
|
|
|
|
| 862 |
|
| 863 |
</details>
|
| 864 |
|
| 865 |
+
### Trigger word + matching descriptors
|
|
|
|
|
|
|
|
|
|
| 866 |
|
| 867 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 868 |
|
|
|
|
| 916 |
|
| 917 |
</details>
|
| 918 |
|
|
|
|
|
|
|
| 919 |
<details>
|
| 920 |
<summary><strong>Caption v2: singer-tag comparison</strong> (480 renders)</summary>
|
| 921 |
|
|
|
|
| 932 |
|
| 933 |
> **Run note:** Checkpoint coverage differs by leg; empty table cells were not rendered.
|
| 934 |
|
| 935 |
+
### No trigger word / off-genre descriptors
|
|
|
|
| 936 |
|
| 937 |
Caption: `Alternative pop-jazz, 102 BPM, brushed drums, upright bass, clean electric piano, airy male vocals, smooth verse-chorus songwriting, restrained dynamics, polished intimate mix.`
|
| 938 |
|
|
|
|
| 984 |
|
| 985 |
</details>
|
| 986 |
|
| 987 |
+
### Trigger word only
|
|
|
|
|
|
|
|
|
|
| 988 |
|
| 989 |
Caption: `system of a down, soad style music`
|
| 990 |
|
|
|
|
| 1036 |
|
| 1037 |
</details>
|
| 1038 |
|
| 1039 |
+
### Matching descriptors, no trigger word
|
|
|
|
|
|
|
|
|
|
| 1040 |
|
| 1041 |
Caption: `Lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 1042 |
|
|
|
|
| 1088 |
|
| 1089 |
</details>
|
| 1090 |
|
| 1091 |
+
### Trigger word + matching descriptors
|
|
|
|
|
|
|
|
|
|
| 1092 |
|
| 1093 |
Caption: `system of a down, soad style music, lilting hypnotic arpeggios, a gentle melodic mid-tempo groove with layered harmonies, sudden explosive distorted riffs, unstable meter changes, theatrical male vocals, tightly syncopated bass and drums, abrupt shifts between fragile verses and frantic heavy choruses.`
|
| 1094 |
|
|
|
|
| 1142 |
|
| 1143 |
</details>
|
| 1144 |
|
|
|
|
|
|
|
| 1145 |
## Evaluation lyrics
|
| 1146 |
|
| 1147 |
The original comparison lyrics are shared by the first four tournaments. The latest singer-tag
|