mirror of
https://github.com/multimodal-art-projection/YuE.git
synced 2026-10-02 01:34:32 +08:00
Ruibin Yuan
88da114a67
Add Suno v6 and Suno v6 Wild to the WildSongBench comparison and use the author-approved wording **competitive with Suno v5/v6**. The README now covers 17 settings; the detailed table and CSV/JSON provide all nine reported metrics. Figure 1 and its normalization data match the updated demo and paper. Validated all 153 aggregate values against the frozen September 12 results, confirmed the previous 15 settings are unchanged, independently checked the plotted coordinates and Pareto set, and verified image-pixel and PDF/SVG parity with the demo. Inference code, checkpoints, listening examples, and cover/editing results are unchanged.
317 lines
9.7 KiB
JSON
317 lines
9.7 KiB
JSON
{
|
||
"snapshot": "2026-09-12",
|
||
"scope": "15 comparator systems and 2 YuE2 settings on WildSongBench; recorded generation and selection protocols, not matched compute.",
|
||
"datasets": [
|
||
{
|
||
"id": "WSB",
|
||
"label": "WildSongBench",
|
||
"prompts": 192
|
||
}
|
||
],
|
||
"metrics": [
|
||
{
|
||
"key": "songbench_global_avg",
|
||
"label": "SongBench",
|
||
"description": "Seven-dimension global average",
|
||
"maximum": 10,
|
||
"lowerIsBetter": false
|
||
},
|
||
{
|
||
"key": "songbench_musicality",
|
||
"label": "SongBench Musicality",
|
||
"description": "SongBench musicality dimension",
|
||
"maximum": 10,
|
||
"lowerIsBetter": false
|
||
},
|
||
{
|
||
"key": "songeval_global_avg",
|
||
"label": "SongEval",
|
||
"description": "Five-dimension global average",
|
||
"maximum": 5,
|
||
"lowerIsBetter": false
|
||
},
|
||
{
|
||
"key": "songeval_musicality",
|
||
"label": "SongEval Musicality",
|
||
"description": "SongEval musicality dimension",
|
||
"maximum": 5,
|
||
"lowerIsBetter": false
|
||
},
|
||
{
|
||
"key": "audiobox_pq",
|
||
"label": "AudioBox PQ",
|
||
"description": "Production quality",
|
||
"maximum": 10,
|
||
"lowerIsBetter": false
|
||
},
|
||
{
|
||
"key": "mulan",
|
||
"label": "MuLan",
|
||
"description": "Style–audio similarity",
|
||
"maximum": 1,
|
||
"lowerIsBetter": false
|
||
},
|
||
{
|
||
"key": "allmusiccaps",
|
||
"label": "AllMusicCaps",
|
||
"description": "Style–audio similarity",
|
||
"maximum": 1,
|
||
"lowerIsBetter": false
|
||
},
|
||
{
|
||
"key": "control_overall",
|
||
"label": "Q3O Control",
|
||
"description": "Prompt-weighted overall control",
|
||
"maximum": 5,
|
||
"lowerIsBetter": false
|
||
},
|
||
{
|
||
"key": "per",
|
||
"label": "PER",
|
||
"description": "Phoneme error rate",
|
||
"maximum": 1,
|
||
"lowerIsBetter": true
|
||
}
|
||
],
|
||
"rows": [
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "YuE2",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.731599404761909,
|
||
"songbench_musicality": 5.907463020833336,
|
||
"songeval_global_avg": 4.262452916666666,
|
||
"songeval_musicality": 4.2151218749999995,
|
||
"audiobox_pq": 8.25979095697403,
|
||
"mulan": 0.5068482221880307,
|
||
"allmusiccaps": 0.40536119728009606,
|
||
"control_overall": 4.681896986601725,
|
||
"per": 0.08443630690279864
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "YuE2 (best-of-8)",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.963229315476184,
|
||
"songbench_musicality": 6.266602604166664,
|
||
"songeval_global_avg": 4.296048333333334,
|
||
"songeval_musicality": 4.250610416666664,
|
||
"audiobox_pq": 8.271365446348986,
|
||
"mulan": 0.5050805467180908,
|
||
"allmusiccaps": 0.398045457637636,
|
||
"control_overall": 4.70087253751977,
|
||
"per": 0.09786539961704843
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "ACE-Step 1.5",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.011793824404761,
|
||
"songbench_musicality": 5.158814583333332,
|
||
"songeval_global_avg": 3.846484166666666,
|
||
"songeval_musicality": 3.805107291666667,
|
||
"audiobox_pq": 8.051833018660545,
|
||
"mulan": 0.43723272854307044,
|
||
"allmusiccaps": 0.38685302749703016,
|
||
"control_overall": 4.580894501177339,
|
||
"per": 0.07457884998484957
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "DiffRhythm 2",
|
||
"n": 192,
|
||
"songbench_global_avg": 5.242845982142859,
|
||
"songbench_musicality": 4.477539583333331,
|
||
"songeval_global_avg": 3.5244677083333333,
|
||
"songeval_musicality": 3.471093229166668,
|
||
"audiobox_pq": 7.978225273390611,
|
||
"mulan": 0.37817035724098486,
|
||
"allmusiccaps": 0.32549789230688475,
|
||
"control_overall": 4.087044602340498,
|
||
"per": 0.18413265339015286
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "HeartMuLa",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.248347544642855,
|
||
"songbench_musicality": 5.496267708333332,
|
||
"songeval_global_avg": 4.551859166666667,
|
||
"songeval_musicality": 4.5329015625,
|
||
"audiobox_pq": 8.293305036922296,
|
||
"mulan": 0.3823466453177389,
|
||
"allmusiccaps": 0.2786190857962841,
|
||
"control_overall": 3.490655976705673,
|
||
"per": 0.10705881594471056
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "LeVo 2",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.324671800595234,
|
||
"songbench_musicality": 5.4589927083333345,
|
||
"songeval_global_avg": 4.023369479166667,
|
||
"songeval_musicality": 3.981881250000001,
|
||
"audiobox_pq": 8.396623315910498,
|
||
"mulan": 0.3542381529114209,
|
||
"allmusiccaps": 0.2679880544101252,
|
||
"control_overall": 3.9457878565807953,
|
||
"per": 0.26116136186343974
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "MiniMax Music 2.6",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.3221788690476215,
|
||
"songbench_musicality": 5.443674479166666,
|
||
"songeval_global_avg": 4.098453333333334,
|
||
"songeval_musicality": 4.055989583333334,
|
||
"audiobox_pq": 8.171058195332686,
|
||
"mulan": 0.42510899043797207,
|
||
"allmusiccaps": 0.3669951342102043,
|
||
"control_overall": 4.568809841873251,
|
||
"per": 0.24545647955335212
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "MiniMax Music 3",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.282988392857145,
|
||
"songbench_musicality": 5.348229166666669,
|
||
"songeval_global_avg": 4.099352187500001,
|
||
"songeval_musicality": 4.0430729166666675,
|
||
"audiobox_pq": 8.28245102117459,
|
||
"mulan": 0.39279851930526394,
|
||
"allmusiccaps": 0.36085424468425725,
|
||
"control_overall": 4.43616720498247,
|
||
"per": 0.06268414232388418
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "Mureka 9",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.937748809523812,
|
||
"songbench_musicality": 6.048820833333333,
|
||
"songeval_global_avg": 4.411131979166667,
|
||
"songeval_musicality": 4.361940104166667,
|
||
"audiobox_pq": 8.022609589000544,
|
||
"mulan": 0.43942587031051517,
|
||
"allmusiccaps": 0.41017753163275,
|
||
"control_overall": 4.636807564214055,
|
||
"per": 0.11690337887061523
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "Muse",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.034903943452382,
|
||
"songbench_musicality": 5.169220312499998,
|
||
"songeval_global_avg": 3.788667708333335,
|
||
"songeval_musicality": 3.736199999999999,
|
||
"audiobox_pq": 8.051745663086573,
|
||
"mulan": 0.3936510290513979,
|
||
"allmusiccaps": 0.3465821466803997,
|
||
"control_overall": 4.403758449771435,
|
||
"per": 0.33418648580693827
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "SongBloom",
|
||
"n": 192,
|
||
"songbench_global_avg": 4.235043750000001,
|
||
"songbench_musicality": 3.4492635416666655,
|
||
"songeval_global_avg": 3.2051130208333345,
|
||
"songeval_musicality": 3.204842187499999,
|
||
"audiobox_pq": 8.153916046023369,
|
||
"mulan": 0.2697471845106823,
|
||
"allmusiccaps": 0.19264957021005102,
|
||
"control_overall": 3.0286910546158463,
|
||
"per": 0.19193351857701904
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "Suno v4.5",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.699528199404766,
|
||
"songbench_musicality": 5.831729166666669,
|
||
"songeval_global_avg": 4.366563958333332,
|
||
"songeval_musicality": 4.319757291666664,
|
||
"audiobox_pq": 8.254062737027803,
|
||
"mulan": 0.5021767822715143,
|
||
"allmusiccaps": 0.38734816436772235,
|
||
"control_overall": 4.41490242329515,
|
||
"per": 0.05798428222369787
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "Suno v5",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.872074330357141,
|
||
"songbench_musicality": 5.9918354166666665,
|
||
"songeval_global_avg": 4.357936874999999,
|
||
"songeval_musicality": 4.305118229166664,
|
||
"audiobox_pq": 8.169838989774385,
|
||
"mulan": 0.5428044088184834,
|
||
"allmusiccaps": 0.43525151138116297,
|
||
"control_overall": 4.5907186541713685,
|
||
"per": 0.080993611026942
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "Suno v5.5",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.714983407738093,
|
||
"songbench_musicality": 5.808658854166672,
|
||
"songeval_global_avg": 4.2151607291666675,
|
||
"songeval_musicality": 4.149727604166669,
|
||
"audiobox_pq": 8.195489337046942,
|
||
"mulan": 0.5088621161606474,
|
||
"allmusiccaps": 0.3917035845806822,
|
||
"control_overall": 4.591446755445953,
|
||
"per": 0.05958313723720433
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "Suno v6",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.556171130952383,
|
||
"songbench_musicality": 5.6557864583333375,
|
||
"songeval_global_avg": 4.308628645833337,
|
||
"songeval_musicality": 4.263533854166667,
|
||
"audiobox_pq": 8.129587392012278,
|
||
"mulan": 0.4916386870124067,
|
||
"allmusiccaps": 0.4304573973834825,
|
||
"control_overall": 4.625818883206533,
|
||
"per": 0.07580377922906821
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "Suno v6 Wild",
|
||
"n": 192,
|
||
"songbench_global_avg": 6.419507440476191,
|
||
"songbench_musicality": 5.564441145833332,
|
||
"songeval_global_avg": 4.219912500000002,
|
||
"songeval_musicality": 4.171640104166664,
|
||
"audiobox_pq": 8.178526304662228,
|
||
"mulan": 0.49989257991546765,
|
||
"allmusiccaps": 0.43164352465343353,
|
||
"control_overall": 4.589767688464477,
|
||
"per": 0.07454661500473854
|
||
},
|
||
{
|
||
"dataset": "WSB",
|
||
"system": "YuE 1",
|
||
"n": 192,
|
||
"songbench_global_avg": 4.916463244047622,
|
||
"songbench_musicality": 4.084726562499998,
|
||
"songeval_global_avg": 3.214967395833334,
|
||
"songeval_musicality": 3.1524088541666675,
|
||
"audiobox_pq": 7.868339642882347,
|
||
"mulan": 0.262323199029197,
|
||
"allmusiccaps": 0.2881557388876293,
|
||
"control_overall": 3.7300764989993715,
|
||
"per": 0.3637846445448743
|
||
}
|
||
]
|
||
}
|