3: MuQ / ECAPA / STARS モデル統合ランキング

入力音源・特徴量を明示した単体ランキングと、z-score標準化後の等重み統合ランキング

<<<表示名の見方>>>

各クリップ名は、楽曲名・フレーズ番号・区間ラベル・開始時刻・終了時刻を連結したものです。 ランキング算出への入力がmixture・vocalsのどちらであっても、結果欄では同一区間のvocals音声とmixture音声を両方再生できます。

例: BigTroubles_-_Phantom / BigTroubles_-_Phantom_0001_a_0009000_0019000
BigTroubles_-_Phantom
アーティスト名 + 曲名
0001
フレーズ / クリップ番号
a / b / c
Lyrics Extensionに含まれるsectionラベル
0009000
開始時刻 ms表記。0009000 = 9.0秒
0019000
終了時刻 ms表記。0019000 = 19.0秒

Query 1: Nerve_9_-_Pray_For_The_Rain / Nerve_9_-_Pray_For_The_Rain_0025_a_0205000_0209000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9781
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0023_a_0196000_0201000
vocals
mixture
2 0.9715
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0022_a_0192000_0196000
vocals
mixture
3 0.9666
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0024_a_0201000_0205000
vocals
mixture
4 0.9646
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0014_a_0106000_0112000
vocals
mixture
5 0.9632
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0026_a_0196000_0201000
vocals
mixture
6 0.9611
Sweet_Lights_-_You_Let_Me_Down
Sweet_Lights_-_You_Let_Me_Down_0021_b_0225000_0229000
vocals
mixture
7 0.9611
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0012_a_0090000_0097000
vocals
mixture
8 0.9607
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0015_c_0127000_0131000
vocals
mixture
9 0.9604
Music_Delta_-_Country1
Music_Delta_-_Country1_0003_a_0018000_0022000
vocals
mixture
10 0.9592
Drumtracks_-_Ghost_Bitch
Drumtracks_-_Ghost_Bitch_0045_c_0294000_0300000
vocals
mixture
11 0.9587
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0030_a_0223000_0228000
vocals
mixture
12 0.9576
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0021_a_0184000_0192000
vocals
mixture
13 0.9573
Sweet_Lights_-_You_Let_Me_Down
Sweet_Lights_-_You_Let_Me_Down_0006_c_0089000_0094000
vocals
mixture
14 0.9570
Ben_Carrigan_-_We_ll_Talk_About_It_All_Tonight
Ben_Carrigan_-_We_ll_Talk_About_It_All_Tonight_0017_a_0116000_0120000
vocals
mixture
15 0.9568
Creepoid_-_OldTree
Creepoid_-_OldTree_0014_a_0116000_0121000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9783
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0022_a_0192000_0196000
vocals
mixture
2 0.9741
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0023_a_0196000_0201000
vocals
mixture
3 0.9721
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0024_a_0201000_0205000
vocals
mixture
4 0.9711
Jay_Menon_-_Through_My_Eyes
Jay_Menon_-_Through_My_Eyes_0006_a_0038000_0042000
vocals
mixture
5 0.9700
Actions_-_Devil_s_Words
Actions_-_Devil_s_Words_0011_a_0072000_0078000
vocals
mixture
6 0.9696
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0026_a_0172000_0178000
vocals
mixture
7 0.9694
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0009_b_0060000_0065000
vocals
mixture
8 0.9683
Hop_Along_-_Sister_Cities
Hop_Along_-_Sister_Cities_0014_a_0120000_0125000
vocals
mixture
9 0.9673
Actions_-_South_Of_The_Water
Actions_-_South_Of_The_Water_0002_a_0015000_0019000
vocals
mixture
10 0.9673
The_Long_Wait_-_Dark_Horses
The_Long_Wait_-_Dark_Horses_0015_b_0122000_0127000
vocals
mixture
11 0.9671
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0027_a_0178000_0184000
vocals
mixture
12 0.9669
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0013_b_0102000_0108000
vocals
mixture
13 0.9668
AM_Contra_-_Heart_Peripheral
AM_Contra_-_Heart_Peripheral_0024_a_0138000_0143000
vocals
mixture
14 0.9668
Hop_Along_-_Sister_Cities
Hop_Along_-_Sister_Cities_0019_a_0162000_0168000
vocals
mixture
15 0.9667
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0026_a_0196000_0201000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 3.3925
MuQ_vocals: raw=0.9783, z=1.7223
ECAPA: raw=0.6060, z=5.0627
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0022_a_0192000_0196000
vocals
mixture
2 3.1633
MuQ_vocals: raw=0.9741, z=1.5793
ECAPA: raw=0.5768, z=4.7473
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0023_a_0196000_0201000
vocals
mixture
3 2.6618
MuQ_vocals: raw=0.9448, z=0.5880
ECAPA: raw=0.5757, z=4.7356
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0021_a_0184000_0192000
vocals
mixture
4 2.6483
MuQ_vocals: raw=0.9563, z=0.9767
ECAPA: raw=0.5372, z=4.3198
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0014_a_0106000_0112000
vocals
mixture
5 2.5799
MuQ_vocals: raw=0.9289, z=0.0536
ECAPA: raw=0.6100, z=5.1062
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0026_a_0209000_0218000
vocals
mixture
6 2.3889
MuQ_vocals: raw=0.9422, z=0.5005
ECAPA: raw=0.5333, z=4.2774
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0002_a_0017000_0025000
vocals
mixture
7 2.3228
MuQ_vocals: raw=0.9644, z=1.2518
ECAPA: raw=0.4515, z=3.3938
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0007_a_0062000_0066000
vocals
mixture
8 2.2339
MuQ_vocals: raw=0.9388, z=0.3860
ECAPA: raw=0.5152, z=4.0817
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0013_a_0097000_0106000
vocals
mixture
9 2.0780
MuQ_vocals: raw=0.9530, z=0.8665
ECAPA: raw=0.4418, z=3.2895
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0011_a_0077000_0086000
vocals
mixture
10 2.0759
MuQ_vocals: raw=0.9343, z=0.2355
ECAPA: raw=0.4999, z=3.9164
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0005_a_0038000_0047000
vocals
mixture
11 2.0454
MuQ_vocals: raw=0.9621, z=1.1732
ECAPA: raw=0.4074, z=2.9176
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0025_a_0189000_0196000
vocals
mixture
12 2.0084
MuQ_vocals: raw=0.9721, z=1.5122
ECAPA: raw=0.3692, z=2.5045
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0024_a_0201000_0205000
vocals
mixture
13 1.9750
MuQ_vocals: raw=0.9587, z=1.0586
ECAPA: raw=0.4050, z=2.8915
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0003_a_0025000_0032000
vocals
mixture
14 1.9566
MuQ_vocals: raw=0.9540, z=0.8990
ECAPA: raw=0.4163, z=3.0141
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0006_a_0043000_0048000
vocals
mixture
15 1.9515
MuQ_vocals: raw=0.9601, z=1.1070
ECAPA: raw=0.3962, z=2.7961
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0012_a_0090000_0097000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 1.4926
MuQ_vocals: raw=0.9711, z=1.4772
STARS_phone_hidden: raw=0.7509, z=1.5080
Jay_Menon_-_Through_My_Eyes
Jay_Menon_-_Through_My_Eyes_0006_a_0038000_0042000
vocals
mixture
2 1.3841
MuQ_vocals: raw=0.9632, z=1.2110
STARS_phone_hidden: raw=0.7590, z=1.5572
Hop_Along_-_Sister_Cities
Hop_Along_-_Sister_Cities_0009_a_0080000_0085000
vocals
mixture
3 1.3009
MuQ_vocals: raw=0.9641, z=1.2419
STARS_phone_hidden: raw=0.7267, z=1.3599
The_Long_Wait_-_Back_Home_To_Blue
The_Long_Wait_-_Back_Home_To_Blue_0021_a_0152000_0158000
vocals
mixture
4 1.3003
MuQ_vocals: raw=0.9667, z=1.3280
STARS_phone_hidden: raw=0.7124, z=1.2725
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0026_a_0196000_0201000
vocals
mixture
5 1.2422
MuQ_vocals: raw=0.9620, z=1.1696
STARS_phone_hidden: raw=0.7193, z=1.3147
Young_Griffo_-_Facade
Young_Griffo_-_Facade_0011_a_0072000_0079000
vocals
mixture
6 1.2416
MuQ_vocals: raw=0.9668, z=1.3325
STARS_phone_hidden: raw=0.6925, z=1.1507
Hop_Along_-_Sister_Cities
Hop_Along_-_Sister_Cities_0019_a_0162000_0168000
vocals
mixture
7 1.2041
MuQ_vocals: raw=0.9683, z=1.3823
STARS_phone_hidden: raw=0.6721, z=1.0260
Hop_Along_-_Sister_Cities
Hop_Along_-_Sister_Cities_0014_a_0120000_0125000
vocals
mixture
8 1.2011
MuQ_vocals: raw=0.9621, z=1.1734
STARS_phone_hidden: raw=0.7053, z=1.2288
The_Districts_-_Vermont
The_Districts_-_Vermont_0026_b_0186000_0190000
vocals
mixture
9 1.1861
MuQ_vocals: raw=0.9538, z=0.8950
STARS_phone_hidden: raw=0.7459, z=1.4773
AM_Contra_-_Heart_Peripheral
AM_Contra_-_Heart_Peripheral_0011_b_0066000_0072000
vocals
mixture
10 1.1754
MuQ_vocals: raw=0.9550, z=0.9343
STARS_phone_hidden: raw=0.7360, z=1.4164
The_Long_Wait_-_Back_Home_To_Blue
The_Long_Wait_-_Back_Home_To_Blue_0006_a_0043000_0049000
vocals
mixture
11 1.1653
MuQ_vocals: raw=0.9638, z=1.2303
STARS_phone_hidden: raw=0.6842, z=1.1003
Music_Delta_-_Beatles
Music_Delta_-_Beatles_0000_a_0000000_0004000
vocals
mixture
12 1.1613
MuQ_vocals: raw=0.9625, z=1.1879
STARS_phone_hidden: raw=0.6899, z=1.1347
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0052_a_0364000_0369000
vocals
mixture
13 1.1573
MuQ_vocals: raw=0.9622, z=1.1783
STARS_phone_hidden: raw=0.6901, z=1.1363
Young_Griffo_-_Facade
Young_Griffo_-_Facade_0022_a_0149000_0154000
vocals
mixture
14 1.1555
MuQ_vocals: raw=0.9741, z=1.5793
STARS_phone_hidden: raw=0.6239, z=0.7317
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0023_a_0196000_0201000
vocals
mixture
15 1.1473
MuQ_vocals: raw=0.9604, z=1.1179
STARS_phone_hidden: raw=0.6967, z=1.1767
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0027_b_0197000_0202000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 2.3528
MuQ_vocals: raw=0.9741, z=1.5793
ECAPA: raw=0.5768, z=4.7473
STARS_phone_hidden: raw=0.6239, z=0.7317
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0023_a_0196000_0201000
vocals
mixture
2 2.3148
MuQ_vocals: raw=0.9783, z=1.7223
ECAPA: raw=0.6060, z=5.0627
STARS_phone_hidden: raw=0.5302, z=0.1593
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0022_a_0192000_0196000
vocals
mixture
3 2.1049
MuQ_vocals: raw=0.9448, z=0.5880
ECAPA: raw=0.5757, z=4.7356
STARS_phone_hidden: raw=0.6664, z=0.9912
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0021_a_0184000_0192000
vocals
mixture
4 2.0362
MuQ_vocals: raw=0.9289, z=0.0536
ECAPA: raw=0.6100, z=5.1062
STARS_phone_hidden: raw=0.6594, z=0.9489
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0026_a_0209000_0218000
vocals
mixture
5 1.8997
MuQ_vocals: raw=0.9563, z=0.9767
ECAPA: raw=0.5372, z=4.3198
STARS_phone_hidden: raw=0.5700, z=0.4024
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0014_a_0106000_0112000
vocals
mixture
6 1.7092
MuQ_vocals: raw=0.9343, z=0.2355
ECAPA: raw=0.4999, z=3.9164
STARS_phone_hidden: raw=0.6639, z=0.9759
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0005_a_0038000_0047000
vocals
mixture
7 1.7053
MuQ_vocals: raw=0.9530, z=0.8665
ECAPA: raw=0.4418, z=3.2895
STARS_phone_hidden: raw=0.6612, z=0.9598
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0011_a_0077000_0086000
vocals
mixture
8 1.6885
MuQ_vocals: raw=0.9601, z=1.1070
ECAPA: raw=0.3962, z=2.7961
STARS_phone_hidden: raw=0.6944, z=1.1624
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0012_a_0090000_0097000
vocals
mixture
9 1.6432
MuQ_vocals: raw=0.9621, z=1.1732
ECAPA: raw=0.4074, z=2.9176
STARS_phone_hidden: raw=0.6414, z=0.8388
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0025_a_0189000_0196000
vocals
mixture
10 1.6392
MuQ_vocals: raw=0.9322, z=0.1646
ECAPA: raw=0.4507, z=3.3852
STARS_phone_hidden: raw=0.7280, z=1.3678
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0005_a_0035000_0043000
vocals
mixture
11 1.6325
MuQ_vocals: raw=0.9422, z=0.5005
ECAPA: raw=0.5333, z=4.2774
STARS_phone_hidden: raw=0.5237, z=0.1198
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0002_a_0017000_0025000
vocals
mixture
12 1.5818
MuQ_vocals: raw=0.9388, z=0.3860
ECAPA: raw=0.5152, z=4.0817
STARS_phone_hidden: raw=0.5496, z=0.2776
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0013_a_0097000_0106000
vocals
mixture
13 1.5799
MuQ_vocals: raw=0.9721, z=1.5122
ECAPA: raw=0.3692, z=2.5045
STARS_phone_hidden: raw=0.6225, z=0.7229
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0024_a_0201000_0205000
vocals
mixture
14 1.5622
MuQ_vocals: raw=0.9405, z=0.4448
ECAPA: raw=0.3758, z=2.5760
STARS_phone_hidden: raw=0.7768, z=1.6659
AM_Contra_-_Heart_Peripheral
AM_Contra_-_Heart_Peripheral_0019_b_0110000_0118000
vocals
mixture
15 1.5256
MuQ_vocals: raw=0.9482, z=0.7058
ECAPA: raw=0.4275, z=3.1350
STARS_phone_hidden: raw=0.6246, z=0.7360
Actions_-_One_Minute_Smile
Actions_-_One_Minute_Smile_0012_b_0074000_0080000
vocals
mixture

Query 2: Secretariat_-_Borderline / Secretariat_-_Borderline_0012_a_0108000_0114000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9643
Secretariat_-_Borderline
Secretariat_-_Borderline_0024_a_0201000_0209000
vocals
mixture
2 0.9642
The_Scarlet_Brand_-_Les_Fleurs_Du_Mal
The_Scarlet_Brand_-_Les_Fleurs_Du_Mal_0014_a_0089000_0095000
vocals
mixture
3 0.9637
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0002_a_0017000_0023000
vocals
mixture
4 0.9633
Invisible_Familiars_-_Disturbing_Wildlife
Invisible_Familiars_-_Disturbing_Wildlife_0023_c_0170000_0177000
vocals
mixture
5 0.9632
Invisible_Familiars_-_Disturbing_Wildlife
Invisible_Familiars_-_Disturbing_Wildlife_0015_b_0116000_0123000
vocals
mixture
6 0.9626
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0027_a_0169000_0176000
vocals
mixture
7 0.9626
The_So_So_Glos_-_Emergency
The_So_So_Glos_-_Emergency_0005_a_0027000_0033000
vocals
mixture
8 0.9624
Invisible_Familiars_-_Disturbing_Wildlife
Invisible_Familiars_-_Disturbing_Wildlife_0017_b_0129000_0136000
vocals
mixture
9 0.9602
Drumtracks_-_Ghost_Bitch
Drumtracks_-_Ghost_Bitch_0007_a_0047000_0053000
vocals
mixture
10 0.9601
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0004_a_0028000_0034000
vocals
mixture
11 0.9600
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0012_a_0075000_0081000
vocals
mixture
12 0.9598
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0011_a_0069000_0076000
vocals
mixture
13 0.9598
Hezekiah_Jones_-_Borrowed_Heart
Hezekiah_Jones_-_Borrowed_Heart_0023_a_0172000_0177000
vocals
mixture
14 0.9597
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0022_a_0172000_0179000
vocals
mixture
15 0.9596
Invisible_Familiars_-_Disturbing_Wildlife
Invisible_Familiars_-_Disturbing_Wildlife_0021_a_0157000_0164000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9801
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0031_a_0228000_0234000
vocals
mixture
2 0.9764
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0012_a_0083000_0090000
vocals
mixture
3 0.9743
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0008_a_0058000_0064000
vocals
mixture
4 0.9729
The_Scarlet_Brand_-_Les_Fleurs_Du_Mal
The_Scarlet_Brand_-_Les_Fleurs_Du_Mal_0014_a_0089000_0095000
vocals
mixture
5 0.9720
Mu_-_Too_Bright
Mu_-_Too_Bright_0018_a_0140000_0146000
vocals
mixture
6 0.9707
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0039_a_0257000_0264000
vocals
mixture
7 0.9706
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0005_a_0034000_0041000
vocals
mixture
8 0.9705
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0011_a_0077000_0086000
vocals
mixture
9 0.9701
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0006_a_0046000_0052000
vocals
mixture
10 0.9700
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0004_a_0028000_0034000
vocals
mixture
11 0.9698
Auctioneer_-_Our_Future_Faces
Auctioneer_-_Our_Future_Faces_0000_a_0000000_0008000
vocals
mixture
12 0.9698
Secretariat_-_Borderline
Secretariat_-_Borderline_0024_a_0201000_0209000
vocals
mixture
13 0.9698
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0005_a_0036000_0044000
vocals
mixture
14 0.9698
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0029_a_0216000_0223000
vocals
mixture
15 0.9697
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0020_a_0146000_0154000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 3.1873
MuQ_vocals: raw=0.9686, z=1.3460
ECAPA: raw=0.6795, z=5.0286
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0006_a_0041000_0048000
vocals
mixture
2 3.1487
MuQ_vocals: raw=0.9592, z=0.9765
ECAPA: raw=0.7103, z=5.3209
Secretariat_-_Borderline
Secretariat_-_Borderline_0014_a_0118000_0128000
vocals
mixture
3 3.1003
MuQ_vocals: raw=0.9656, z=1.2295
ECAPA: raw=0.6734, z=4.9711
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0005_a_0034000_0041000
vocals
mixture
4 3.0646
MuQ_vocals: raw=0.9700, z=1.4017
ECAPA: raw=0.6478, z=4.7275
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0004_a_0028000_0034000
vocals
mixture
5 2.8694
MuQ_vocals: raw=0.9764, z=1.6513
ECAPA: raw=0.5803, z=4.0874
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0012_a_0083000_0090000
vocals
mixture
6 2.7730
MuQ_vocals: raw=0.9729, z=1.5142
ECAPA: raw=0.5745, z=4.0319
The_Scarlet_Brand_-_Les_Fleurs_Du_Mal
The_Scarlet_Brand_-_Les_Fleurs_Du_Mal_0014_a_0089000_0095000
vocals
mixture
7 2.7494
MuQ_vocals: raw=0.9616, z=1.0714
ECAPA: raw=0.6162, z=4.4274
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0014_b_0104000_0110000
vocals
mixture
8 2.7431
MuQ_vocals: raw=0.9698, z=1.3932
ECAPA: raw=0.5809, z=4.0929
Secretariat_-_Borderline
Secretariat_-_Borderline_0024_a_0201000_0209000
vocals
mixture
9 2.7390
MuQ_vocals: raw=0.9801, z=1.7965
ECAPA: raw=0.5376, z=3.6815
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0031_a_0228000_0234000
vocals
mixture
10 2.7261
MuQ_vocals: raw=0.9528, z=0.7268
ECAPA: raw=0.6476, z=4.7255
Secretariat_-_Borderline
Secretariat_-_Borderline_0005_a_0049000_0057000
vocals
mixture
11 2.7192
MuQ_vocals: raw=0.9458, z=0.4517
ECAPA: raw=0.6751, z=4.9868
Secretariat_-_Borderline
Secretariat_-_Borderline_0015_b_0128000_0139000
vocals
mixture
12 2.6419
MuQ_vocals: raw=0.9502, z=0.6230
ECAPA: raw=0.6407, z=4.6607
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0009_b_0058000_0068000
vocals
mixture
13 2.6136
MuQ_vocals: raw=0.9431, z=0.3436
ECAPA: raw=0.6642, z=4.8835
Secretariat_-_Borderline
Secretariat_-_Borderline_0016_a_0138000_0147000
vocals
mixture
14 2.5106
MuQ_vocals: raw=0.9695, z=1.3824
ECAPA: raw=0.5331, z=3.6387
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0020_a_0142000_0149000
vocals
mixture
15 2.4596
MuQ_vocals: raw=0.9585, z=0.9512
ECAPA: raw=0.5678, z=3.9680
Secretariat_-_Borderline
Secretariat_-_Borderline_0010_b_0090000_0098000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 1.7077
MuQ_vocals: raw=0.9585, z=0.9513
STARS_phone_hidden: raw=0.9157, z=2.4642
Girls_Under_Glass_-_We_Feel_Alright
Girls_Under_Glass_-_We_Feel_Alright_0021_b_0135000_0142000
vocals
mixture
2 1.6615
MuQ_vocals: raw=0.9604, z=1.0258
STARS_phone_hidden: raw=0.8867, z=2.2972
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0023_c_0152000_0157000
vocals
mixture
3 1.5294
MuQ_vocals: raw=0.9620, z=1.0863
STARS_phone_hidden: raw=0.8302, z=1.9725
Georgia_Wonder_-_Siren
Georgia_Wonder_-_Siren_0008_a_0055000_0062000
vocals
mixture
4 1.5294
MuQ_vocals: raw=0.9525, z=0.7140
STARS_phone_hidden: raw=0.8949, z=2.3447
The_Doppler_Shift_-_Atrophy
The_Doppler_Shift_-_Atrophy_0007_a_0052000_0057000
vocals
mixture
5 1.5106
MuQ_vocals: raw=0.9519, z=0.6909
STARS_phone_hidden: raw=0.8924, z=2.3302
Carlos_Gonzalez_-_A_Place_For_Us
Carlos_Gonzalez_-_A_Place_For_Us_0024_c_0170000_0177000
vocals
mixture
6 1.4946
MuQ_vocals: raw=0.9615, z=1.0686
STARS_phone_hidden: raw=0.8212, z=1.9206
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0011_a_0082000_0090000
vocals
mixture
7 1.4846
MuQ_vocals: raw=0.9462, z=0.4660
STARS_phone_hidden: raw=0.9225, z=2.5031
Secretariat_-_Borderline
Secretariat_-_Borderline_0006_b_0057000_0067000
vocals
mixture
8 1.4659
MuQ_vocals: raw=0.9588, z=0.9617
STARS_phone_hidden: raw=0.8298, z=1.9701
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0019_a_0136000_0142000
vocals
mixture
9 1.4225
MuQ_vocals: raw=0.9500, z=0.6145
STARS_phone_hidden: raw=0.8751, z=2.2305
The_Mountaineering_Club_-_Mallory
The_Mountaineering_Club_-_Mallory_0030_b_0206000_0213000
vocals
mixture
10 1.4191
MuQ_vocals: raw=0.9505, z=0.6353
STARS_phone_hidden: raw=0.8703, z=2.2028
BKS_-_Bulldozer
BKS_-_Bulldozer_0023_a_0165000_0170000
vocals
mixture
11 1.4130
MuQ_vocals: raw=0.9477, z=0.5259
STARS_phone_hidden: raw=0.8872, z=2.3000
Ben_Carrigan_-_We_ll_Talk_About_It_All_Tonight
Ben_Carrigan_-_We_ll_Talk_About_It_All_Tonight_0011_c_0071000_0076000
vocals
mixture
12 1.4124
MuQ_vocals: raw=0.9495, z=0.5952
STARS_phone_hidden: raw=0.8749, z=2.2296
The_Mountaineering_Club_-_Mallory
The_Mountaineering_Club_-_Mallory_0028_b_0193000_0199000
vocals
mixture
13 1.4105
MuQ_vocals: raw=0.9497, z=0.6059
STARS_phone_hidden: raw=0.8724, z=2.2150
Girls_Under_Glass_-_We_Feel_Alright
Girls_Under_Glass_-_We_Feel_Alright_0013_b_0087000_0095000
vocals
mixture
14 1.4084
MuQ_vocals: raw=0.9643, z=1.1780
STARS_phone_hidden: raw=0.7722, z=1.6389
Al_James_-_Schoolboy_Facination
Al_James_-_Schoolboy_Facination_0003_a_0020000_0027000
vocals
mixture
15 1.4008
MuQ_vocals: raw=0.9629, z=1.1211
STARS_phone_hidden: raw=0.7795, z=1.6805
St_Vitus_-_Word_Gets_Around
St_Vitus_-_Word_Gets_Around_0008_a_0048000_0055000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 2.3679
MuQ_vocals: raw=0.9592, z=0.9765
ECAPA: raw=0.7103, z=5.3209
STARS_phone_hidden: raw=0.6275, z=0.8062
Secretariat_-_Borderline
Secretariat_-_Borderline_0014_a_0118000_0128000
vocals
mixture
2 2.2487
MuQ_vocals: raw=0.9528, z=0.7268
ECAPA: raw=0.6476, z=4.7255
STARS_phone_hidden: raw=0.7123, z=1.2939
Secretariat_-_Borderline
Secretariat_-_Borderline_0005_a_0049000_0057000
vocals
mixture
3 2.2101
MuQ_vocals: raw=0.9656, z=1.2295
ECAPA: raw=0.6734, z=4.9711
STARS_phone_hidden: raw=0.5620, z=0.4295
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0005_a_0034000_0041000
vocals
mixture
4 2.0882
MuQ_vocals: raw=0.9700, z=1.4017
ECAPA: raw=0.6478, z=4.7275
STARS_phone_hidden: raw=0.5108, z=0.1352
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0004_a_0028000_0034000
vocals
mixture
5 2.0668
MuQ_vocals: raw=0.9473, z=0.5094
ECAPA: raw=0.5944, z=4.2209
STARS_phone_hidden: raw=0.7429, z=1.4701
Secretariat_-_Borderline
Secretariat_-_Borderline_0009_a_0085000_0090000
vocals
mixture
6 2.0539
MuQ_vocals: raw=0.9801, z=1.7965
ECAPA: raw=0.5376, z=3.6815
STARS_phone_hidden: raw=0.6062, z=0.6837
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0031_a_0228000_0234000
vocals
mixture
7 2.0464
MuQ_vocals: raw=0.9764, z=1.6513
ECAPA: raw=0.5803, z=4.0874
STARS_phone_hidden: raw=0.5569, z=0.4003
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0012_a_0083000_0090000
vocals
mixture
8 2.0299
MuQ_vocals: raw=0.9686, z=1.3460
ECAPA: raw=0.6795, z=5.0286
STARS_phone_hidden: raw=0.4378, z=-0.2849
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0006_a_0041000_0048000
vocals
mixture
9 1.9818
MuQ_vocals: raw=0.9729, z=1.5142
ECAPA: raw=0.5745, z=4.0319
STARS_phone_hidden: raw=0.5567, z=0.3993
The_Scarlet_Brand_-_Les_Fleurs_Du_Mal
The_Scarlet_Brand_-_Les_Fleurs_Du_Mal_0014_a_0089000_0095000
vocals
mixture
10 1.9583
MuQ_vocals: raw=0.9431, z=0.3436
ECAPA: raw=0.6642, z=4.8835
STARS_phone_hidden: raw=0.6000, z=0.6479
Secretariat_-_Borderline
Secretariat_-_Borderline_0016_a_0138000_0147000
vocals
mixture
11 1.8936
MuQ_vocals: raw=0.9588, z=0.9617
ECAPA: raw=0.4394, z=2.7491
STARS_phone_hidden: raw=0.8298, z=1.9701
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0019_a_0136000_0142000
vocals
mixture
12 1.8893
MuQ_vocals: raw=0.9585, z=0.9512
ECAPA: raw=0.5678, z=3.9680
STARS_phone_hidden: raw=0.6175, z=0.7488
Secretariat_-_Borderline
Secretariat_-_Borderline_0010_b_0090000_0098000
vocals
mixture
13 1.8408
MuQ_vocals: raw=0.9488, z=0.5681
ECAPA: raw=0.5718, z=4.0060
STARS_phone_hidden: raw=0.6522, z=0.9483
Secretariat_-_Borderline
Secretariat_-_Borderline_0029_b_0250000_0260000
vocals
mixture
14 1.8370
MuQ_vocals: raw=0.9642, z=1.1734
ECAPA: raw=0.5159, z=3.4757
STARS_phone_hidden: raw=0.6372, z=0.8621
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0004_a_0028000_0035000
vocals
mixture
15 1.8121
MuQ_vocals: raw=0.9697, z=1.3903
ECAPA: raw=0.5125, z=3.4435
STARS_phone_hidden: raw=0.5921, z=0.6026
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0020_a_0146000_0154000
vocals
mixture

Query 3: Port_St_Willow_-_Stay_Even / Port_St_Willow_-_Stay_Even_0015_a_0115000_0121000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9693
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0018_a_0146000_0155000
vocals
mixture
2 0.9693
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0017_a_0135000_0146000
vocals
mixture
3 0.9691
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0002_a_0016000_0023000
vocals
mixture
4 0.9684
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0001_a_0010000_0016000
vocals
mixture
5 0.9672
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0016_a_0121000_0132000
vocals
mixture
6 0.9657
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0014_a_0110000_0115000
vocals
mixture
7 0.9589
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0003_a_0023000_0033000
vocals
mixture
8 0.9581
James_May_-_On_The_Line
James_May_-_On_The_Line_0017_a_0112000_0119000
vocals
mixture
9 0.9574
Carlos_Gonzalez_-_A_Place_For_Us
Carlos_Gonzalez_-_A_Place_For_Us_0013_b_0087000_0095000
vocals
mixture
10 0.9564
Voelund_-_Comfort_Lives_In_Belief
Voelund_-_Comfort_Lives_In_Belief_0004_b_0034000_0041000
vocals
mixture
11 0.9560
Auctioneer_-_Our_Future_Faces
Auctioneer_-_Our_Future_Faces_0006_a_0051000_0059000
vocals
mixture
12 0.9557
The_Easton_Ellises_-_Falcon_69
The_Easton_Ellises_-_Falcon_69_0004_a_0033000_0038000
vocals
mixture
13 0.9557
Carlos_Gonzalez_-_A_Place_For_Us
Carlos_Gonzalez_-_A_Place_For_Us_0011_b_0076000_0084000
vocals
mixture
14 0.9554
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0005_a_0044000_0048000
vocals
mixture
15 0.9554
James_May_-_On_The_Line
James_May_-_On_The_Line_0024_a_0158000_0164000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9784
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0001_a_0010000_0016000
vocals
mixture
2 0.9762
Patrick_Talbot_-_A_Reason_To_Leave
Patrick_Talbot_-_A_Reason_To_Leave_0007_a_0045000_0051000
vocals
mixture
3 0.9722
Dreamers_Of_The_Ghetto_-_Heavy_Love
Dreamers_Of_The_Ghetto_-_Heavy_Love_0011_a_0085000_0092000
vocals
mixture
4 0.9712
The_Long_Wait_-_Dark_Horses
The_Long_Wait_-_Dark_Horses_0011_b_0089000_0097000
vocals
mixture
5 0.9705
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0012_b_0076000_0083000
vocals
mixture
6 0.9704
Patrick_Talbot_-_A_Reason_To_Leave
Patrick_Talbot_-_A_Reason_To_Leave_0006_a_0039000_0045000
vocals
mixture
7 0.9704
Carlos_Gonzalez_-_A_Place_For_Us
Carlos_Gonzalez_-_A_Place_For_Us_0003_a_0024000_0029000
vocals
mixture
8 0.9704
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0007_b_0044000_0051000
vocals
mixture
9 0.9702
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0018_b_0120000_0127000
vocals
mixture
10 0.9701
AvaLuna_-_Waterduct
AvaLuna_-_Waterduct_0008_a_0054000_0059000
vocals
mixture
11 0.9697
Mu_-_Too_Bright
Mu_-_Too_Bright_0018_a_0140000_0146000
vocals
mixture
12 0.9697
Cristina_Vane_-_So_Easy
Cristina_Vane_-_So_Easy_0020_a_0131000_0136000
vocals
mixture
13 0.9695
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0004_a_0036000_0042000
vocals
mixture
14 0.9695
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0003_a_0026000_0031000
vocals
mixture
15 0.9695
James_May_-_On_The_Line
James_May_-_On_The_Line_0024_a_0158000_0164000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 2.7391
MuQ_vocals: raw=0.9613, z=1.0078
ECAPA: raw=0.7319, z=4.4705
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0002_a_0016000_0023000
vocals
mixture
2 2.7172
MuQ_vocals: raw=0.9577, z=0.8845
ECAPA: raw=0.7412, z=4.5499
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0020_a_0160000_0167000
vocals
mixture
3 2.6149
MuQ_vocals: raw=0.9630, z=1.0669
ECAPA: raw=0.6956, z=4.1629
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0023_a_0178000_0186000
vocals
mixture
4 2.5626
MuQ_vocals: raw=0.9475, z=0.5351
ECAPA: raw=0.7459, z=4.5901
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0016_a_0121000_0132000
vocals
mixture
5 2.4898
MuQ_vocals: raw=0.9500, z=0.6209
ECAPA: raw=0.7187, z=4.3587
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0025_a_0195000_0205000
vocals
mixture
6 2.4758
MuQ_vocals: raw=0.9784, z=1.5937
ECAPA: raw=0.6009, z=3.3579
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0001_a_0010000_0016000
vocals
mixture
7 2.3929
MuQ_vocals: raw=0.9576, z=0.8809
ECAPA: raw=0.6653, z=3.9049
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0018_a_0146000_0155000
vocals
mixture
8 2.3428
MuQ_vocals: raw=0.9466, z=0.5040
ECAPA: raw=0.6978, z=4.1816
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0003_a_0023000_0033000
vocals
mixture
9 2.0757
MuQ_vocals: raw=0.9401, z=0.2829
ECAPA: raw=0.6610, z=3.8684
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0017_a_0135000_0146000
vocals
mixture
10 2.0608
MuQ_vocals: raw=0.9558, z=0.8209
ECAPA: raw=0.5942, z=3.3006
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0021_a_0166000_0170000
vocals
mixture
11 1.9754
MuQ_vocals: raw=0.9556, z=0.8118
ECAPA: raw=0.5751, z=3.1389
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0014_a_0110000_0115000
vocals
mixture
12 1.8720
MuQ_vocals: raw=0.9624, z=1.0444
ECAPA: raw=0.5234, z=2.6996
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0005_a_0044000_0048000
vocals
mixture
13 1.8672
MuQ_vocals: raw=0.9673, z=1.2140
ECAPA: raw=0.5023, z=2.5205
Dreamers_Of_The_Ghetto_-_Heavy_Love
Dreamers_Of_The_Ghetto_-_Heavy_Love_0008_a_0062000_0069000
vocals
mixture
14 1.6589
MuQ_vocals: raw=0.9576, z=0.8804
ECAPA: raw=0.4926, z=2.4374
Strand_Of_Oaks_-_Spacestation
Strand_Of_Oaks_-_Spacestation_0004_a_0032000_0042000
vocals
mixture
15 1.6387
MuQ_vocals: raw=0.9353, z=0.1169
ECAPA: raw=0.5777, z=3.1604
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0006_a_0048000_0058000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 1.7674
MuQ_vocals: raw=0.9667, z=1.1928
STARS_phone_hidden: raw=0.8419, z=2.3420
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0028_a_0210000_0216000
vocals
mixture
2 1.7470
MuQ_vocals: raw=0.9630, z=1.0669
STARS_phone_hidden: raw=0.8508, z=2.4271
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0023_a_0178000_0186000
vocals
mixture
3 1.6426
MuQ_vocals: raw=0.9784, z=1.5937
STARS_phone_hidden: raw=0.7745, z=1.6915
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0001_a_0010000_0016000
vocals
mixture
4 1.5740
MuQ_vocals: raw=0.9624, z=1.0469
STARS_phone_hidden: raw=0.8170, z=2.1011
Remember_December_-_C_U_Next_Time
Remember_December_-_C_U_Next_Time_0027_a_0179000_0185000
vocals
mixture
5 1.5610
MuQ_vocals: raw=0.9539, z=0.7556
STARS_phone_hidden: raw=0.8445, z=2.3664
The_Long_Wait_-_Back_Home_To_Blue
The_Long_Wait_-_Back_Home_To_Blue_0032_a_0231000_0237000
vocals
mixture
6 1.5423
MuQ_vocals: raw=0.9663, z=1.1812
STARS_phone_hidden: raw=0.7964, z=1.9033
St_Vitus_-_Word_Gets_Around
St_Vitus_-_Word_Gets_Around_0023_a_0174000_0180000
vocals
mixture
7 1.4704
MuQ_vocals: raw=0.9613, z=1.0078
STARS_phone_hidden: raw=0.7995, z=1.9331
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0002_a_0016000_0023000
vocals
mixture
8 1.4570
MuQ_vocals: raw=0.9659, z=1.1660
STARS_phone_hidden: raw=0.7803, z=1.7481
James_May_-_On_The_Line
James_May_-_On_The_Line_0016_a_0106000_0112000
vocals
mixture
9 1.4310
MuQ_vocals: raw=0.9691, z=1.2769
STARS_phone_hidden: raw=0.7634, z=1.5851
James_May_-_On_The_Line
James_May_-_On_The_Line_0001_a_0007000_0013000
vocals
mixture
10 1.4103
MuQ_vocals: raw=0.9695, z=1.2879
STARS_phone_hidden: raw=0.7580, z=1.5328
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0006_a_0044000_0050000
vocals
mixture
11 1.3939
MuQ_vocals: raw=0.9722, z=1.3811
STARS_phone_hidden: raw=0.7449, z=1.4067
Dreamers_Of_The_Ghetto_-_Heavy_Love
Dreamers_Of_The_Ghetto_-_Heavy_Love_0011_a_0085000_0092000
vocals
mixture
12 1.3915
MuQ_vocals: raw=0.9675, z=1.2213
STARS_phone_hidden: raw=0.7610, z=1.5617
Mu_-_Too_Bright
Mu_-_Too_Bright_0012_a_0107000_0114000
vocals
mixture
13 1.3803
MuQ_vocals: raw=0.9516, z=0.6768
STARS_phone_hidden: raw=0.8152, z=2.0838
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0012_b_0082000_0090000
vocals
mixture
14 1.3783
MuQ_vocals: raw=0.9695, z=1.2904
STARS_phone_hidden: raw=0.7511, z=1.4663
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0003_a_0026000_0031000
vocals
mixture
15 1.3649
MuQ_vocals: raw=0.9667, z=1.1936
STARS_phone_hidden: raw=0.7583, z=1.5361
Actions_-_Devil_s_Words
Actions_-_Devil_s_Words_0011_a_0072000_0078000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 2.5523
MuQ_vocals: raw=0.9630, z=1.0669
ECAPA: raw=0.6956, z=4.1629
STARS_phone_hidden: raw=0.8508, z=2.4271
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0023_a_0178000_0186000
vocals
mixture
2 2.4704
MuQ_vocals: raw=0.9613, z=1.0078
ECAPA: raw=0.7319, z=4.4705
STARS_phone_hidden: raw=0.7995, z=1.9331
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0002_a_0016000_0023000
vocals
mixture
3 2.4088
MuQ_vocals: raw=0.9577, z=0.8845
ECAPA: raw=0.7412, z=4.5499
STARS_phone_hidden: raw=0.7849, z=1.7919
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0020_a_0160000_0167000
vocals
mixture
4 2.2207
MuQ_vocals: raw=0.9475, z=0.5351
ECAPA: raw=0.7459, z=4.5901
STARS_phone_hidden: raw=0.7584, z=1.5370
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0016_a_0121000_0132000
vocals
mixture
5 2.2144
MuQ_vocals: raw=0.9784, z=1.5937
ECAPA: raw=0.6009, z=3.3579
STARS_phone_hidden: raw=0.7745, z=1.6915
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0001_a_0010000_0016000
vocals
mixture
6 2.0410
MuQ_vocals: raw=0.9466, z=0.5040
ECAPA: raw=0.6978, z=4.1816
STARS_phone_hidden: raw=0.7481, z=1.4374
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0003_a_0023000_0033000
vocals
mixture
7 1.9372
MuQ_vocals: raw=0.9500, z=0.6209
ECAPA: raw=0.7187, z=4.3587
STARS_phone_hidden: raw=0.6853, z=0.8320
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0025_a_0195000_0205000
vocals
mixture
8 1.8765
MuQ_vocals: raw=0.9558, z=0.8209
ECAPA: raw=0.5942, z=3.3006
STARS_phone_hidden: raw=0.7554, z=1.5081
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0021_a_0166000_0170000
vocals
mixture
9 1.8761
MuQ_vocals: raw=0.9576, z=0.8809
ECAPA: raw=0.6653, z=3.9049
STARS_phone_hidden: raw=0.6864, z=0.8424
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0018_a_0146000_0155000
vocals
mixture
10 1.7292
MuQ_vocals: raw=0.9353, z=0.1169
ECAPA: raw=0.5777, z=3.1604
STARS_phone_hidden: raw=0.7972, z=1.9102
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0006_a_0048000_0058000
vocals
mixture
11 1.6668
MuQ_vocals: raw=0.9401, z=0.2829
ECAPA: raw=0.6610, z=3.8684
STARS_phone_hidden: raw=0.6871, z=0.8491
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0017_a_0135000_0146000
vocals
mixture
12 1.5262
MuQ_vocals: raw=0.9722, z=1.3811
ECAPA: raw=0.4164, z=1.7907
STARS_phone_hidden: raw=0.7449, z=1.4067
Dreamers_Of_The_Ghetto_-_Heavy_Love
Dreamers_Of_The_Ghetto_-_Heavy_Love_0011_a_0085000_0092000
vocals
mixture
13 1.4609
MuQ_vocals: raw=0.9667, z=1.1928
ECAPA: raw=0.3055, z=0.8479
STARS_phone_hidden: raw=0.8419, z=2.3420
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0028_a_0210000_0216000
vocals
mixture
14 1.4379
MuQ_vocals: raw=0.9516, z=0.6768
ECAPA: raw=0.3885, z=1.5532
STARS_phone_hidden: raw=0.8152, z=2.0838
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0012_b_0082000_0090000
vocals
mixture
15 1.4086
MuQ_vocals: raw=0.9576, z=0.8804
ECAPA: raw=0.4926, z=2.4374
STARS_phone_hidden: raw=0.6932, z=0.9079
Strand_Of_Oaks_-_Spacestation
Strand_Of_Oaks_-_Spacestation_0004_a_0032000_0042000
vocals
mixture

Query 4: James_May_-_Dont_Let_Go / James_May_-_Dont_Let_Go_0019_a_0125000_0133000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9832
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0021_a_0142000_0151000
vocals
mixture
2 0.9816
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0004_a_0026000_0033000
vocals
mixture
3 0.9806
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0006_a_0039000_0047000
vocals
mixture
4 0.9798
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0020_a_0133000_0142000
vocals
mixture
5 0.9784
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0010_a_0066000_0074000
vocals
mixture
6 0.9759
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0008_a_0052000_0061000
vocals
mixture
7 0.9726
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0029_a_0201000_0209000
vocals
mixture
8 0.9714
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0031_a_0213000_0220000
vocals
mixture
9 0.9706
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0015_a_0097000_0105000
vocals
mixture
10 0.9701
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0027_a_0185000_0193000
vocals
mixture
11 0.9699
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0005_a_0033000_0039000
vocals
mixture
12 0.9695
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0017_a_0112000_0121000
vocals
mixture
13 0.9675
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0026_a_0184000_0190000
vocals
mixture
14 0.9674
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0013_a_0086000_0093000
vocals
mixture
15 0.9664
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0027_a_0190000_0196000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9839
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0004_a_0026000_0033000
vocals
mixture
2 0.9747
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0010_a_0066000_0074000
vocals
mixture
3 0.9746
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0006_a_0039000_0047000
vocals
mixture
4 0.9741
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0031_a_0213000_0220000
vocals
mixture
5 0.9736
James_May_-_If_You_Say
James_May_-_If_You_Say_0008_a_0061000_0071000
vocals
mixture
6 0.9733
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0004_a_0030000_0040000
vocals
mixture
7 0.9732
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0021_a_0142000_0151000
vocals
mixture
8 0.9731
James_May_-_If_You_Say
James_May_-_If_You_Say_0026_a_0186000_0195000
vocals
mixture
9 0.9726
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0029_a_0200000_0208000
vocals
mixture
10 0.9722
James_May_-_If_You_Say
James_May_-_If_You_Say_0016_a_0111000_0120000
vocals
mixture
11 0.9719
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0027_a_0185000_0193000
vocals
mixture
12 0.9716
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0008_a_0052000_0061000
vocals
mixture
13 0.9706
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0028_a_0192000_0200000
vocals
mixture
14 0.9702
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0011_a_0069000_0076000
vocals
mixture
15 0.9701
A_Classic_Education_-_NightOwl
A_Classic_Education_-_NightOwl_0020_a_0158000_0168000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 3.2345
MuQ_vocals: raw=0.9839, z=1.8930
ECAPA: raw=0.8326, z=4.5761
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0004_a_0026000_0033000
vocals
mixture
2 2.9076
MuQ_vocals: raw=0.9746, z=1.5881
ECAPA: raw=0.7831, z=4.2272
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0006_a_0039000_0047000
vocals
mixture
3 2.7689
MuQ_vocals: raw=0.9719, z=1.4982
ECAPA: raw=0.7565, z=4.0396
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0027_a_0185000_0193000
vocals
mixture
4 2.7259
MuQ_vocals: raw=0.9747, z=1.5902
ECAPA: raw=0.7312, z=3.8615
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0010_a_0066000_0074000
vocals
mixture
5 2.7100
MuQ_vocals: raw=0.9706, z=1.4551
ECAPA: raw=0.7459, z=3.9649
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0028_a_0192000_0200000
vocals
mixture
6 2.6762
MuQ_vocals: raw=0.9726, z=1.5212
ECAPA: raw=0.7269, z=3.8312
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0029_a_0200000_0208000
vocals
mixture
7 2.6565
MuQ_vocals: raw=0.9683, z=1.3827
ECAPA: raw=0.7410, z=3.9304
James_May_-_On_The_Line
James_May_-_On_The_Line_0011_b_0073000_0083000
vocals
mixture
8 2.6475
MuQ_vocals: raw=0.9694, z=1.4178
ECAPA: raw=0.7335, z=3.8772
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0009_a_0064000_0071000
vocals
mixture
9 2.6416
MuQ_vocals: raw=0.9632, z=1.2153
ECAPA: raw=0.7605, z=4.0680
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0007_a_0047000_0052000
vocals
mixture
10 2.5851
MuQ_vocals: raw=0.9697, z=1.4279
ECAPA: raw=0.7143, z=3.7424
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0026_a_0176000_0185000
vocals
mixture
11 2.5755
MuQ_vocals: raw=0.9716, z=1.4893
ECAPA: raw=0.7029, z=3.6617
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0008_a_0052000_0061000
vocals
mixture
12 2.5560
MuQ_vocals: raw=0.9732, z=1.5420
ECAPA: raw=0.6899, z=3.5700
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0021_a_0142000_0151000
vocals
mixture
13 2.5151
MuQ_vocals: raw=0.9653, z=1.2830
ECAPA: raw=0.7150, z=3.7472
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0017_a_0112000_0121000
vocals
mixture
14 2.4557
MuQ_vocals: raw=0.9686, z=1.3924
ECAPA: raw=0.6826, z=3.5190
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0016_a_0111000_0117000
vocals
mixture
15 2.4306
MuQ_vocals: raw=0.9700, z=1.4367
ECAPA: raw=0.6692, z=3.4245
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0020_a_0133000_0142000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 1.9041
MuQ_vocals: raw=0.9839, z=1.8930
STARS_phone_hidden: raw=0.9223, z=1.9152
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0004_a_0026000_0033000
vocals
mixture
2 1.6229
MuQ_vocals: raw=0.9746, z=1.5881
STARS_phone_hidden: raw=0.8874, z=1.6578
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0006_a_0039000_0047000
vocals
mixture
3 1.5394
MuQ_vocals: raw=0.9747, z=1.5902
STARS_phone_hidden: raw=0.8644, z=1.4885
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0010_a_0066000_0074000
vocals
mixture
4 1.5209
MuQ_vocals: raw=0.9736, z=1.5535
STARS_phone_hidden: raw=0.8644, z=1.4884
James_May_-_If_You_Say
James_May_-_If_You_Say_0008_a_0061000_0071000
vocals
mixture
5 1.5034
MuQ_vocals: raw=0.9716, z=1.4893
STARS_phone_hidden: raw=0.8684, z=1.5176
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0008_a_0052000_0061000
vocals
mixture
6 1.4575
MuQ_vocals: raw=0.9683, z=1.3827
STARS_phone_hidden: raw=0.8704, z=1.5324
James_May_-_On_The_Line
James_May_-_On_The_Line_0011_b_0073000_0083000
vocals
mixture
7 1.4258
MuQ_vocals: raw=0.9726, z=1.5212
STARS_phone_hidden: raw=0.8430, z=1.3303
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0029_a_0200000_0208000
vocals
mixture
8 1.3831
MuQ_vocals: raw=0.9700, z=1.4367
STARS_phone_hidden: raw=0.8429, z=1.3295
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0020_a_0133000_0142000
vocals
mixture
9 1.3819
MuQ_vocals: raw=0.9691, z=1.4088
STARS_phone_hidden: raw=0.8463, z=1.3550
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0015_a_0097000_0105000
vocals
mixture
10 1.3585
MuQ_vocals: raw=0.9731, z=1.5374
STARS_phone_hidden: raw=0.8226, z=1.1795
James_May_-_If_You_Say
James_May_-_If_You_Say_0026_a_0186000_0195000
vocals
mixture
11 1.3436
MuQ_vocals: raw=0.9602, z=1.1159
STARS_phone_hidden: raw=0.8757, z=1.5713
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0003_a_0029000_0036000
vocals
mixture
12 1.3366
MuQ_vocals: raw=0.9701, z=1.4410
STARS_phone_hidden: raw=0.8297, z=1.2321
A_Classic_Education_-_NightOwl
A_Classic_Education_-_NightOwl_0020_a_0158000_0168000
vocals
mixture
13 1.3178
MuQ_vocals: raw=0.9694, z=1.4178
STARS_phone_hidden: raw=0.8277, z=1.2177
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0009_a_0064000_0071000
vocals
mixture
14 1.3099
MuQ_vocals: raw=0.9686, z=1.3924
STARS_phone_hidden: raw=0.8290, z=1.2273
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0016_a_0111000_0117000
vocals
mixture
15 1.3036
MuQ_vocals: raw=0.9732, z=1.5420
STARS_phone_hidden: raw=0.8071, z=1.0652
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0021_a_0142000_0151000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 2.7947
MuQ_vocals: raw=0.9839, z=1.8930
ECAPA: raw=0.8326, z=4.5761
STARS_phone_hidden: raw=0.9223, z=1.9152
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0004_a_0026000_0033000
vocals
mixture
2 2.4910
MuQ_vocals: raw=0.9746, z=1.5881
ECAPA: raw=0.7831, z=4.2272
STARS_phone_hidden: raw=0.8874, z=1.6578
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0006_a_0039000_0047000
vocals
mixture
3 2.3134
MuQ_vocals: raw=0.9747, z=1.5902
ECAPA: raw=0.7312, z=3.8615
STARS_phone_hidden: raw=0.8644, z=1.4885
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0010_a_0066000_0074000
vocals
mixture
4 2.2818
MuQ_vocals: raw=0.9683, z=1.3827
ECAPA: raw=0.7410, z=3.9304
STARS_phone_hidden: raw=0.8704, z=1.5324
James_May_-_On_The_Line
James_May_-_On_The_Line_0011_b_0073000_0083000
vocals
mixture
5 2.2276
MuQ_vocals: raw=0.9726, z=1.5212
ECAPA: raw=0.7269, z=3.8312
STARS_phone_hidden: raw=0.8430, z=1.3303
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0029_a_0200000_0208000
vocals
mixture
6 2.2229
MuQ_vocals: raw=0.9716, z=1.4893
ECAPA: raw=0.7029, z=3.6617
STARS_phone_hidden: raw=0.8684, z=1.5176
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0008_a_0052000_0061000
vocals
mixture
7 2.2147
MuQ_vocals: raw=0.9719, z=1.4982
ECAPA: raw=0.7565, z=4.0396
STARS_phone_hidden: raw=0.8127, z=1.1064
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0027_a_0185000_0193000
vocals
mixture
8 2.1871
MuQ_vocals: raw=0.9632, z=1.2153
ECAPA: raw=0.7605, z=4.0680
STARS_phone_hidden: raw=0.8359, z=1.2779
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0007_a_0047000_0052000
vocals
mixture
9 2.1822
MuQ_vocals: raw=0.9706, z=1.4551
ECAPA: raw=0.7459, z=3.9649
STARS_phone_hidden: raw=0.8154, z=1.1265
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0028_a_0192000_0200000
vocals
mixture
10 2.1709
MuQ_vocals: raw=0.9694, z=1.4178
ECAPA: raw=0.7335, z=3.8772
STARS_phone_hidden: raw=0.8277, z=1.2177
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0009_a_0064000_0071000
vocals
mixture
11 2.0983
MuQ_vocals: raw=0.9697, z=1.4279
ECAPA: raw=0.7143, z=3.7424
STARS_phone_hidden: raw=0.8151, z=1.1246
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0026_a_0176000_0185000
vocals
mixture
12 2.0640
MuQ_vocals: raw=0.9602, z=1.1159
ECAPA: raw=0.6806, z=3.5047
STARS_phone_hidden: raw=0.8757, z=1.5713
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0003_a_0029000_0036000
vocals
mixture
13 2.0636
MuQ_vocals: raw=0.9700, z=1.4367
ECAPA: raw=0.6692, z=3.4245
STARS_phone_hidden: raw=0.8429, z=1.3295
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0020_a_0133000_0142000
vocals
mixture
14 2.0591
MuQ_vocals: raw=0.9732, z=1.5420
ECAPA: raw=0.6899, z=3.5700
STARS_phone_hidden: raw=0.8071, z=1.0652
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0021_a_0142000_0151000
vocals
mixture
15 2.0463
MuQ_vocals: raw=0.9686, z=1.3924
ECAPA: raw=0.6826, z=3.5190
STARS_phone_hidden: raw=0.8290, z=1.2273
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0016_a_0111000_0117000
vocals
mixture

Query 5: James_May_-_All_Souls_Moon / James_May_-_All_Souls_Moon_0013_b_0090000_0099000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9763
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0007_b_0053000_0061000
vocals
mixture
2 0.9691
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0010_a_0066000_0074000
vocals
mixture
3 0.9685
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0027_a_0185000_0193000
vocals
mixture
4 0.9685
The_Long_Wait_-_Dark_Horses
The_Long_Wait_-_Dark_Horses_0007_b_0056000_0065000
vocals
mixture
5 0.9673
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0028_a_0192000_0200000
vocals
mixture
6 0.9662
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0006_a_0039000_0047000
vocals
mixture
7 0.9659
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0026_a_0176000_0185000
vocals
mixture
8 0.9650
Secretariat_-_Borderline
Secretariat_-_Borderline_0014_a_0118000_0128000
vocals
mixture
9 0.9648
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0020_a_0133000_0142000
vocals
mixture
10 0.9643
The_Long_Wait_-_Dark_Horses
The_Long_Wait_-_Dark_Horses_0017_b_0135000_0145000
vocals
mixture
11 0.9640
The_Long_Wait_-_Dark_Horses
The_Long_Wait_-_Dark_Horses_0012_b_0097000_0105000
vocals
mixture
12 0.9640
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0019_a_0125000_0133000
vocals
mixture
13 0.9638
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0008_a_0052000_0061000
vocals
mixture
14 0.9630
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0010_a_0071000_0078000
vocals
mixture
15 0.9630
The_Long_Wait_-_Back_Home_To_Blue
The_Long_Wait_-_Back_Home_To_Blue_0013_c_0086000_0097000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9768
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0007_b_0053000_0061000
vocals
mixture
2 0.9647
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0015_a_0097000_0105000
vocals
mixture
3 0.9637
BKS_-_Too_Much
BKS_-_Too_Much_0013_b_0087000_0096000
vocals
mixture
4 0.9631
BKS_-_Too_Much
BKS_-_Too_Much_0024_b_0160000_0168000
vocals
mixture
5 0.9628
Jay_Menon_-_Through_My_Eyes
Jay_Menon_-_Through_My_Eyes_0013_a_0082000_0090000
vocals
mixture
6 0.9611
BKS_-_Too_Much
BKS_-_Too_Much_0027_b_0180000_0188000
vocals
mixture
7 0.9601
Jay_Menon_-_Through_My_Eyes
Jay_Menon_-_Through_My_Eyes_0003_a_0015000_0022000
vocals
mixture
8 0.9601
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0031_a_0213000_0220000
vocals
mixture
9 0.9595
BKS_-_Too_Much
BKS_-_Too_Much_0002_a_0020000_0028000
vocals
mixture
10 0.9585
AM_Contra_-_Heart_Peripheral
AM_Contra_-_Heart_Peripheral_0004_b_0030000_0038000
vocals
mixture
11 0.9584
Jay_Menon_-_Through_My_Eyes
Jay_Menon_-_Through_My_Eyes_0009_a_0053000_0060000
vocals
mixture
12 0.9584
BKS_-_Too_Much
BKS_-_Too_Much_0007_b_0051000_0059000
vocals
mixture
13 0.9581
BKS_-_Too_Much
BKS_-_Too_Much_0011_a_0076000_0083000
vocals
mixture
14 0.9578
Jay_Menon_-_Through_My_Eyes
Jay_Menon_-_Through_My_Eyes_0018_a_0120000_0127000
vocals
mixture
15 0.9576
BKS_-_Too_Much
BKS_-_Too_Much_0004_b_0032000_0040000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 2.8957
MuQ_vocals: raw=0.9572, z=1.5449
ECAPA: raw=0.7811, z=4.2464
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0026_a_0176000_0185000
vocals
mixture
2 2.8284
MuQ_vocals: raw=0.9647, z=1.8525
ECAPA: raw=0.7209, z=3.8043
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0015_a_0097000_0105000
vocals
mixture
3 2.7062
MuQ_vocals: raw=0.9601, z=1.6648
ECAPA: raw=0.7131, z=3.7475
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0031_a_0213000_0220000
vocals
mixture
4 2.6519
MuQ_vocals: raw=0.9509, z=1.2872
ECAPA: raw=0.7498, z=4.0167
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0027_a_0185000_0193000
vocals
mixture
5 2.6117
MuQ_vocals: raw=0.9502, z=1.2582
ECAPA: raw=0.7428, z=3.9651
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0009_a_0064000_0071000
vocals
mixture
6 2.6041
MuQ_vocals: raw=0.9516, z=1.3178
ECAPA: raw=0.7326, z=3.8905
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0010_a_0071000_0078000
vocals
mixture
7 2.5618
MuQ_vocals: raw=0.9568, z=1.5303
ECAPA: raw=0.6921, z=3.5934
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0020_a_0133000_0142000
vocals
mixture
8 2.5446
MuQ_vocals: raw=0.9553, z=1.4696
ECAPA: raw=0.6957, z=3.6197
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0029_a_0200000_0208000
vocals
mixture
9 2.5166
MuQ_vocals: raw=0.9569, z=1.5336
ECAPA: raw=0.6794, z=3.4996
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0028_a_0192000_0200000
vocals
mixture
10 2.4856
MuQ_vocals: raw=0.9489, z=1.2077
ECAPA: raw=0.7153, z=3.7635
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0021_a_0142000_0151000
vocals
mixture
11 2.4074
MuQ_vocals: raw=0.9475, z=1.1490
ECAPA: raw=0.7020, z=3.6657
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0019_a_0125000_0133000
vocals
mixture
12 2.4072
MuQ_vocals: raw=0.9471, z=1.1343
ECAPA: raw=0.7040, z=3.6801
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0025_a_0170000_0176000
vocals
mixture
13 2.3931
MuQ_vocals: raw=0.9484, z=1.1870
ECAPA: raw=0.6929, z=3.5993
James_May_-_On_The_Line
James_May_-_On_The_Line_0011_b_0073000_0083000
vocals
mixture
14 2.3668
MuQ_vocals: raw=0.9543, z=1.4282
ECAPA: raw=0.6529, z=3.3053
James_May_-_On_The_Line
James_May_-_On_The_Line_0028_b_0183000_0190000
vocals
mixture
15 2.3565
MuQ_vocals: raw=0.9540, z=1.4158
ECAPA: raw=0.6518, z=3.2971
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0008_a_0052000_0061000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 1.6514
MuQ_vocals: raw=0.9568, z=1.5305
STARS_phone_hidden: raw=0.8627, z=1.7722
Alexander_Ross_-_Velvet_Curtain
Alexander_Ross_-_Velvet_Curtain_0004_a_0030000_0040000
vocals
mixture
2 1.5834
MuQ_vocals: raw=0.9628, z=1.7739
STARS_phone_hidden: raw=0.8146, z=1.3929
Jay_Menon_-_Through_My_Eyes
Jay_Menon_-_Through_My_Eyes_0013_a_0082000_0090000
vocals
mixture
3 1.5411
MuQ_vocals: raw=0.9561, z=1.5003
STARS_phone_hidden: raw=0.8386, z=1.5819
Jay_Menon_-_Through_My_Eyes
Jay_Menon_-_Through_My_Eyes_0002_a_0008000_0015000
vocals
mixture
4 1.5383
MuQ_vocals: raw=0.9647, z=1.8525
STARS_phone_hidden: raw=0.7933, z=1.2241
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0015_a_0097000_0105000
vocals
mixture
5 1.4705
MuQ_vocals: raw=0.9584, z=1.5921
STARS_phone_hidden: raw=0.8091, z=1.3489
BKS_-_Too_Much
BKS_-_Too_Much_0007_b_0051000_0059000
vocals
mixture
6 1.4674
MuQ_vocals: raw=0.9471, z=1.1343
STARS_phone_hidden: raw=0.8663, z=1.8006
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0025_a_0170000_0176000
vocals
mixture
7 1.4625
MuQ_vocals: raw=0.9611, z=1.7040
STARS_phone_hidden: raw=0.7929, z=1.2211
BKS_-_Too_Much
BKS_-_Too_Much_0027_b_0180000_0188000
vocals
mixture
8 1.4513
MuQ_vocals: raw=0.9485, z=1.1916
STARS_phone_hidden: raw=0.8549, z=1.7111
Flags_-_54
Flags_-_54_0017_a_0140000_0149000
vocals
mixture
9 1.4025
MuQ_vocals: raw=0.9483, z=1.1835
STARS_phone_hidden: raw=0.8436, z=1.6216
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0006_b_0048000_0053000
vocals
mixture
10 1.3900
MuQ_vocals: raw=0.9503, z=1.2661
STARS_phone_hidden: raw=0.8300, z=1.5140
Strand_Of_Oaks_-_Spacestation
Strand_Of_Oaks_-_Spacestation_0025_b_0185000_0194000
vocals
mixture
11 1.3869
MuQ_vocals: raw=0.9473, z=1.1417
STARS_phone_hidden: raw=0.8449, z=1.6320
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0021_b_0146000_0156000
vocals
mixture
12 1.3744
MuQ_vocals: raw=0.9601, z=1.6648
STARS_phone_hidden: raw=0.7755, z=1.0840
Jay_Menon_-_Through_My_Eyes
Jay_Menon_-_Through_My_Eyes_0003_a_0015000_0022000
vocals
mixture
13 1.3673
MuQ_vocals: raw=0.9510, z=1.2917
STARS_phone_hidden: raw=0.8210, z=1.4428
Patrick_Talbot_-_A_Reason_To_Leave
Patrick_Talbot_-_A_Reason_To_Leave_0004_a_0029000_0036000
vocals
mixture
14 1.3638
MuQ_vocals: raw=0.9504, z=1.2663
STARS_phone_hidden: raw=0.8233, z=1.4613
Patrick_Talbot_-_A_Reason_To_Leave
Patrick_Talbot_-_A_Reason_To_Leave_0008_b_0051000_0060000
vocals
mixture
15 1.3546
MuQ_vocals: raw=0.9472, z=1.1391
STARS_phone_hidden: raw=0.8371, z=1.5700
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0016_a_0111000_0117000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 2.2936
MuQ_vocals: raw=0.9647, z=1.8525
ECAPA: raw=0.7209, z=3.8043
STARS_phone_hidden: raw=0.7933, z=1.2241
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0015_a_0097000_0105000
vocals
mixture
2 2.2050
MuQ_vocals: raw=0.9471, z=1.1343
ECAPA: raw=0.7040, z=3.6801
STARS_phone_hidden: raw=0.8663, z=1.8006
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0025_a_0170000_0176000
vocals
mixture
3 2.2025
MuQ_vocals: raw=0.9572, z=1.5449
ECAPA: raw=0.7811, z=4.2464
STARS_phone_hidden: raw=0.7416, z=0.8161
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0026_a_0176000_0185000
vocals
mixture
4 2.1912
MuQ_vocals: raw=0.9509, z=1.2872
ECAPA: raw=0.7498, z=4.0167
STARS_phone_hidden: raw=0.7991, z=1.2699
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0027_a_0185000_0193000
vocals
mixture
5 2.1734
MuQ_vocals: raw=0.9516, z=1.3178
ECAPA: raw=0.7326, z=3.8905
STARS_phone_hidden: raw=0.8044, z=1.3118
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0010_a_0071000_0078000
vocals
mixture
6 2.1067
MuQ_vocals: raw=0.9483, z=1.1835
ECAPA: raw=0.6814, z=3.5149
STARS_phone_hidden: raw=0.8436, z=1.6216
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0006_b_0048000_0053000
vocals
mixture
7 2.0707
MuQ_vocals: raw=0.9553, z=1.4696
ECAPA: raw=0.6957, z=3.6197
STARS_phone_hidden: raw=0.7804, z=1.1227
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0029_a_0200000_0208000
vocals
mixture
8 2.0701
MuQ_vocals: raw=0.9568, z=1.5303
ECAPA: raw=0.6921, z=3.5934
STARS_phone_hidden: raw=0.7759, z=1.0867
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0020_a_0133000_0142000
vocals
mixture
9 2.0681
MuQ_vocals: raw=0.9489, z=1.2077
ECAPA: raw=0.7153, z=3.7635
STARS_phone_hidden: raw=0.7944, z=1.2330
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0021_a_0142000_0151000
vocals
mixture
10 2.0189
MuQ_vocals: raw=0.9502, z=1.2582
ECAPA: raw=0.7428, z=3.9651
STARS_phone_hidden: raw=0.7438, z=0.8335
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0009_a_0064000_0071000
vocals
mixture
11 1.9572
MuQ_vocals: raw=0.9434, z=0.9823
ECAPA: raw=0.6439, z=3.2395
STARS_phone_hidden: raw=0.8472, z=1.6498
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0002_a_0021000_0029000
vocals
mixture
12 1.9452
MuQ_vocals: raw=0.9461, z=1.0945
ECAPA: raw=0.6501, z=3.2851
STARS_phone_hidden: raw=0.8226, z=1.4560
James_May_-_On_The_Line
James_May_-_On_The_Line_0007_a_0046000_0053000
vocals
mixture
13 1.9371
MuQ_vocals: raw=0.9449, z=1.0461
ECAPA: raw=0.6845, z=3.5376
STARS_phone_hidden: raw=0.7937, z=1.2274
James_May_-_If_You_Say
James_May_-_If_You_Say_0021_a_0149000_0158000
vocals
mixture
14 1.9335
MuQ_vocals: raw=0.9475, z=1.1490
ECAPA: raw=0.7020, z=3.6657
STARS_phone_hidden: raw=0.7631, z=0.9859
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0019_a_0125000_0133000
vocals
mixture
15 1.9036
MuQ_vocals: raw=0.9482, z=1.1779
ECAPA: raw=0.6555, z=3.3247
STARS_phone_hidden: raw=0.7913, z=1.2083
James_May_-_Dont_Let_Go
James_May_-_Dont_Let_Go_0010_a_0066000_0074000
vocals
mixture

Query 6: Triviul_-_Angelsaint / Triviul_-_Angelsaint_0008_a_0071000_0080000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9524
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0007_c_0061000_0071000
vocals
mixture
2 0.9497
James_Elder_Mark_M_Thompson_-_The_English_Actor
James_Elder_Mark_M_Thompson_-_The_English_Actor_0007_c_0047000_0056000
vocals
mixture
3 0.9487
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0014_a_0107000_0116000
vocals
mixture
4 0.9481
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0022_c_0209000_0220000
vocals
mixture
5 0.9479
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0015_c_0116000_0125000
vocals
mixture
6 0.9478
James_Elder_Mark_M_Thompson_-_The_English_Actor
James_Elder_Mark_M_Thompson_-_The_English_Actor_0013_b_0092000_0101000
vocals
mixture
7 0.9461
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0009_a_0080000_0086000
vocals
mixture
8 0.9455
Speak_Softly_-_Broken_Man
Speak_Softly_-_Broken_Man_0017_a_0145000_0155000
vocals
mixture
9 0.9449
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0005_a_0038000_0047000
vocals
mixture
10 0.9446
James_May_-_All_Souls_Moon
James_May_-_All_Souls_Moon_0027_a_0185000_0193000
vocals
mixture
11 0.9445
Secret_Mountains_-_High_Horse
Secret_Mountains_-_High_Horse_0020_a_0162000_0170000
vocals
mixture
12 0.9444
Mu_-_Too_Bright
Mu_-_Too_Bright_0030_b_0225000_0235000
vocals
mixture
13 0.9441
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0013_a_0097000_0106000
vocals
mixture
14 0.9435
The_Easton_Ellises_-_Falcon_69
The_Easton_Ellises_-_Falcon_69_0023_a_0160000_0169000
vocals
mixture
15 0.9433
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0006_c_0045000_0053000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9503
Louis_Cressy_Band_-_Good_Time
Louis_Cressy_Band_-_Good_Time_0014_a_0101000_0111000
vocals
mixture
2 0.9479
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0013_a_0097000_0104000
vocals
mixture
3 0.9468
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0010_b_0064000_0074000
vocals
mixture
4 0.9448
Speak_Softly_-_Like_Horses
Speak_Softly_-_Like_Horses_0021_b_0155000_0166000
vocals
mixture
5 0.9446
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0022_b_0163000_0173000
vocals
mixture
6 0.9446
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0004_a_0031000_0039000
vocals
mixture
7 0.9441
Dark_Ride_-_Burning_Bridges
Dark_Ride_-_Burning_Bridges_0008_a_0074000_0084000
vocals
mixture
8 0.9435
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0006_a_0044000_0052000
vocals
mixture
9 0.9435
Flags_-_54
Flags_-_54_0011_a_0097000_0107000
vocals
mixture
10 0.9427
The_Easton_Ellises_-_Falcon_69
The_Easton_Ellises_-_Falcon_69_0015_c_0101000_0112000
vocals
mixture
11 0.9426
Hop_Along_-_Sister_Cities
Hop_Along_-_Sister_Cities_0006_a_0052000_0061000
vocals
mixture
12 0.9423
Hop_Along_-_Sister_Cities
Hop_Along_-_Sister_Cities_0005_a_0042000_0052000
vocals
mixture
13 0.9422
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0022_c_0209000_0220000
vocals
mixture
14 0.9417
AM_Contra_-_Heart_Peripheral
AM_Contra_-_Heart_Peripheral_0025_a_0142000_0152000
vocals
mixture
15 0.9403
James_May_-_If_You_Say
James_May_-_If_You_Say_0026_a_0186000_0195000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 2.3512
MuQ_vocals: raw=0.9335, z=1.4961
ECAPA: raw=0.7012, z=3.2063
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0009_a_0080000_0086000
vocals
mixture
2 2.2962
MuQ_vocals: raw=0.9468, z=2.0478
ECAPA: raw=0.6127, z=2.5445
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0010_b_0064000_0074000
vocals
mixture
3 2.1970
MuQ_vocals: raw=0.9317, z=1.4208
ECAPA: raw=0.6700, z=2.9731
Traffic_Experiment_-_Once_More_With_Feeling
Traffic_Experiment_-_Once_More_With_Feeling_0016_c_0150000_0159000
vocals
mixture
4 2.1703
MuQ_vocals: raw=0.9296, z=1.3334
ECAPA: raw=0.6746, z=3.0071
Traffic_Experiment_-_Once_More_With_Feeling
Traffic_Experiment_-_Once_More_With_Feeling_0007_c_0068000_0077000
vocals
mixture
5 2.1113
MuQ_vocals: raw=0.9282, z=1.2728
ECAPA: raw=0.6669, z=2.9499
Secretariat_-_Borderline
Secretariat_-_Borderline_0007_b_0069000_0079000
vocals
mixture
6 2.0807
MuQ_vocals: raw=0.9281, z=1.2696
ECAPA: raw=0.6592, z=2.8918
Secretariat_-_Borderline
Secretariat_-_Borderline_0027_b_0231000_0241000
vocals
mixture
7 2.0479
MuQ_vocals: raw=0.9392, z=1.7335
ECAPA: raw=0.5883, z=2.3623
Mu_-_Too_Bright
Mu_-_Too_Bright_0019_b_0146000_0156000
vocals
mixture
8 2.0288
MuQ_vocals: raw=0.9388, z=1.7147
ECAPA: raw=0.5857, z=2.3429
Mu_-_Too_Bright
Mu_-_Too_Bright_0006_b_0057000_0067000
vocals
mixture
9 1.9961
MuQ_vocals: raw=0.9479, z=2.0956
ECAPA: raw=0.5260, z=1.8966
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0013_a_0097000_0104000
vocals
mixture
10 1.9939
MuQ_vocals: raw=0.9357, z=1.5885
ECAPA: raw=0.5932, z=2.3992
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0017_a_0135000_0146000
vocals
mixture
11 1.9816
MuQ_vocals: raw=0.9324, z=1.4509
ECAPA: raw=0.6084, z=2.5123
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0013_b_0090000_0100000
vocals
mixture
12 1.9755
MuQ_vocals: raw=0.9435, z=1.9125
ECAPA: raw=0.5450, z=2.0385
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0006_a_0044000_0052000
vocals
mixture
13 1.9632
MuQ_vocals: raw=0.9370, z=1.6427
ECAPA: raw=0.5778, z=2.2836
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0004_a_0030000_0038000
vocals
mixture
14 1.9625
MuQ_vocals: raw=0.9378, z=1.6757
ECAPA: raw=0.5732, z=2.2494
Mu_-_Too_Bright
Mu_-_Too_Bright_0030_b_0225000_0235000
vocals
mixture
15 1.9495
MuQ_vocals: raw=0.9360, z=1.5979
ECAPA: raw=0.5801, z=2.3010
Mu_-_Too_Bright
Mu_-_Too_Bright_0031_b_0235000_0245000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 1.8803
MuQ_vocals: raw=0.9384, z=1.6982
STARS_phone_hidden: raw=0.7332, z=2.0624
Johnny_Lokke_-_Whisper_To_A_Scream
Johnny_Lokke_-_Whisper_To_A_Scream_0025_b_0179000_0187000
vocals
mixture
2 1.7132
MuQ_vocals: raw=0.9293, z=1.3214
STARS_phone_hidden: raw=0.7376, z=2.1050
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0018_a_0146000_0155000
vocals
mixture
3 1.6365
MuQ_vocals: raw=0.9435, z=1.9125
STARS_phone_hidden: raw=0.6605, z=1.3606
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0006_a_0044000_0052000
vocals
mixture
4 1.6166
MuQ_vocals: raw=0.9342, z=1.5240
STARS_phone_hidden: raw=0.6966, z=1.7092
Creepoid_-_OldTree
Creepoid_-_OldTree_0011_a_0093000_0101000
vocals
mixture
5 1.6155
MuQ_vocals: raw=0.9355, z=1.5793
STARS_phone_hidden: raw=0.6906, z=1.6518
The_Doppler_Shift_-_Atrophy
The_Doppler_Shift_-_Atrophy_0035_a_0261000_0271000
vocals
mixture
6 1.6137
MuQ_vocals: raw=0.9249, z=1.1367
STARS_phone_hidden: raw=0.7362, z=2.0908
Chris_Durban_-_Celebrate
Chris_Durban_-_Celebrate_0012_a_0120000_0128000
vocals
mixture
7 1.6059
MuQ_vocals: raw=0.9403, z=1.7794
STARS_phone_hidden: raw=0.6679, z=1.4323
James_May_-_If_You_Say
James_May_-_If_You_Say_0026_a_0186000_0195000
vocals
mixture
8 1.6048
MuQ_vocals: raw=0.9294, z=1.3230
STARS_phone_hidden: raw=0.7150, z=1.8866
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0014_a_0107000_0116000
vocals
mixture
9 1.5619
MuQ_vocals: raw=0.9503, z=2.1961
STARS_phone_hidden: raw=0.6156, z=0.9278
Louis_Cressy_Band_-_Good_Time
Louis_Cressy_Band_-_Good_Time_0014_a_0101000_0111000
vocals
mixture
10 1.5602
MuQ_vocals: raw=0.9357, z=1.5865
STARS_phone_hidden: raw=0.6784, z=1.5339
Remember_December_-_C_U_Next_Time
Remember_December_-_C_U_Next_Time_0010_b_0065000_0074000
vocals
mixture
11 1.5387
MuQ_vocals: raw=0.9319, z=1.4276
STARS_phone_hidden: raw=0.6904, z=1.6499
Clara_Berry_And_Wooldog_-_Air_Traffic
Clara_Berry_And_Wooldog_-_Air_Traffic_0009_c_0078000_0088000
vocals
mixture
12 1.5382
MuQ_vocals: raw=0.9446, z=1.9578
STARS_phone_hidden: raw=0.6354, z=1.1186
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0022_b_0163000_0173000
vocals
mixture
13 1.5149
MuQ_vocals: raw=0.9340, z=1.5177
STARS_phone_hidden: raw=0.6762, z=1.5122
Remember_December_-_C_U_Next_Time
Remember_December_-_C_U_Next_Time_0026_a_0169000_0179000
vocals
mixture
14 1.5122
MuQ_vocals: raw=0.9240, z=1.0980
STARS_phone_hidden: raw=0.7191, z=1.9263
The_Wrong_Uns_-_Rothko
The_Wrong_Uns_-_Rothko_0010_a_0078000_0085000
vocals
mixture
15 1.4942
MuQ_vocals: raw=0.9345, z=1.5357
STARS_phone_hidden: raw=0.6700, z=1.4528
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0020_b_0150000_0158000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 1.8329
MuQ_vocals: raw=0.9294, z=1.3230
ECAPA: raw=0.5785, z=2.2890
STARS_phone_hidden: raw=0.7150, z=1.8866
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0014_a_0107000_0116000
vocals
mixture
2 1.7724
MuQ_vocals: raw=0.9317, z=1.4208
ECAPA: raw=0.6700, z=2.9731
STARS_phone_hidden: raw=0.6151, z=0.9234
Traffic_Experiment_-_Once_More_With_Feeling
Traffic_Experiment_-_Once_More_With_Feeling_0016_c_0150000_0159000
vocals
mixture
3 1.7705
MuQ_vocals: raw=0.9435, z=1.9125
ECAPA: raw=0.5450, z=2.0385
STARS_phone_hidden: raw=0.6605, z=1.3606
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0006_a_0044000_0052000
vocals
mixture
4 1.7472
MuQ_vocals: raw=0.9355, z=1.5793
ECAPA: raw=0.5412, z=2.0105
STARS_phone_hidden: raw=0.6906, z=1.6518
The_Doppler_Shift_-_Atrophy
The_Doppler_Shift_-_Atrophy_0035_a_0261000_0271000
vocals
mixture
5 1.7321
MuQ_vocals: raw=0.9392, z=1.7335
ECAPA: raw=0.5883, z=2.3623
STARS_phone_hidden: raw=0.6335, z=1.1005
Mu_-_Too_Bright
Mu_-_Too_Bright_0019_b_0146000_0156000
vocals
mixture
6 1.7307
MuQ_vocals: raw=0.9281, z=1.2696
ECAPA: raw=0.6592, z=2.8918
STARS_phone_hidden: raw=0.6263, z=1.0308
Secretariat_-_Borderline
Secretariat_-_Borderline_0027_b_0231000_0241000
vocals
mixture
7 1.7159
MuQ_vocals: raw=0.9345, z=1.5357
ECAPA: raw=0.5611, z=2.1592
STARS_phone_hidden: raw=0.6700, z=1.4528
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0020_b_0150000_0158000
vocals
mixture
8 1.7089
MuQ_vocals: raw=0.9468, z=2.0478
ECAPA: raw=0.6127, z=2.5445
STARS_phone_hidden: raw=0.5748, z=0.5344
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0010_b_0064000_0074000
vocals
mixture
9 1.6503
MuQ_vocals: raw=0.9446, z=1.9578
ECAPA: raw=0.5230, z=1.8746
STARS_phone_hidden: raw=0.6354, z=1.1186
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0022_b_0163000_0173000
vocals
mixture
10 1.6389
MuQ_vocals: raw=0.9293, z=1.3214
ECAPA: raw=0.4716, z=1.4903
STARS_phone_hidden: raw=0.7376, z=2.1050
Port_St_Willow_-_Stay_Even
Port_St_Willow_-_Stay_Even_0018_a_0146000_0155000
vocals
mixture
11 1.6354
MuQ_vocals: raw=0.9324, z=1.4509
ECAPA: raw=0.6084, z=2.5123
STARS_phone_hidden: raw=0.6172, z=0.9432
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0013_b_0090000_0100000
vocals
mixture
12 1.6202
MuQ_vocals: raw=0.9195, z=0.9098
ECAPA: raw=0.6704, z=2.9756
STARS_phone_hidden: raw=0.6205, z=0.9752
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0011_a_0093000_0099000
vocals
mixture
13 1.6173
MuQ_vocals: raw=0.9479, z=2.0956
ECAPA: raw=0.5260, z=1.8966
STARS_phone_hidden: raw=0.6085, z=0.8598
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0013_a_0097000_0104000
vocals
mixture
14 1.6171
MuQ_vocals: raw=0.9296, z=1.3334
ECAPA: raw=0.6746, z=3.0071
STARS_phone_hidden: raw=0.5724, z=0.5108
Traffic_Experiment_-_Once_More_With_Feeling
Traffic_Experiment_-_Once_More_With_Feeling_0007_c_0068000_0077000
vocals
mixture
15 1.6156
MuQ_vocals: raw=0.9335, z=1.4961
ECAPA: raw=0.7012, z=3.2063
STARS_phone_hidden: raw=0.5344, z=0.1443
Triviul_-_Angelsaint
Triviul_-_Angelsaint_0009_a_0080000_0086000
vocals
mixture

Query 7: Forkupines_-_Semantics / Forkupines_-_Semantics_0009_b_0062000_0069000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9910
Forkupines_-_Semantics
Forkupines_-_Semantics_0025_b_0167000_0174000
vocals
mixture
2 0.9892
Forkupines_-_Semantics
Forkupines_-_Semantics_0023_b_0155000_0161000
vocals
mixture
3 0.9835
Forkupines_-_Semantics
Forkupines_-_Semantics_0024_b_0161000_0167000
vocals
mixture
4 0.9724
Forkupines_-_Semantics
Forkupines_-_Semantics_0008_b_0055000_0062000
vocals
mixture
5 0.9619
Forkupines_-_Semantics
Forkupines_-_Semantics_0022_b_0151000_0155000
vocals
mixture
6 0.9540
Jokers_Jacks_Kings_-_Sea_Of_Leaves
Jokers_Jacks_Kings_-_Sea_Of_Leaves_0010_b_0068000_0076000
vocals
mixture
7 0.9536
North_To_Alaska_-_All_The_Same
North_To_Alaska_-_All_The_Same_0029_b_0207000_0214000
vocals
mixture
8 0.9516
Forkupines_-_Semantics
Forkupines_-_Semantics_0012_b_0083000_0090000
vocals
mixture
9 0.9514
The_Easton_Ellises_Baumi_-_SDRNR
The_Easton_Ellises_Baumi_-_SDRNR_0016_c_0128000_0137000
vocals
mixture
10 0.9512
Jokers_Jacks_Kings_-_Sea_Of_Leaves
Jokers_Jacks_Kings_-_Sea_Of_Leaves_0012_b_0081000_0091000
vocals
mixture
11 0.9509
BKS_-_Bulldozer
BKS_-_Bulldozer_0043_c_0296000_0307000
vocals
mixture
12 0.9508
The_Easton_Ellises_Baumi_-_SDRNR
The_Easton_Ellises_Baumi_-_SDRNR_0011_c_0087000_0096000
vocals
mixture
13 0.9507
BKS_-_Bulldozer
BKS_-_Bulldozer_0020_c_0147000_0156000
vocals
mixture
14 0.9503
Jokers_Jacks_Kings_-_Sea_Of_Leaves
Jokers_Jacks_Kings_-_Sea_Of_Leaves_0023_b_0165000_0174000
vocals
mixture
15 0.9503
Atlantis_Bound_-_It_Was_My_Fault_For_Waiting
Atlantis_Bound_-_It_Was_My_Fault_For_Waiting_0030_a_0195000_0203000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9918
Forkupines_-_Semantics
Forkupines_-_Semantics_0025_b_0167000_0174000
vocals
mixture
2 0.9828
Forkupines_-_Semantics
Forkupines_-_Semantics_0023_b_0155000_0161000
vocals
mixture
3 0.9725
Forkupines_-_Semantics
Forkupines_-_Semantics_0024_b_0161000_0167000
vocals
mixture
4 0.9699
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0018_b_0127000_0134000
vocals
mixture
5 0.9670
Actions_-_Devil_s_Words
Actions_-_Devil_s_Words_0006_b_0046000_0053000
vocals
mixture
6 0.9644
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0012_b_0085000_0091000
vocals
mixture
7 0.9630
Forkupines_-_Semantics
Forkupines_-_Semantics_0008_b_0055000_0062000
vocals
mixture
8 0.9629
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0011_b_0079000_0085000
vocals
mixture
9 0.9624
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0005_a_0035000_0042000
vocals
mixture
10 0.9606
AM_Contra_-_Heart_Peripheral
AM_Contra_-_Heart_Peripheral_0003_a_0023000_0030000
vocals
mixture
11 0.9596
Wall_Of_Death_-_Femme
Wall_Of_Death_-_Femme_0017_a_0122000_0129000
vocals
mixture
12 0.9595
Johnny_Lokke_-_Whisper_To_A_Scream
Johnny_Lokke_-_Whisper_To_A_Scream_0004_b_0027000_0034000
vocals
mixture
13 0.9586
Actions_-_Devil_s_Words
Actions_-_Devil_s_Words_0023_a_0142000_0149000
vocals
mixture
14 0.9583
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0024_a_0151000_0158000
vocals
mixture
15 0.9582
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0008_a_0052000_0058000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 3.3430
MuQ_vocals: raw=0.9918, z=3.2203
ECAPA: raw=0.7729, z=3.4658
Forkupines_-_Semantics
Forkupines_-_Semantics_0025_b_0167000_0174000
vocals
mixture
2 3.1870
MuQ_vocals: raw=0.9828, z=2.7984
ECAPA: raw=0.7895, z=3.5755
Forkupines_-_Semantics
Forkupines_-_Semantics_0023_b_0155000_0161000
vocals
mixture
3 2.3792
MuQ_vocals: raw=0.9725, z=2.3148
ECAPA: raw=0.6183, z=2.4436
Forkupines_-_Semantics
Forkupines_-_Semantics_0024_b_0161000_0167000
vocals
mixture
4 2.3298
MuQ_vocals: raw=0.9630, z=1.8711
ECAPA: raw=0.6705, z=2.7885
Forkupines_-_Semantics
Forkupines_-_Semantics_0008_b_0055000_0062000
vocals
mixture
5 2.1484
MuQ_vocals: raw=0.9461, z=1.0839
ECAPA: raw=0.7347, z=3.2129
Forkupines_-_Semantics
Forkupines_-_Semantics_0038_a_0263000_0269000
vocals
mixture
6 1.9185
MuQ_vocals: raw=0.9574, z=1.6111
ECAPA: raw=0.5854, z=2.2259
Motor_Tapes_-_Shore
Motor_Tapes_-_Shore_0015_b_0132000_0139000
vocals
mixture
7 1.9177
MuQ_vocals: raw=0.9397, z=0.7845
ECAPA: raw=0.7102, z=3.0509
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0035_a_0256000_0261000
vocals
mixture
8 1.9104
MuQ_vocals: raw=0.9549, z=1.4927
ECAPA: raw=0.6009, z=2.3281
Forkupines_-_Semantics
Forkupines_-_Semantics_0022_b_0151000_0155000
vocals
mixture
9 1.8724
MuQ_vocals: raw=0.9579, z=1.6317
ECAPA: raw=0.5684, z=2.1132
Wall_Of_Death_-_Femme
Wall_Of_Death_-_Femme_0005_a_0037000_0044000
vocals
mixture
10 1.8210
MuQ_vocals: raw=0.9596, z=1.7135
ECAPA: raw=0.5405, z=1.9285
Wall_Of_Death_-_Femme
Wall_Of_Death_-_Femme_0017_a_0122000_0129000
vocals
mixture
11 1.8203
MuQ_vocals: raw=0.9494, z=1.2344
ECAPA: raw=0.6127, z=2.4063
Forkupines_-_Semantics
Forkupines_-_Semantics_0016_b_0111000_0116000
vocals
mixture
12 1.8150
MuQ_vocals: raw=0.9498, z=1.2536
ECAPA: raw=0.6082, z=2.3764
We_Fell_From_The_Sky_-_Not_You
We_Fell_From_The_Sky_-_Not_You_0026_c_0186000_0191000
vocals
mixture
13 1.7981
MuQ_vocals: raw=0.9516, z=1.3404
ECAPA: raw=0.5900, z=2.2559
Johnny_Lokke_-_Whisper_To_A_Scream
Johnny_Lokke_-_Whisper_To_A_Scream_0031_b_0218000_0225000
vocals
mixture
14 1.7971
MuQ_vocals: raw=0.9547, z=1.4830
ECAPA: raw=0.5681, z=2.1113
Forkupines_-_Semantics
Forkupines_-_Semantics_0018_b_0122000_0128000
vocals
mixture
15 1.7923
MuQ_vocals: raw=0.9582, z=1.6468
ECAPA: raw=0.5419, z=1.9379
Wall_Of_Death_-_Femme
Wall_Of_Death_-_Femme_0012_a_0085000_0092000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 2.3935
MuQ_vocals: raw=0.9918, z=3.2203
STARS_phone_hidden: raw=0.7814, z=1.5667
Forkupines_-_Semantics
Forkupines_-_Semantics_0025_b_0167000_0174000
vocals
mixture
2 2.3269
MuQ_vocals: raw=0.9828, z=2.7984
STARS_phone_hidden: raw=0.8130, z=1.8554
Forkupines_-_Semantics
Forkupines_-_Semantics_0023_b_0155000_0161000
vocals
mixture
3 1.8305
MuQ_vocals: raw=0.9699, z=2.1964
STARS_phone_hidden: raw=0.7703, z=1.4646
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0018_b_0127000_0134000
vocals
mixture
4 1.8151
MuQ_vocals: raw=0.9491, z=1.2210
STARS_phone_hidden: raw=0.8736, z=2.4093
Louis_Cressy_Band_-_Good_Time
Louis_Cressy_Band_-_Good_Time_0009_a_0066000_0072000
vocals
mixture
5 1.7857
MuQ_vocals: raw=0.9562, z=1.5533
STARS_phone_hidden: raw=0.8308, z=2.0181
Georgia_Wonder_-_Siren
Georgia_Wonder_-_Siren_0008_a_0055000_0062000
vocals
mixture
6 1.7620
MuQ_vocals: raw=0.9606, z=1.7593
STARS_phone_hidden: raw=0.8031, z=1.7646
AM_Contra_-_Heart_Peripheral
AM_Contra_-_Heart_Peripheral_0003_a_0023000_0030000
vocals
mixture
7 1.7374
MuQ_vocals: raw=0.9644, z=1.9391
STARS_phone_hidden: raw=0.7780, z=1.5356
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0012_b_0085000_0091000
vocals
mixture
8 1.7362
MuQ_vocals: raw=0.9496, z=1.2446
STARS_phone_hidden: raw=0.8538, z=2.2279
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0030_b_0250000_0254000
vocals
mixture
9 1.7275
MuQ_vocals: raw=0.9725, z=2.3148
STARS_phone_hidden: raw=0.7348, z=1.1402
Forkupines_-_Semantics
Forkupines_-_Semantics_0024_b_0161000_0167000
vocals
mixture
10 1.7098
MuQ_vocals: raw=0.9409, z=0.8387
STARS_phone_hidden: raw=0.8924, z=2.5809
Girls_Under_Glass_-_We_Feel_Alright
Girls_Under_Glass_-_We_Feel_Alright_0016_b_0106000_0111000
vocals
mixture
11 1.7021
MuQ_vocals: raw=0.9511, z=1.3153
STARS_phone_hidden: raw=0.8386, z=2.0889
Bill_Chudziak_-_Children_Of_No-one
Bill_Chudziak_-_Children_Of_No-one_0023_b_0163000_0171000
vocals
mixture
12 1.6985
MuQ_vocals: raw=0.9514, z=1.3286
STARS_phone_hidden: raw=0.8363, z=2.0685
Louis_Cressy_Band_-_Good_Time
Louis_Cressy_Band_-_Good_Time_0010_a_0072000_0079000
vocals
mixture
13 1.6616
MuQ_vocals: raw=0.9423, z=0.9042
STARS_phone_hidden: raw=0.8747, z=2.4190
Grants_-_PunchDrunk
Grants_-_PunchDrunk_0027_b_0180000_0186000
vocals
mixture
14 1.6453
MuQ_vocals: raw=0.9477, z=1.1582
STARS_phone_hidden: raw=0.8434, z=2.1325
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0005_b_0032000_0038000
vocals
mixture
15 1.6328
MuQ_vocals: raw=0.9457, z=1.0630
STARS_phone_hidden: raw=0.8510, z=2.2026
Grants_-_PunchDrunk
Grants_-_PunchDrunk_0004_b_0029000_0036000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 2.7509
MuQ_vocals: raw=0.9918, z=3.2203
ECAPA: raw=0.7729, z=3.4658
STARS_phone_hidden: raw=0.7814, z=1.5667
Forkupines_-_Semantics
Forkupines_-_Semantics_0025_b_0167000_0174000
vocals
mixture
2 2.7431
MuQ_vocals: raw=0.9828, z=2.7984
ECAPA: raw=0.7895, z=3.5755
STARS_phone_hidden: raw=0.8130, z=1.8554
Forkupines_-_Semantics
Forkupines_-_Semantics_0023_b_0155000_0161000
vocals
mixture
3 1.9662
MuQ_vocals: raw=0.9725, z=2.3148
ECAPA: raw=0.6183, z=2.4436
STARS_phone_hidden: raw=0.7348, z=1.1402
Forkupines_-_Semantics
Forkupines_-_Semantics_0024_b_0161000_0167000
vocals
mixture
4 1.9455
MuQ_vocals: raw=0.9630, z=1.8711
ECAPA: raw=0.6705, z=2.7885
STARS_phone_hidden: raw=0.7388, z=1.1769
Forkupines_-_Semantics
Forkupines_-_Semantics_0008_b_0055000_0062000
vocals
mixture
5 1.8673
MuQ_vocals: raw=0.9358, z=0.5980
ECAPA: raw=0.6133, z=2.4100
STARS_phone_hidden: raw=0.8938, z=2.5938
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0029_b_0238000_0248000
vocals
mixture
6 1.8606
MuQ_vocals: raw=0.9498, z=1.2536
ECAPA: raw=0.6082, z=2.3764
STARS_phone_hidden: raw=0.8236, z=1.9517
We_Fell_From_The_Sky_-_Not_You
We_Fell_From_The_Sky_-_Not_You_0026_c_0186000_0191000
vocals
mixture
7 1.8156
MuQ_vocals: raw=0.9326, z=0.4507
ECAPA: raw=0.6306, z=2.5247
STARS_phone_hidden: raw=0.8804, z=2.4713
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0025_b_0208000_0212000
vocals
mixture
8 1.8060
MuQ_vocals: raw=0.9496, z=1.2446
ECAPA: raw=0.5430, z=1.9456
STARS_phone_hidden: raw=0.8538, z=2.2279
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0030_b_0250000_0254000
vocals
mixture
9 1.7908
MuQ_vocals: raw=0.9511, z=1.3153
ECAPA: raw=0.5464, z=1.9681
STARS_phone_hidden: raw=0.8386, z=2.0889
Bill_Chudziak_-_Children_Of_No-one
Bill_Chudziak_-_Children_Of_No-one_0023_b_0163000_0171000
vocals
mixture
10 1.7429
MuQ_vocals: raw=0.9253, z=0.1085
ECAPA: raw=0.6915, z=2.9275
STARS_phone_hidden: raw=0.8499, z=2.1927
We_Fell_From_The_Sky_-_Not_You
We_Fell_From_The_Sky_-_Not_You_0028_a_0200000_0206000
vocals
mixture
11 1.7416
MuQ_vocals: raw=0.9549, z=1.4927
ECAPA: raw=0.6009, z=2.3281
STARS_phone_hidden: raw=0.7636, z=1.4040
Forkupines_-_Semantics
Forkupines_-_Semantics_0022_b_0151000_0155000
vocals
mixture
12 1.7400
MuQ_vocals: raw=0.9478, z=1.1596
ECAPA: raw=0.5809, z=2.1959
STARS_phone_hidden: raw=0.8140, z=1.8645
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0027_b_0217000_0224000
vocals
mixture
13 1.6784
MuQ_vocals: raw=0.9517, z=1.3460
ECAPA: raw=0.5727, z=2.1416
STARS_phone_hidden: raw=0.7794, z=1.5477
Meaxic_-_You_Listen
Meaxic_-_You_Listen_0012_b_0082000_0090000
vocals
mixture
14 1.6601
MuQ_vocals: raw=0.9341, z=0.5218
ECAPA: raw=0.6069, z=2.3678
STARS_phone_hidden: raw=0.8388, z=2.0905
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0023_b_0189000_0199000
vocals
mixture
15 1.6365
MuQ_vocals: raw=0.9366, z=0.6384
ECAPA: raw=0.5260, z=1.8332
STARS_phone_hidden: raw=0.8768, z=2.4380
The_Easton_Ellises_Baumi_-_SDRNR
The_Easton_Ellises_Baumi_-_SDRNR_0016_c_0128000_0137000
vocals
mixture

Query 8: BKS_-_Too_Much / BKS_-_Too_Much_0008_b_0059000_0066000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9869
BKS_-_Too_Much
BKS_-_Too_Much_0016_b_0106000_0114000
vocals
mixture
2 0.9867
BKS_-_Too_Much
BKS_-_Too_Much_0029_b_0191000_0200000
vocals
mixture
3 0.9776
BKS_-_Too_Much
BKS_-_Too_Much_0027_b_0180000_0188000
vocals
mixture
4 0.9759
BKS_-_Too_Much
BKS_-_Too_Much_0006_a_0044000_0051000
vocals
mixture
5 0.9724
BKS_-_Too_Much
BKS_-_Too_Much_0004_b_0032000_0040000
vocals
mixture
6 0.9717
Al_James_-_Schoolboy_Facination
Al_James_-_Schoolboy_Facination_0009_c_0061000_0069000
vocals
mixture
7 0.9703
Secretariat_-_Borderline
Secretariat_-_Borderline_0010_b_0090000_0098000
vocals
mixture
8 0.9699
Al_James_-_Schoolboy_Facination
Al_James_-_Schoolboy_Facination_0023_b_0154000_0161000
vocals
mixture
9 0.9696
Al_James_-_Schoolboy_Facination
Al_James_-_Schoolboy_Facination_0026_c_0175000_0183000
vocals
mixture
10 0.9686
Secretariat_-_Over_The_Top
Secretariat_-_Over_The_Top_0005_a_0034000_0041000
vocals
mixture
11 0.9686
The_So_So_Glos_-_Emergency
The_So_So_Glos_-_Emergency_0017_a_0114000_0121000
vocals
mixture
12 0.9684
BKS_-_Too_Much
BKS_-_Too_Much_0009_a_0066000_0072000
vocals
mixture
13 0.9683
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0017_b_0132000_0139000
vocals
mixture
14 0.9682
BKS_-_Too_Much
BKS_-_Too_Much_0014_a_0095000_0099000
vocals
mixture
15 0.9678
Leaf_-_Wicked
Leaf_-_Wicked_0013_c_0134000_0140000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9813
BKS_-_Too_Much
BKS_-_Too_Much_0016_b_0106000_0114000
vocals
mixture
2 0.9787
BKS_-_Too_Much
BKS_-_Too_Much_0029_b_0191000_0200000
vocals
mixture
3 0.9729
BKS_-_Too_Much
BKS_-_Too_Much_0026_b_0172000_0180000
vocals
mixture
4 0.9725
BKS_-_Too_Much
BKS_-_Too_Much_0004_b_0032000_0040000
vocals
mixture
5 0.9724
BKS_-_Too_Much
BKS_-_Too_Much_0025_b_0168000_0172000
vocals
mixture
6 0.9716
Fergessen_-_The_Wind
Fergessen_-_The_Wind_0013_b_0107000_0114000
vocals
mixture
7 0.9712
Bobby_Nobody_-_Stitch_Up
Bobby_Nobody_-_Stitch_Up_0016_b_0105000_0109000
vocals
mixture
8 0.9708
BKS_-_Too_Much
BKS_-_Too_Much_0012_b_0083000_0087000
vocals
mixture
9 0.9706
BKS_-_Too_Much
BKS_-_Too_Much_0027_b_0180000_0188000
vocals
mixture
10 0.9703
Bobby_Nobody_-_Stitch_Up
Bobby_Nobody_-_Stitch_Up_0027_b_0188000_0195000
vocals
mixture
11 0.9702
Bobby_Nobody_-_Stitch_Up
Bobby_Nobody_-_Stitch_Up_0026_b_0181000_0188000
vocals
mixture
12 0.9699
Bobby_Nobody_-_Stitch_Up
Bobby_Nobody_-_Stitch_Up_0017_c_0109000_0113000
vocals
mixture
13 0.9694
Steven_Clark_-_Bounty
Steven_Clark_-_Bounty_0009_b_0066000_0073000
vocals
mixture
14 0.9693
BKS_-_Too_Much
BKS_-_Too_Much_0003_a_0028000_0032000
vocals
mixture
15 0.9693
Steven_Clark_-_Bounty
Steven_Clark_-_Bounty_0030_b_0237000_0245000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 3.1950
MuQ_vocals: raw=0.9787, z=1.4965
ECAPA: raw=0.7733, z=4.8935
BKS_-_Too_Much
BKS_-_Too_Much_0029_b_0191000_0200000
vocals
mixture
2 3.1062
MuQ_vocals: raw=0.9813, z=1.5795
ECAPA: raw=0.7429, z=4.6328
BKS_-_Too_Much
BKS_-_Too_Much_0016_b_0106000_0114000
vocals
mixture
3 2.9325
MuQ_vocals: raw=0.9725, z=1.2992
ECAPA: raw=0.7350, z=4.5658
BKS_-_Too_Much
BKS_-_Too_Much_0004_b_0032000_0040000
vocals
mixture
4 2.8651
MuQ_vocals: raw=0.9729, z=1.3114
ECAPA: raw=0.7179, z=4.4189
BKS_-_Too_Much
BKS_-_Too_Much_0026_b_0172000_0180000
vocals
mixture
5 2.7554
MuQ_vocals: raw=0.9627, z=0.9843
ECAPA: raw=0.7305, z=4.5266
BKS_-_Too_Much
BKS_-_Too_Much_0011_a_0076000_0083000
vocals
mixture
6 2.7334
MuQ_vocals: raw=0.9640, z=1.0275
ECAPA: raw=0.7203, z=4.4392
BKS_-_Too_Much
BKS_-_Too_Much_0006_a_0044000_0051000
vocals
mixture
7 2.7274
MuQ_vocals: raw=0.9673, z=1.1302
ECAPA: raw=0.7069, z=4.3245
BKS_-_Too_Much
BKS_-_Too_Much_0007_b_0051000_0059000
vocals
mixture
8 2.7240
MuQ_vocals: raw=0.9693, z=1.1965
ECAPA: raw=0.6983, z=4.2515
BKS_-_Too_Much
BKS_-_Too_Much_0003_a_0028000_0032000
vocals
mixture
9 2.6717
MuQ_vocals: raw=0.9706, z=1.2364
ECAPA: raw=0.6815, z=4.1071
BKS_-_Too_Much
BKS_-_Too_Much_0027_b_0180000_0188000
vocals
mixture
10 2.5848
MuQ_vocals: raw=0.9646, z=1.0467
ECAPA: raw=0.6833, z=4.1229
BKS_-_Too_Much
BKS_-_Too_Much_0013_b_0087000_0096000
vocals
mixture
11 2.3933
MuQ_vocals: raw=0.9592, z=0.8737
ECAPA: raw=0.6588, z=3.9129
BKS_-_Too_Much
BKS_-_Too_Much_0014_a_0095000_0099000
vocals
mixture
12 2.3546
MuQ_vocals: raw=0.9659, z=1.0862
ECAPA: raw=0.6250, z=3.6229
BKS_-_Too_Much
BKS_-_Too_Much_0005_b_0040000_0044000
vocals
mixture
13 2.3294
MuQ_vocals: raw=0.9642, z=1.0335
ECAPA: raw=0.6253, z=3.6253
BKS_-_Too_Much
BKS_-_Too_Much_0023_a_0152000_0160000
vocals
mixture
14 2.3189
MuQ_vocals: raw=0.9708, z=1.2441
ECAPA: raw=0.5982, z=3.3936
BKS_-_Too_Much
BKS_-_Too_Much_0012_b_0083000_0087000
vocals
mixture
15 2.2984
MuQ_vocals: raw=0.9501, z=0.5815
ECAPA: raw=0.6708, z=4.0153
BKS_-_Bulldozer
BKS_-_Bulldozer_0008_a_0060000_0066000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 1.3928
MuQ_vocals: raw=0.9725, z=1.2992
STARS_phone_hidden: raw=0.8814, z=1.4864
BKS_-_Too_Much
BKS_-_Too_Much_0004_b_0032000_0040000
vocals
mixture
2 1.3614
MuQ_vocals: raw=0.9706, z=1.2364
STARS_phone_hidden: raw=0.8814, z=1.4863
BKS_-_Too_Much
BKS_-_Too_Much_0027_b_0180000_0188000
vocals
mixture
3 1.3600
MuQ_vocals: raw=0.9813, z=1.5795
STARS_phone_hidden: raw=0.8251, z=1.1404
BKS_-_Too_Much
BKS_-_Too_Much_0016_b_0106000_0114000
vocals
mixture
4 1.3138
MuQ_vocals: raw=0.9648, z=1.0527
STARS_phone_hidden: raw=0.8958, z=1.5748
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0015_c_0110000_0116000
vocals
mixture
5 1.2430
MuQ_vocals: raw=0.9684, z=1.1656
STARS_phone_hidden: raw=0.8544, z=1.3205
Bobby_Nobody_-_Stitch_Up
Bobby_Nobody_-_Stitch_Up_0012_b_0079000_0087000
vocals
mixture
6 1.2344
MuQ_vocals: raw=0.9787, z=1.4965
STARS_phone_hidden: raw=0.7977, z=0.9723
BKS_-_Too_Much
BKS_-_Too_Much_0029_b_0191000_0200000
vocals
mixture
7 1.2190
MuQ_vocals: raw=0.9702, z=1.2250
STARS_phone_hidden: raw=0.8369, z=1.2130
Bobby_Nobody_-_Stitch_Up
Bobby_Nobody_-_Stitch_Up_0026_b_0181000_0188000
vocals
mixture
8 1.2140
MuQ_vocals: raw=0.9619, z=0.9576
STARS_phone_hidden: raw=0.8788, z=1.4703
Cristina_Vane_-_So_Easy
Cristina_Vane_-_So_Easy_0018_a_0117000_0124000
vocals
mixture
9 1.2093
MuQ_vocals: raw=0.9588, z=0.8594
STARS_phone_hidden: raw=0.8933, z=1.5592
Dreamers_Of_The_Ghetto_-_Heavy_Love
Dreamers_Of_The_Ghetto_-_Heavy_Love_0018_c_0135000_0141000
vocals
mixture
10 1.1997
MuQ_vocals: raw=0.9729, z=1.3114
STARS_phone_hidden: raw=0.8166, z=1.0881
BKS_-_Too_Much
BKS_-_Too_Much_0026_b_0172000_0180000
vocals
mixture
11 1.1992
MuQ_vocals: raw=0.9703, z=1.2260
STARS_phone_hidden: raw=0.8303, z=1.1724
Bobby_Nobody_-_Stitch_Up
Bobby_Nobody_-_Stitch_Up_0027_b_0188000_0195000
vocals
mixture
12 1.1904
MuQ_vocals: raw=0.9682, z=1.1588
STARS_phone_hidden: raw=0.8384, z=1.2220
ANiMAL_-_Easy_Tiger
ANiMAL_-_Easy_Tiger_0012_b_0084000_0088000
vocals
mixture
13 1.1837
MuQ_vocals: raw=0.9673, z=1.1302
STARS_phone_hidden: raw=0.8408, z=1.2371
BKS_-_Too_Much
BKS_-_Too_Much_0007_b_0051000_0059000
vocals
mixture
14 1.1817
MuQ_vocals: raw=0.9673, z=1.1312
STARS_phone_hidden: raw=0.8400, z=1.2322
Bobby_Nobody_-_Stitch_Up
Bobby_Nobody_-_Stitch_Up_0024_a_0159000_0165000
vocals
mixture
15 1.1765
MuQ_vocals: raw=0.9612, z=0.9358
STARS_phone_hidden: raw=0.8701, z=1.4171
Bobby_Nobody_-_Stitch_Up
Bobby_Nobody_-_Stitch_Up_0022_b_0142000_0151000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 2.4541
MuQ_vocals: raw=0.9787, z=1.4965
ECAPA: raw=0.7733, z=4.8935
STARS_phone_hidden: raw=0.7977, z=0.9723
BKS_-_Too_Much
BKS_-_Too_Much_0029_b_0191000_0200000
vocals
mixture
2 2.4509
MuQ_vocals: raw=0.9813, z=1.5795
ECAPA: raw=0.7429, z=4.6328
STARS_phone_hidden: raw=0.8251, z=1.1404
BKS_-_Too_Much
BKS_-_Too_Much_0016_b_0106000_0114000
vocals
mixture
3 2.4504
MuQ_vocals: raw=0.9725, z=1.2992
ECAPA: raw=0.7350, z=4.5658
STARS_phone_hidden: raw=0.8814, z=1.4864
BKS_-_Too_Much
BKS_-_Too_Much_0004_b_0032000_0040000
vocals
mixture
4 2.2766
MuQ_vocals: raw=0.9706, z=1.2364
ECAPA: raw=0.6815, z=4.1071
STARS_phone_hidden: raw=0.8814, z=1.4863
BKS_-_Too_Much
BKS_-_Too_Much_0027_b_0180000_0188000
vocals
mixture
5 2.2728
MuQ_vocals: raw=0.9729, z=1.3114
ECAPA: raw=0.7179, z=4.4189
STARS_phone_hidden: raw=0.8166, z=1.0881
BKS_-_Too_Much
BKS_-_Too_Much_0026_b_0172000_0180000
vocals
mixture
6 2.2306
MuQ_vocals: raw=0.9673, z=1.1302
ECAPA: raw=0.7069, z=4.3245
STARS_phone_hidden: raw=0.8408, z=1.2371
BKS_-_Too_Much
BKS_-_Too_Much_0007_b_0051000_0059000
vocals
mixture
7 2.2048
MuQ_vocals: raw=0.9640, z=1.0275
ECAPA: raw=0.7203, z=4.4392
STARS_phone_hidden: raw=0.8263, z=1.1476
BKS_-_Too_Much
BKS_-_Too_Much_0006_a_0044000_0051000
vocals
mixture
8 2.1613
MuQ_vocals: raw=0.9627, z=0.9843
ECAPA: raw=0.7305, z=4.5266
STARS_phone_hidden: raw=0.7978, z=0.9729
BKS_-_Too_Much
BKS_-_Too_Much_0011_a_0076000_0083000
vocals
mixture
9 2.1469
MuQ_vocals: raw=0.9693, z=1.1965
ECAPA: raw=0.6983, z=4.2515
STARS_phone_hidden: raw=0.8011, z=0.9928
BKS_-_Too_Much
BKS_-_Too_Much_0003_a_0028000_0032000
vocals
mixture
10 2.1093
MuQ_vocals: raw=0.9646, z=1.0467
ECAPA: raw=0.6833, z=4.1229
STARS_phone_hidden: raw=0.8280, z=1.1584
BKS_-_Too_Much
BKS_-_Too_Much_0013_b_0087000_0096000
vocals
mixture
11 1.9683
MuQ_vocals: raw=0.9659, z=1.0862
ECAPA: raw=0.6250, z=3.6229
STARS_phone_hidden: raw=0.8341, z=1.1956
BKS_-_Too_Much
BKS_-_Too_Much_0005_b_0040000_0044000
vocals
mixture
12 1.9407
MuQ_vocals: raw=0.9592, z=0.8737
ECAPA: raw=0.6588, z=3.9129
STARS_phone_hidden: raw=0.8080, z=1.0356
BKS_-_Too_Much
BKS_-_Too_Much_0014_a_0095000_0099000
vocals
mixture
13 1.8829
MuQ_vocals: raw=0.9708, z=1.2441
ECAPA: raw=0.5982, z=3.3936
STARS_phone_hidden: raw=0.8040, z=1.0111
BKS_-_Too_Much
BKS_-_Too_Much_0012_b_0083000_0087000
vocals
mixture
14 1.8091
MuQ_vocals: raw=0.9501, z=0.5815
ECAPA: raw=0.6708, z=4.0153
STARS_phone_hidden: raw=0.7746, z=0.8305
BKS_-_Bulldozer
BKS_-_Bulldozer_0008_a_0060000_0066000
vocals
mixture
15 1.7846
MuQ_vocals: raw=0.9558, z=0.7655
ECAPA: raw=0.5781, z=3.2210
STARS_phone_hidden: raw=0.8620, z=1.3673
James_May_-_On_The_Line
James_May_-_On_The_Line_0007_a_0046000_0053000
vocals
mixture

Query 9: Punkdisco_-_Oral_Hygiene / Punkdisco_-_Oral_Hygiene_0010_a_0053000_0057000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9846
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0003_a_0013000_0017000
vocals
mixture
2 0.9818
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0009_a_0049000_0053000
vocals
mixture
3 0.9783
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0011_a_0057000_0065000
vocals
mixture
4 0.9773
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0026_b_0169000_0173000
vocals
mixture
5 0.9757
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0014_a_0076000_0080000
vocals
mixture
6 0.9749
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0023_a_0131000_0136000
vocals
mixture
7 0.9730
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0006_a_0032000_0037000
vocals
mixture
8 0.9716
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0002_a_0010000_0014000
vocals
mixture
9 0.9673
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0020_a_0113000_0118000
vocals
mixture
10 0.9671
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0007_a_0037000_0042000
vocals
mixture
11 0.9664
ANiMAL_-_Easy_Tiger
ANiMAL_-_Easy_Tiger_0003_b_0022000_0027000
vocals
mixture
12 0.9662
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0015_a_0080000_0085000
vocals
mixture
13 0.9658
Leaf_-_Wicked
Leaf_-_Wicked_0005_b_0068000_0072000
vocals
mixture
14 0.9642
Music_Delta_-_Disco
Music_Delta_-_Disco_0005_a_0075000_0080000
vocals
mixture
15 0.9642
Leaf_-_Wicked
Leaf_-_Wicked_0009_c_0111000_0115000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9836
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0014_a_0076000_0080000
vocals
mixture
2 0.9829
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0003_a_0013000_0017000
vocals
mixture
3 0.9772
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0009_a_0049000_0053000
vocals
mixture
4 0.9770
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0026_b_0169000_0173000
vocals
mixture
5 0.9755
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0023_a_0131000_0136000
vocals
mixture
6 0.9728
Leaf_-_Wicked
Leaf_-_Wicked_0005_b_0068000_0072000
vocals
mixture
7 0.9720
Leaf_-_Wicked
Leaf_-_Wicked_0008_c_0107000_0112000
vocals
mixture
8 0.9707
Music_Delta_-_Disco
Music_Delta_-_Disco_0004_a_0062000_0066000
vocals
mixture
9 0.9700
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0002_a_0010000_0014000
vocals
mixture
10 0.9699
Angels_In_Amplifiers_-_I_m_Alright
Angels_In_Amplifiers_-_I_m_Alright_0011_a_0080000_0085000
vocals
mixture
11 0.9689
Leaf_-_Wicked
Leaf_-_Wicked_0004_b_0064000_0068000
vocals
mixture
12 0.9681
Music_Delta_-_Disco
Music_Delta_-_Disco_0006_a_0095000_0099000
vocals
mixture
13 0.9679
Leaf_-_Wicked
Leaf_-_Wicked_0009_c_0111000_0115000
vocals
mixture
14 0.9677
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0003_b_0019000_0025000
vocals
mixture
15 0.9664
Fergessen_-_Back_From_The_Start
Fergessen_-_Back_From_The_Start_0015_a_0084000_0088000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 5.0147
MuQ_vocals: raw=0.9543, z=1.1711
ECAPA: raw=0.8689, z=8.8584
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0011_a_0057000_0065000
vocals
mixture
2 4.5014
MuQ_vocals: raw=0.9829, z=2.0911
ECAPA: raw=0.6841, z=6.9117
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0003_a_0013000_0017000
vocals
mixture
3 4.4309
MuQ_vocals: raw=0.9772, z=1.9069
ECAPA: raw=0.6882, z=6.9548
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0009_a_0049000_0053000
vocals
mixture
4 4.0709
MuQ_vocals: raw=0.9836, z=2.1146
ECAPA: raw=0.6002, z=6.0272
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0014_a_0076000_0080000
vocals
mixture
5 4.0119
MuQ_vocals: raw=0.9355, z=0.5657
ECAPA: raw=0.7360, z=7.4580
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0022_a_0123000_0131000
vocals
mixture
6 3.9790
MuQ_vocals: raw=0.9755, z=1.8515
ECAPA: raw=0.6077, z=6.1064
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0023_a_0131000_0136000
vocals
mixture
7 3.6444
MuQ_vocals: raw=0.9282, z=0.3308
ECAPA: raw=0.6885, z=6.9580
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0013_a_0068000_0076000
vocals
mixture
8 3.4231
MuQ_vocals: raw=0.9700, z=1.6754
ECAPA: raw=0.5189, z=5.1708
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0002_a_0010000_0014000
vocals
mixture
9 3.2601
MuQ_vocals: raw=0.9637, z=1.4723
ECAPA: raw=0.5072, z=5.0479
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0024_a_0152000_0159000
vocals
mixture
10 3.1259
MuQ_vocals: raw=0.9238, z=0.1891
ECAPA: raw=0.6036, z=6.0628
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0004_a_0017000_0026000
vocals
mixture
11 2.9499
MuQ_vocals: raw=0.9298, z=0.3820
ECAPA: raw=0.5518, z=5.5178
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0006_a_0032000_0037000
vocals
mixture
12 2.9305
MuQ_vocals: raw=0.9159, z=-0.0673
ECAPA: raw=0.5908, z=5.9283
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0021_a_0118000_0123000
vocals
mixture
13 2.9257
MuQ_vocals: raw=0.9453, z=0.8809
ECAPA: raw=0.4999, z=4.9705
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0015_a_0080000_0085000
vocals
mixture
14 2.5487
MuQ_vocals: raw=0.9770, z=1.9021
ECAPA: raw=0.3314, z=3.1953
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0026_b_0169000_0173000
vocals
mixture
15 2.2008
MuQ_vocals: raw=0.9311, z=0.4232
ECAPA: raw=0.4057, z=3.9784
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0007_a_0037000_0042000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 1.7251
MuQ_vocals: raw=0.9772, z=1.9069
STARS_phone_hidden: raw=0.8513, z=1.5432
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0009_a_0049000_0053000
vocals
mixture
2 1.6780
MuQ_vocals: raw=0.9653, z=1.5248
STARS_phone_hidden: raw=0.8882, z=1.8313
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0006_b_0038000_0044000
vocals
mixture
3 1.6707
MuQ_vocals: raw=0.9623, z=1.4292
STARS_phone_hidden: raw=0.8986, z=1.9122
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0017_a_0113000_0120000
vocals
mixture
4 1.6488
MuQ_vocals: raw=0.9627, z=1.4409
STARS_phone_hidden: raw=0.8915, z=1.8567
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0004_b_0025000_0032000
vocals
mixture
5 1.6325
MuQ_vocals: raw=0.9661, z=1.5507
STARS_phone_hidden: raw=0.8732, z=1.7142
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0008_b_0051000_0057000
vocals
mixture
6 1.6035
MuQ_vocals: raw=0.9755, z=1.8515
STARS_phone_hidden: raw=0.8272, z=1.3555
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0023_a_0131000_0136000
vocals
mixture
7 1.5140
MuQ_vocals: raw=0.9605, z=1.3711
STARS_phone_hidden: raw=0.8659, z=1.6570
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0015_b_0095000_0102000
vocals
mixture
8 1.5070
MuQ_vocals: raw=0.9829, z=2.0911
STARS_phone_hidden: raw=0.7718, z=0.9229
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0003_a_0013000_0017000
vocals
mixture
9 1.5017
MuQ_vocals: raw=0.9616, z=1.4062
STARS_phone_hidden: raw=0.8582, z=1.5971
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0011_b_0070000_0076000
vocals
mixture
10 1.5015
MuQ_vocals: raw=0.9578, z=1.2824
STARS_phone_hidden: raw=0.8740, z=1.7206
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0012_b_0076000_0083000
vocals
mixture
11 1.4503
MuQ_vocals: raw=0.9523, z=1.1063
STARS_phone_hidden: raw=0.8835, z=1.7943
BKS_-_Too_Much
BKS_-_Too_Much_0012_b_0083000_0087000
vocals
mixture
12 1.4301
MuQ_vocals: raw=0.9517, z=1.0851
STARS_phone_hidden: raw=0.8810, z=1.7750
Angels_In_Amplifiers_-_I_m_Alright
Angels_In_Amplifiers_-_I_m_Alright_0019_b_0138000_0143000
vocals
mixture
13 1.4164
MuQ_vocals: raw=0.9543, z=1.1711
STARS_phone_hidden: raw=0.8665, z=1.6617
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0011_a_0057000_0065000
vocals
mixture
14 1.4031
MuQ_vocals: raw=0.9607, z=1.3770
STARS_phone_hidden: raw=0.8367, z=1.4293
Patrick_Talbot_-_A_Reason_To_Leave
Patrick_Talbot_-_A_Reason_To_Leave_0007_a_0045000_0051000
vocals
mixture
15 1.3919
MuQ_vocals: raw=0.9594, z=1.3344
STARS_phone_hidden: raw=0.8392, z=1.4494
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0007_b_0044000_0051000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 3.8971
MuQ_vocals: raw=0.9543, z=1.1711
ECAPA: raw=0.8689, z=8.8584
STARS_phone_hidden: raw=0.8665, z=1.6617
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0011_a_0057000_0065000
vocals
mixture
2 3.4683
MuQ_vocals: raw=0.9772, z=1.9069
ECAPA: raw=0.6882, z=6.9548
STARS_phone_hidden: raw=0.8513, z=1.5432
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0009_a_0049000_0053000
vocals
mixture
3 3.3086
MuQ_vocals: raw=0.9829, z=2.0911
ECAPA: raw=0.6841, z=6.9117
STARS_phone_hidden: raw=0.7718, z=0.9229
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0003_a_0013000_0017000
vocals
mixture
4 3.1045
MuQ_vocals: raw=0.9755, z=1.8515
ECAPA: raw=0.6077, z=6.1064
STARS_phone_hidden: raw=0.8272, z=1.3555
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0023_a_0131000_0136000
vocals
mixture
5 2.7898
MuQ_vocals: raw=0.9836, z=2.1146
ECAPA: raw=0.6002, z=6.0272
STARS_phone_hidden: raw=0.6826, z=0.2275
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0014_a_0076000_0080000
vocals
mixture
6 2.7792
MuQ_vocals: raw=0.9282, z=0.3308
ECAPA: raw=0.6885, z=6.9580
STARS_phone_hidden: raw=0.7879, z=1.0489
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0013_a_0068000_0076000
vocals
mixture
7 2.5547
MuQ_vocals: raw=0.9238, z=0.1891
ECAPA: raw=0.6036, z=6.0628
STARS_phone_hidden: raw=0.8345, z=1.4121
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0004_a_0017000_0026000
vocals
mixture
8 2.4901
MuQ_vocals: raw=0.9637, z=1.4723
ECAPA: raw=0.5072, z=5.0479
STARS_phone_hidden: raw=0.7753, z=0.9502
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0024_a_0152000_0159000
vocals
mixture
9 2.4669
MuQ_vocals: raw=0.9355, z=0.5657
ECAPA: raw=0.7360, z=7.4580
STARS_phone_hidden: raw=0.5736, z=-0.6230
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0022_a_0123000_0131000
vocals
mixture
10 2.3472
MuQ_vocals: raw=0.9700, z=1.6754
ECAPA: raw=0.5189, z=5.1708
STARS_phone_hidden: raw=0.6785, z=0.1952
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0002_a_0010000_0014000
vocals
mixture
11 2.2252
MuQ_vocals: raw=0.9159, z=-0.0673
ECAPA: raw=0.5908, z=5.9283
STARS_phone_hidden: raw=0.7579, z=0.8145
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0021_a_0118000_0123000
vocals
mixture
12 2.0864
MuQ_vocals: raw=0.9298, z=0.3820
ECAPA: raw=0.5518, z=5.5178
STARS_phone_hidden: raw=0.6995, z=0.3594
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0006_a_0032000_0037000
vocals
mixture
13 1.9525
MuQ_vocals: raw=0.9453, z=0.8809
ECAPA: raw=0.4999, z=4.9705
STARS_phone_hidden: raw=0.6543, z=0.0062
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0015_a_0080000_0085000
vocals
mixture
14 1.9482
MuQ_vocals: raw=0.9311, z=0.4232
ECAPA: raw=0.4057, z=3.9784
STARS_phone_hidden: raw=0.8384, z=1.4429
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0007_a_0037000_0042000
vocals
mixture
15 1.9363
MuQ_vocals: raw=0.9770, z=1.9021
ECAPA: raw=0.3314, z=3.1953
STARS_phone_hidden: raw=0.7447, z=0.7116
Punkdisco_-_Oral_Hygiene
Punkdisco_-_Oral_Hygiene_0026_b_0169000_0173000
vocals
mixture

Query 10: Lyndsey_Ollard_-_Catching_Up / Lyndsey_Ollard_-_Catching_Up_0021_c_0140000_0146000

Query vocals
Query mixture
common candidates: 2535 / fusion: per-query z-score, equal weights

MuQ(mixture入力) Top 15

ランキングへの入力
MuQ入力: mixture(ボーカル+伴奏)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9871
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0023_c_0152000_0157000
vocals
mixture
2 0.9842
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0027_a_0178000_0184000
vocals
mixture
3 0.9783
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0026_a_0172000_0178000
vocals
mixture
4 0.9775
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0025_a_0165000_0172000
vocals
mixture
5 0.9730
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0014_b_0089000_0095000
vocals
mixture
6 0.9728
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0018_a_0116000_0122000
vocals
mixture
7 0.9716
Al_James_-_Schoolboy_Facination
Al_James_-_Schoolboy_Facination_0016_b_0107000_0113000
vocals
mixture
8 0.9715
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0023_a_0145000_0151000
vocals
mixture
9 0.9709
Alexander_Ross_-_Goodbye_Bolero
Alexander_Ross_-_Goodbye_Bolero_0021_a_0135000_0140000
vocals
mixture
10 0.9708
Al_James_-_Schoolboy_Facination
Al_James_-_Schoolboy_Facination_0006_b_0044000_0050000
vocals
mixture
11 0.9703
Al_James_-_Schoolboy_Facination
Al_James_-_Schoolboy_Facination_0023_b_0154000_0161000
vocals
mixture
12 0.9703
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0011_b_0070000_0076000
vocals
mixture
13 0.9693
BKS_-_Bulldozer
BKS_-_Bulldozer_0023_a_0165000_0170000
vocals
mixture
14 0.9692
BKS_-_Bulldozer
BKS_-_Bulldozer_0012_a_0092000_0098000
vocals
mixture
15 0.9691
James_May_-_On_The_Line
James_May_-_On_The_Line_0026_a_0170000_0176000
vocals
mixture

MuQ(vocals入力) Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ランキング算出: コサイン類似度
Rank Cosine Audio / Clip
1 0.9785
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0023_c_0152000_0157000
vocals
mixture
2 0.9779
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0026_a_0172000_0178000
vocals
mixture
3 0.9766
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0027_a_0178000_0184000
vocals
mixture
4 0.9747
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0025_a_0165000_0172000
vocals
mixture
5 0.9744
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0029_b_0191000_0197000
vocals
mixture
6 0.9713
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0012_b_0076000_0083000
vocals
mixture
7 0.9709
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0027_b_0197000_0202000
vocals
mixture
8 0.9708
Actions_-_Devil_s_Words
Actions_-_Devil_s_Words_0031_b_0184000_0189000
vocals
mixture
9 0.9706
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0013_b_0102000_0108000
vocals
mixture
10 0.9706
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0011_b_0088000_0095000
vocals
mixture
11 0.9700
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0026_a_0196000_0201000
vocals
mixture
12 0.9696
Actions_-_Devil_s_Words
Actions_-_Devil_s_Words_0011_a_0072000_0078000
vocals
mixture
13 0.9693
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0021_b_0161000_0167000
vocals
mixture
14 0.9692
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0017_b_0132000_0139000
vocals
mixture
15 0.9691
The_Sunshine_Garcia_Band_-_For_I_Am_The_Moon
The_Sunshine_Garcia_Band_-_For_I_Am_The_Moon_0009_a_0189000_0196000
vocals
mixture

MuQ(vocals)+ECAPA Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 2.7973
MuQ_vocals: raw=0.9785, z=1.7005
ECAPA: raw=0.5865, z=3.8940
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0023_c_0152000_0157000
vocals
mixture
2 2.6560
MuQ_vocals: raw=0.9747, z=1.5671
ECAPA: raw=0.5702, z=3.7449
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0025_a_0165000_0172000
vocals
mixture
3 2.6488
MuQ_vocals: raw=0.9779, z=1.6806
ECAPA: raw=0.5562, z=3.6170
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0026_a_0172000_0178000
vocals
mixture
4 2.6280
MuQ_vocals: raw=0.9744, z=1.5573
ECAPA: raw=0.5652, z=3.6988
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0029_b_0191000_0197000
vocals
mixture
5 2.5671
MuQ_vocals: raw=0.9670, z=1.2997
ECAPA: raw=0.5800, z=3.8344
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0031_b_0203000_0208000
vocals
mixture
6 2.5201
MuQ_vocals: raw=0.9766, z=1.6340
ECAPA: raw=0.5333, z=3.4062
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0027_a_0178000_0184000
vocals
mixture
7 2.5045
MuQ_vocals: raw=0.9665, z=1.2831
ECAPA: raw=0.5681, z=3.7259
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0024_c_0157000_0165000
vocals
mixture
8 2.1788
MuQ_vocals: raw=0.9682, z=1.3406
ECAPA: raw=0.4908, z=3.0171
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0010_b_0064000_0070000
vocals
mixture
9 2.1438
MuQ_vocals: raw=0.9675, z=1.3162
ECAPA: raw=0.4858, z=2.9714
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0002_b_0014000_0019000
vocals
mixture
10 2.0761
MuQ_vocals: raw=0.9510, z=0.7412
ECAPA: raw=0.5338, z=3.4111
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0028_a_0184000_0191000
vocals
mixture
11 2.0713
MuQ_vocals: raw=0.9664, z=1.2788
ECAPA: raw=0.4741, z=2.8638
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0011_b_0070000_0076000
vocals
mixture
12 1.9143
MuQ_vocals: raw=0.9591, z=1.0224
ECAPA: raw=0.4678, z=2.8062
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0011_a_0077000_0086000
vocals
mixture
13 1.9027
MuQ_vocals: raw=0.9706, z=1.4240
ECAPA: raw=0.4215, z=2.3814
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0011_b_0088000_0095000
vocals
mixture
14 1.8990
MuQ_vocals: raw=0.9713, z=1.4484
ECAPA: raw=0.4180, z=2.3497
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0012_b_0076000_0083000
vocals
mixture
15 1.8513
MuQ_vocals: raw=0.9548, z=0.8723
ECAPA: raw=0.4704, z=2.8303
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0015_b_0095000_0102000
vocals
mixture

MuQ(vocals)+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、等重み平均
Rank Fused z Component scores Audio / Clip
1 1.8439
MuQ_vocals: raw=0.9665, z=1.2831
STARS_phone_hidden: raw=0.8110, z=2.4048
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0024_c_0157000_0165000
vocals
mixture
2 1.7309
MuQ_vocals: raw=0.9682, z=1.3406
STARS_phone_hidden: raw=0.7751, z=2.1212
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0010_b_0064000_0070000
vocals
mixture
3 1.6727
MuQ_vocals: raw=0.9655, z=1.2467
STARS_phone_hidden: raw=0.7723, z=2.0988
Johnny_Lokke_-_Promises_Lies
Johnny_Lokke_-_Promises_Lies_0006_a_0057000_0063000
vocals
mixture
4 1.6601
MuQ_vocals: raw=0.9659, z=1.2619
STARS_phone_hidden: raw=0.7672, z=2.0584
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0005_b_0037000_0042000
vocals
mixture
5 1.6478
MuQ_vocals: raw=0.9646, z=1.2161
STARS_phone_hidden: raw=0.7699, z=2.0795
Actions_-_One_Minute_Smile
Actions_-_One_Minute_Smile_0001_a_0006000_0012000
vocals
mixture
6 1.6379
MuQ_vocals: raw=0.9545, z=0.8643
STARS_phone_hidden: raw=0.8119, z=2.4115
Leaf_-_Summerghost
Leaf_-_Summerghost_0015_b_0122000_0128000
vocals
mixture
7 1.6059
MuQ_vocals: raw=0.9637, z=1.1846
STARS_phone_hidden: raw=0.7633, z=2.0272
James_May_-_On_The_Line
James_May_-_On_The_Line_0024_a_0158000_0164000
vocals
mixture
8 1.5518
MuQ_vocals: raw=0.9568, z=0.9424
STARS_phone_hidden: raw=0.7802, z=2.1613
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0006_a_0043000_0048000
vocals
mixture
9 1.5418
MuQ_vocals: raw=0.9589, z=1.0171
STARS_phone_hidden: raw=0.7682, z=2.0666
Nerve_9_-_Pray_For_The_Rain
Nerve_9_-_Pray_For_The_Rain_0014_a_0106000_0112000
vocals
mixture
10 1.5281
MuQ_vocals: raw=0.9546, z=0.8657
STARS_phone_hidden: raw=0.7839, z=2.1905
Young_Griffo_-_Blood_To_Bone
Young_Griffo_-_Blood_To_Bone_0014_c_0104000_0110000
vocals
mixture
11 1.5241
MuQ_vocals: raw=0.9779, z=1.6806
STARS_phone_hidden: raw=0.6798, z=1.3677
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0026_a_0172000_0178000
vocals
mixture
12 1.4659
MuQ_vocals: raw=0.9691, z=1.3725
STARS_phone_hidden: raw=0.7040, z=1.5593
The_Sunshine_Garcia_Band_-_For_I_Am_The_Moon
The_Sunshine_Garcia_Band_-_For_I_Am_The_Moon_0009_a_0189000_0196000
vocals
mixture
13 1.4278
MuQ_vocals: raw=0.9599, z=1.0503
STARS_phone_hidden: raw=0.7352, z=1.8053
Swinging_Steaks_-_Lost_My_Way
Swinging_Steaks_-_Lost_My_Way_0020_b_0140000_0146000
vocals
mixture
14 1.4160
MuQ_vocals: raw=0.9581, z=0.9900
STARS_phone_hidden: raw=0.7398, z=1.8420
Remember_December_-_C_U_Next_Time
Remember_December_-_C_U_Next_Time_0029_b_0192000_0197000
vocals
mixture
15 1.3941
MuQ_vocals: raw=0.9553, z=0.8906
STARS_phone_hidden: raw=0.7468, z=1.8976
The_Long_Wait_-_Dark_Horses
The_Long_Wait_-_Dark_Horses_0013_b_0105000_0112000
vocals
mixture

MuQ(vocals)+ECAPA+STARS phone_hidden Top 15

ランキングへの入力
MuQ入力: vocals(ボーカルのみ)特徴量: Conformer Encoder 最終層
ECAPA入力: vocals(ボーカルのみ)特徴量: 192次元の話者・声質特徴量
STARS入力: vocals(ボーカルのみ)特徴量: phone_hidden
ランキング算出: 各類似度をz-score標準化後、3モデルを等重み平均
Rank Fused z Component scores Audio / Clip
1 2.4712
MuQ_vocals: raw=0.9665, z=1.2831
ECAPA: raw=0.5681, z=3.7259
STARS_phone_hidden: raw=0.8110, z=2.4048
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0024_c_0157000_0165000
vocals
mixture
2 2.2218
MuQ_vocals: raw=0.9779, z=1.6806
ECAPA: raw=0.5562, z=3.6170
STARS_phone_hidden: raw=0.6798, z=1.3677
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0026_a_0172000_0178000
vocals
mixture
3 2.1596
MuQ_vocals: raw=0.9682, z=1.3406
ECAPA: raw=0.4908, z=3.0171
STARS_phone_hidden: raw=0.7751, z=2.1212
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0010_b_0064000_0070000
vocals
mixture
4 2.0965
MuQ_vocals: raw=0.9670, z=1.2997
ECAPA: raw=0.5800, z=3.8344
STARS_phone_hidden: raw=0.6529, z=1.1553
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0031_b_0203000_0208000
vocals
mixture
5 2.0495
MuQ_vocals: raw=0.9744, z=1.5573
ECAPA: raw=0.5652, z=3.6988
STARS_phone_hidden: raw=0.6196, z=0.8924
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0029_b_0191000_0197000
vocals
mixture
6 1.9399
MuQ_vocals: raw=0.9545, z=0.8643
ECAPA: raw=0.4392, z=2.5440
STARS_phone_hidden: raw=0.8119, z=2.4115
Leaf_-_Summerghost
Leaf_-_Summerghost_0015_b_0122000_0128000
vocals
mixture
7 1.9315
MuQ_vocals: raw=0.9747, z=1.5671
ECAPA: raw=0.5702, z=3.7449
STARS_phone_hidden: raw=0.5677, z=0.4823
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0025_a_0165000_0172000
vocals
mixture
8 1.7665
MuQ_vocals: raw=0.9591, z=1.0224
ECAPA: raw=0.4678, z=2.8062
STARS_phone_hidden: raw=0.6929, z=1.4710
Angela_Thomas_Wade_-_Milk_Cow_Blues
Angela_Thomas_Wade_-_Milk_Cow_Blues_0011_a_0077000_0086000
vocals
mixture
9 1.7591
MuQ_vocals: raw=0.9766, z=1.6340
ECAPA: raw=0.5333, z=3.4062
STARS_phone_hidden: raw=0.5367, z=0.2371
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0027_a_0178000_0184000
vocals
mixture
10 1.7453
MuQ_vocals: raw=0.9691, z=1.3725
ECAPA: raw=0.4131, z=2.3040
STARS_phone_hidden: raw=0.7040, z=1.5593
The_Sunshine_Garcia_Band_-_For_I_Am_The_Moon
The_Sunshine_Garcia_Band_-_For_I_Am_The_Moon_0009_a_0189000_0196000
vocals
mixture
11 1.6991
MuQ_vocals: raw=0.9566, z=0.9343
ECAPA: raw=0.4186, z=2.3551
STARS_phone_hidden: raw=0.7355, z=1.8080
Motor_Tapes_-_Shore
Motor_Tapes_-_Shore_0009_b_0081000_0087000
vocals
mixture
12 1.6542
MuQ_vocals: raw=0.9664, z=1.2788
ECAPA: raw=0.4741, z=2.8638
STARS_phone_hidden: raw=0.6105, z=0.8199
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0011_b_0070000_0076000
vocals
mixture
13 1.6192
MuQ_vocals: raw=0.9581, z=0.9900
ECAPA: raw=0.3827, z=2.0256
STARS_phone_hidden: raw=0.7398, z=1.8420
Remember_December_-_C_U_Next_Time
Remember_December_-_C_U_Next_Time_0029_b_0192000_0197000
vocals
mixture
14 1.6074
MuQ_vocals: raw=0.9692, z=1.3760
ECAPA: raw=0.4133, z=2.3066
STARS_phone_hidden: raw=0.6509, z=1.1397
Juliet_s_Rescue_-_Heartbeats
Juliet_s_Rescue_-_Heartbeats_0017_b_0132000_0139000
vocals
mixture
15 1.5710
MuQ_vocals: raw=0.9713, z=1.4484
ECAPA: raw=0.4180, z=2.3497
STARS_phone_hidden: raw=0.6225, z=0.9150
Lyndsey_Ollard_-_Catching_Up
Lyndsey_Ollard_-_Catching_Up_0012_b_0076000_0083000
vocals
mixture