Upload SSL Layer MOS checkpoints (part 5)

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run2/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab412a40c6a2a7fb78b6ba80e6a890b9b5dc58db1ebf1b0ea16c266fa97849e5
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run2/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7698ab5576e7ac936a82b6fbf0546adbc3155970ce8bf6022014b940076b93e5
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run2/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.4333679804053062
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3561 | LCC = 0.8064 | SRCC = 0.8111 ]
+INFO:root:Average loss=0.27634325871864956
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3084 | LCC = 0.8394 | SRCC = 0.8386 ]
+INFO:root:Average loss=0.2417834808047001
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3485 | LCC = 0.8483 | SRCC = 0.8507 ]
+INFO:root:Average loss=0.2363869091257071
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2711 | LCC = 0.8551 | SRCC = 0.8538 ]
+INFO:root:Average loss=0.21763800313839546
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2544 | LCC = 0.8584 | SRCC = 0.8584 ]
+INFO:root:Average loss=0.19965359090994567
+INFO:root:
+[Valid][5][UTT][ MSE = 0.3070 | LCC = 0.8598 | SRCC = 0.8588 ]
+INFO:root:Average loss=0.18454301662934133
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2501 | LCC = 0.8572 | SRCC = 0.8555 ]
+INFO:root:Average loss=0.17260222862928343
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2483 | LCC = 0.8597 | SRCC = 0.8598 ]
+INFO:root:Average loss=0.16166870964643282
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2490 | LCC = 0.8598 | SRCC = 0.8600 ]
+INFO:root:Average loss=0.14926960261968467
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2632 | LCC = 0.8607 | SRCC = 0.8602 ]
+INFO:root:Average loss=0.14366154783429244
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2406 | LCC = 0.8575 | SRCC = 0.8566 ]
+INFO:root:Average loss=0.13270336131636912
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2480 | LCC = 0.8566 | SRCC = 0.8561 ]
+INFO:root:Average loss=0.12664856838110167
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2659 | LCC = 0.8546 | SRCC = 0.8570 ]
+INFO:root:Average loss=0.11981937614006874
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2470 | LCC = 0.8519 | SRCC = 0.8529 ]
+INFO:root:Average loss=0.1176408742291805
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2629 | LCC = 0.8542 | SRCC = 0.8551 ]
+INFO:root:Average loss=0.10629363157428227
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2790 | LCC = 0.8513 | SRCC = 0.8532 ]
+INFO:root:Average loss=0.09905539023188445
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2683 | LCC = 0.8479 | SRCC = 0.8480 ]
+INFO:root:Average loss=0.09355268398156533
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2807 | LCC = 0.8432 | SRCC = 0.8459 ]
+INFO:root:Average loss=0.0945625029122218
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2507 | LCC = 0.8487 | SRCC = 0.8497 ]
+INFO:root:Average loss=0.0855995132946051
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2653 | LCC = 0.8466 | SRCC = 0.8466 ]
+INFO:root:Average loss=0.08877791110903789
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2444 | LCC = 0.8511 | SRCC = 0.8523 ]
+INFO:root:Average loss=0.08188509148282883
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2533 | LCC = 0.8484 | SRCC = 0.8497 ]
+INFO:root:Average loss=0.07404621523351242
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2624 | LCC = 0.8462 | SRCC = 0.8473 ]
+INFO:root:Average loss=0.07226133031340745
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2529 | LCC = 0.8506 | SRCC = 0.8515 ]
+INFO:root:Average loss=0.06969021642819428
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2734 | LCC = 0.8433 | SRCC = 0.8429 ]
+INFO:root:Average loss=0.0658117977854533
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2387 | LCC = 0.8498 | SRCC = 0.8512 ]
+INFO:root:Average loss=0.06294814812449309
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2505 | LCC = 0.8490 | SRCC = 0.8490 ]
+INFO:root:Average loss=0.05886615440249443
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2603 | LCC = 0.8470 | SRCC = 0.8474 ]
+INFO:root:Average loss=0.06502897827289043
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2498 | LCC = 0.8478 | SRCC = 0.8487 ]
+INFO:root:Average loss=0.05716708899499514
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2474 | LCC = 0.8491 | SRCC = 0.8506 ]
+INFO:root:Average loss=0.05529519208730795
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2509 | LCC = 0.8480 | SRCC = 0.8496 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2618 | LCC = 0.8545 | SRCC = 0.8545 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run3/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run3/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c044c654a235ea22bca38526a79abc04b31e7b75fec343bedeefdf651b13165
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run3/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:799fe32dda6351c6f9545fff54e5d9ffdf44818f21b1e8eb9cb0347b0a8745dc
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run3/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.43758178807986087
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3539 | LCC = 0.8074 | SRCC = 0.8132 ]
+INFO:root:Average loss=0.27459769141979706
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3049 | LCC = 0.8408 | SRCC = 0.8401 ]
+INFO:root:Average loss=0.2429573260820829
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3024 | LCC = 0.8515 | SRCC = 0.8493 ]
+INFO:root:Average loss=0.22346308005925936
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2774 | LCC = 0.8556 | SRCC = 0.8543 ]
+INFO:root:Average loss=0.21441376257019165
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2575 | LCC = 0.8587 | SRCC = 0.8564 ]
+INFO:root:Average loss=0.20390632767707872
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2398 | LCC = 0.8612 | SRCC = 0.8603 ]
+INFO:root:Average loss=0.1827615274068637
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2447 | LCC = 0.8647 | SRCC = 0.8631 ]
+INFO:root:Average loss=0.17666180852131966
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2448 | LCC = 0.8626 | SRCC = 0.8612 ]
+INFO:root:Average loss=0.17401467999204612
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2554 | LCC = 0.8610 | SRCC = 0.8603 ]
+INFO:root:Average loss=0.15118789195250243
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2648 | LCC = 0.8550 | SRCC = 0.8547 ]
+INFO:root:Average loss=0.14323040499137
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2619 | LCC = 0.8552 | SRCC = 0.8540 ]
+INFO:root:Average loss=0.13883352661744142
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2436 | LCC = 0.8555 | SRCC = 0.8536 ]
+INFO:root:Average loss=0.13759343230571502
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2495 | LCC = 0.8558 | SRCC = 0.8549 ]
+INFO:root:Average loss=0.12571560648771432
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2366 | LCC = 0.8557 | SRCC = 0.8539 ]
+INFO:root:Average loss=0.11170433576290424
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2476 | LCC = 0.8527 | SRCC = 0.8527 ]
+INFO:root:Average loss=0.11621865797310303
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2430 | LCC = 0.8506 | SRCC = 0.8492 ]
+INFO:root:Average loss=0.11055554239413677
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2520 | LCC = 0.8499 | SRCC = 0.8491 ]
+INFO:root:Average loss=0.09552048122844635
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2442 | LCC = 0.8482 | SRCC = 0.8477 ]
+INFO:root:Average loss=0.0940097991663676
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2473 | LCC = 0.8450 | SRCC = 0.8442 ]
+INFO:root:Average loss=0.08238960525546318
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2469 | LCC = 0.8447 | SRCC = 0.8448 ]
+INFO:root:Average loss=0.08315739842752616
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2684 | LCC = 0.8465 | SRCC = 0.8446 ]
+INFO:root:Average loss=0.07951096015480849
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2539 | LCC = 0.8463 | SRCC = 0.8448 ]
+INFO:root:Average loss=0.0808035160582035
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2467 | LCC = 0.8466 | SRCC = 0.8462 ]
+INFO:root:Average loss=0.06967812523436852
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2423 | LCC = 0.8485 | SRCC = 0.8481 ]
+INFO:root:Average loss=0.06693637542999707
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2659 | LCC = 0.8458 | SRCC = 0.8452 ]
+INFO:root:Average loss=0.06728526051992025
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2427 | LCC = 0.8468 | SRCC = 0.8467 ]
+INFO:root:Average loss=0.06428273881857212
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2512 | LCC = 0.8420 | SRCC = 0.8412 ]
+INFO:root:Average loss=0.06212889455640928
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2510 | LCC = 0.8475 | SRCC = 0.8468 ]
+INFO:root:Average loss=0.05527650297452242
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2555 | LCC = 0.8418 | SRCC = 0.8410 ]
+INFO:root:Average loss=0.05925052374219283
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2549 | LCC = 0.8511 | SRCC = 0.8518 ]
+INFO:root:Average loss=0.0530470193674167
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2690 | LCC = 0.8456 | SRCC = 0.8437 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2736 | LCC = 0.8476 | SRCC = 0.8479 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run4/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run4/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c005cc74d5043872a925bfe06ff3108e12e70282024c0b9cb22bc4a514fe008
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run4/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e46e158aa41028fdf4ecd033cb845311dcaf0fafaf73575ad130afc911de705
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer7/run4/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.452056314318608
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3180 | LCC = 0.8033 | SRCC = 0.8126 ]
+INFO:root:Average loss=0.28075962494581175
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3133 | LCC = 0.8388 | SRCC = 0.8359 ]
+INFO:root:Average loss=0.24883588747336313
+INFO:root:
+[Valid][2][UTT][ MSE = 0.2865 | LCC = 0.8490 | SRCC = 0.8459 ]
+INFO:root:Average loss=0.22087921412327352
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2560 | LCC = 0.8562 | SRCC = 0.8534 ]
+INFO:root:Average loss=0.20980743165963736
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2687 | LCC = 0.8581 | SRCC = 0.8557 ]
+INFO:root:Average loss=0.19847632696231207
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2448 | LCC = 0.8600 | SRCC = 0.8589 ]
+INFO:root:Average loss=0.17746730502217245
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2413 | LCC = 0.8595 | SRCC = 0.8583 ]
+INFO:root:Average loss=0.16760278779726762
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2404 | LCC = 0.8605 | SRCC = 0.8585 ]
+INFO:root:Average loss=0.15927141522749877
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2363 | LCC = 0.8587 | SRCC = 0.8557 ]
+INFO:root:Average loss=0.14369378086083975
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2501 | LCC = 0.8568 | SRCC = 0.8550 ]
+INFO:root:Average loss=0.13832736101288062
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2567 | LCC = 0.8544 | SRCC = 0.8524 ]
+INFO:root:Average loss=0.12984634582430887
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2514 | LCC = 0.8541 | SRCC = 0.8525 ]
+INFO:root:Average loss=0.11912940853299239
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2415 | LCC = 0.8569 | SRCC = 0.8557 ]
+INFO:root:Average loss=0.11701363540039612
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2581 | LCC = 0.8531 | SRCC = 0.8513 ]
+INFO:root:Average loss=0.10647028144926597
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2474 | LCC = 0.8503 | SRCC = 0.8478 ]
+INFO:root:Average loss=0.09998841378360222
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2427 | LCC = 0.8517 | SRCC = 0.8514 ]
+INFO:root:Average loss=0.09776374286948106
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2416 | LCC = 0.8488 | SRCC = 0.8487 ]
+INFO:root:Average loss=0.09849136690489757
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2510 | LCC = 0.8530 | SRCC = 0.8514 ]
+INFO:root:Average loss=0.08787564326746342
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2531 | LCC = 0.8468 | SRCC = 0.8472 ]
+INFO:root:Average loss=0.08047295696078202
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2503 | LCC = 0.8481 | SRCC = 0.8469 ]
+INFO:root:Average loss=0.0800807113544299
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2806 | LCC = 0.8476 | SRCC = 0.8461 ]
+INFO:root:Average loss=0.07766213654898681
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2480 | LCC = 0.8466 | SRCC = 0.8454 ]
+INFO:root:Average loss=0.06867456698837952
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2421 | LCC = 0.8462 | SRCC = 0.8455 ]
+INFO:root:Average loss=0.07454978598233981
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2873 | LCC = 0.8408 | SRCC = 0.8422 ]
+INFO:root:Average loss=0.06894614456746823
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2485 | LCC = 0.8482 | SRCC = 0.8472 ]
+INFO:root:Average loss=0.06405419498108901
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2435 | LCC = 0.8488 | SRCC = 0.8475 ]
+INFO:root:Average loss=0.06023827629784743
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2630 | LCC = 0.8463 | SRCC = 0.8452 ]
+INFO:root:Average loss=0.05772898284097513
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2517 | LCC = 0.8441 | SRCC = 0.8437 ]
+INFO:root:Average loss=0.05193462480719273
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2475 | LCC = 0.8469 | SRCC = 0.8466 ]
+INFO:root:Average loss=0.05206845416568029
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2494 | LCC = 0.8462 | SRCC = 0.8451 ]
+INFO:root:Average loss=0.048558820803196005
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2480 | LCC = 0.8450 | SRCC = 0.8443 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2674 | LCC = 0.8480 | SRCC = 0.8486 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run0/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run0/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a75ea57d1c60fcc537a34a6c7f3689c31eaec88efc4140bafdc7bfb668461c43
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run0/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cffb8133bf1e41da74124aac9cb28a7eb05e9d22f8f5ddda8411c7e55d06b40
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run0/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.47223941676127607
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3751 | LCC = 0.7904 | SRCC = 0.8011 ]
+INFO:root:Average loss=0.27982731278126055
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3302 | LCC = 0.8303 | SRCC = 0.8308 ]
+INFO:root:Average loss=0.2549691217449995
+INFO:root:
+[Valid][2][UTT][ MSE = 0.2838 | LCC = 0.8463 | SRCC = 0.8453 ]
+INFO:root:Average loss=0.23500471390210664
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2539 | LCC = 0.8503 | SRCC = 0.8517 ]
+INFO:root:Average loss=0.21987843398864454
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2718 | LCC = 0.8506 | SRCC = 0.8522 ]
+INFO:root:Average loss=0.2170359866741376
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2457 | LCC = 0.8586 | SRCC = 0.8584 ]
+INFO:root:Average loss=0.20060518842477065
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2286 | LCC = 0.8626 | SRCC = 0.8630 ]
+INFO:root:Average loss=0.19284183226334742
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2629 | LCC = 0.8586 | SRCC = 0.8582 ]
+INFO:root:Average loss=0.17678190242403594
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2560 | LCC = 0.8595 | SRCC = 0.8614 ]
+INFO:root:Average loss=0.17932205007244378
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2733 | LCC = 0.8643 | SRCC = 0.8641 ]
+INFO:root:Average loss=0.16425106836817202
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2289 | LCC = 0.8613 | SRCC = 0.8604 ]
+INFO:root:Average loss=0.16195728620275474
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2871 | LCC = 0.8548 | SRCC = 0.8567 ]
+INFO:root:Average loss=0.14901328545350295
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2531 | LCC = 0.8595 | SRCC = 0.8583 ]
+INFO:root:Average loss=0.1450820328333439
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2588 | LCC = 0.8582 | SRCC = 0.8622 ]
+INFO:root:Average loss=0.14872629548876715
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2499 | LCC = 0.8616 | SRCC = 0.8629 ]
+INFO:root:Average loss=0.13497571045389542
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2722 | LCC = 0.8587 | SRCC = 0.8610 ]
+INFO:root:Average loss=0.12823205307508126
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2513 | LCC = 0.8575 | SRCC = 0.8599 ]
+INFO:root:Average loss=0.11853662391121571
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2392 | LCC = 0.8583 | SRCC = 0.8596 ]
+INFO:root:Average loss=0.1215384110617332
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2400 | LCC = 0.8624 | SRCC = 0.8627 ]
+INFO:root:Average loss=0.11138051327986595
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2692 | LCC = 0.8510 | SRCC = 0.8559 ]
+INFO:root:Average loss=0.10720421330860028
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2219 | LCC = 0.8603 | SRCC = 0.8609 ]
+INFO:root:Average loss=0.10368176158039998
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2606 | LCC = 0.8563 | SRCC = 0.8613 ]
+INFO:root:Average loss=0.09395950932342273
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2329 | LCC = 0.8539 | SRCC = 0.8549 ]
+INFO:root:Average loss=0.0912096606901823
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2483 | LCC = 0.8518 | SRCC = 0.8572 ]
+INFO:root:Average loss=0.0967549738020469
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2274 | LCC = 0.8592 | SRCC = 0.8603 ]
+INFO:root:Average loss=0.08598366322425696
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2381 | LCC = 0.8554 | SRCC = 0.8547 ]
+INFO:root:Average loss=0.08496240774790446
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2299 | LCC = 0.8567 | SRCC = 0.8560 ]
+INFO:root:Average loss=0.07885864548958264
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2292 | LCC = 0.8565 | SRCC = 0.8591 ]
+INFO:root:Average loss=0.07814654598060326
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2370 | LCC = 0.8523 | SRCC = 0.8532 ]
+INFO:root:Average loss=0.0762549732835629
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2385 | LCC = 0.8525 | SRCC = 0.8537 ]
+INFO:root:Average loss=0.07473513718025807
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2510 | LCC = 0.8476 | SRCC = 0.8498 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2667 | LCC = 0.8646 | SRCC = 0.8623 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run1/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run1/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdafd90b7bc12c729c4df2e3a1609d1346e0c4c2e5ebf3bd6b18b0c23d7c7de3
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run1/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9dd2f52000cdb89be8a5b10597c15b96ad3e0745365a10c38fbe2eb634132fef
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run1/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.41303758246776384
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3732 | LCC = 0.8069 | SRCC = 0.8122 ]
+INFO:root:Average loss=0.285670280838624
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3252 | LCC = 0.8318 | SRCC = 0.8336 ]
+INFO:root:Average loss=0.2465732741432312
+INFO:root:
+[Valid][2][UTT][ MSE = 0.2782 | LCC = 0.8431 | SRCC = 0.8426 ]
+INFO:root:Average loss=0.23031453559031853
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2695 | LCC = 0.8469 | SRCC = 0.8471 ]
+INFO:root:Average loss=0.2165564417074888
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2533 | LCC = 0.8569 | SRCC = 0.8579 ]
+INFO:root:Average loss=0.20914513503129667
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2484 | LCC = 0.8551 | SRCC = 0.8553 ]
+INFO:root:Average loss=0.19378782894748908
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2830 | LCC = 0.8554 | SRCC = 0.8556 ]
+INFO:root:Average loss=0.19045280875303808
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2577 | LCC = 0.8613 | SRCC = 0.8614 ]
+INFO:root:Average loss=0.1825267035418596
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2559 | LCC = 0.8612 | SRCC = 0.8587 ]
+INFO:root:Average loss=0.1781048531142565
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2480 | LCC = 0.8628 | SRCC = 0.8614 ]
+INFO:root:Average loss=0.1567079455424578
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2610 | LCC = 0.8596 | SRCC = 0.8582 ]
+INFO:root:Average loss=0.15464371643387353
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2663 | LCC = 0.8602 | SRCC = 0.8593 ]
+INFO:root:Average loss=0.15943133668639722
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2434 | LCC = 0.8588 | SRCC = 0.8588 ]
+INFO:root:Average loss=0.13974300399422646
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2527 | LCC = 0.8597 | SRCC = 0.8590 ]
+INFO:root:Average loss=0.13362917944024771
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2506 | LCC = 0.8576 | SRCC = 0.8567 ]
+INFO:root:Average loss=0.14108931206357786
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2575 | LCC = 0.8563 | SRCC = 0.8579 ]
+INFO:root:Average loss=0.12080439170583701
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2339 | LCC = 0.8572 | SRCC = 0.8572 ]
+INFO:root:Average loss=0.12141439557457581
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2522 | LCC = 0.8558 | SRCC = 0.8563 ]
+INFO:root:Average loss=0.11084135382985458
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2263 | LCC = 0.8595 | SRCC = 0.8601 ]
+INFO:root:Average loss=0.10768489721111763
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2399 | LCC = 0.8555 | SRCC = 0.8557 ]
+INFO:root:Average loss=0.10416071470349263
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2568 | LCC = 0.8535 | SRCC = 0.8535 ]
+INFO:root:Average loss=0.1003027319526061
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2479 | LCC = 0.8515 | SRCC = 0.8523 ]
+INFO:root:Average loss=0.09413785067124245
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2455 | LCC = 0.8547 | SRCC = 0.8538 ]
+INFO:root:Average loss=0.09474462060592113
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2563 | LCC = 0.8475 | SRCC = 0.8485 ]
+INFO:root:Average loss=0.08715413195582536
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2635 | LCC = 0.8497 | SRCC = 0.8499 ]
+INFO:root:Average loss=0.08358084114316182
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2458 | LCC = 0.8492 | SRCC = 0.8482 ]
+INFO:root:Average loss=0.08539289790086257
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2477 | LCC = 0.8525 | SRCC = 0.8543 ]
+INFO:root:Average loss=0.07988832332193851
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2406 | LCC = 0.8528 | SRCC = 0.8537 ]
+INFO:root:Average loss=0.0792451973240345
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2300 | LCC = 0.8561 | SRCC = 0.8560 ]
+INFO:root:Average loss=0.07189440622161596
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2456 | LCC = 0.8487 | SRCC = 0.8490 ]
+INFO:root:Average loss=0.07597707555844234
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2339 | LCC = 0.8544 | SRCC = 0.8541 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2507 | LCC = 0.8575 | SRCC = 0.8566 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run2/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run2/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28554270ea3746945a70254d3c16fa1891e297fe95b35782b784777732203d60
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run2/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab01059d7aa73d848a850fac307178a448ea569cade2e633670f4f4112b46571
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run2/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.4284714019069305
+INFO:root:
+[Valid][0][UTT][ MSE = 0.4288 | LCC = 0.8063 | SRCC = 0.8101 ]
+INFO:root:Average loss=0.2851868722683344
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3220 | LCC = 0.8339 | SRCC = 0.8350 ]
+INFO:root:Average loss=0.24797301987806955
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3159 | LCC = 0.8416 | SRCC = 0.8408 ]
+INFO:root:Average loss=0.23648603585286018
+INFO:root:
+[Valid][3][UTT][ MSE = 0.3584 | LCC = 0.8436 | SRCC = 0.8444 ]
+INFO:root:Average loss=0.21922305016181406
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3090 | LCC = 0.8522 | SRCC = 0.8512 ]
+INFO:root:Average loss=0.23053583999474844
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2908 | LCC = 0.8516 | SRCC = 0.8494 ]
+INFO:root:Average loss=0.20694475363080317
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2505 | LCC = 0.8596 | SRCC = 0.8567 ]
+INFO:root:Average loss=0.1895792339092646
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2652 | LCC = 0.8616 | SRCC = 0.8609 ]
+INFO:root:Average loss=0.17921728335129908
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2995 | LCC = 0.8570 | SRCC = 0.8550 ]
+INFO:root:Average loss=0.16879250319340292
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2540 | LCC = 0.8542 | SRCC = 0.8517 ]
+INFO:root:Average loss=0.16210990771651268
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2948 | LCC = 0.8568 | SRCC = 0.8579 ]
+INFO:root:Average loss=0.15975914952846673
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3007 | LCC = 0.8533 | SRCC = 0.8534 ]
+INFO:root:Average loss=0.1476324700201169
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2552 | LCC = 0.8567 | SRCC = 0.8565 ]
+INFO:root:Average loss=0.13965487948212868
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2529 | LCC = 0.8545 | SRCC = 0.8533 ]
+INFO:root:Average loss=0.1314195753672184
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2787 | LCC = 0.8546 | SRCC = 0.8535 ]
+INFO:root:Average loss=0.12969911471009254
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2740 | LCC = 0.8548 | SRCC = 0.8525 ]
+INFO:root:Average loss=0.12821087948022744
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2554 | LCC = 0.8507 | SRCC = 0.8486 ]
+INFO:root:Average loss=0.12309678414693245
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2384 | LCC = 0.8510 | SRCC = 0.8487 ]
+INFO:root:Average loss=0.10895248215932113
+INFO:root:
+[Valid][18][UTT][ MSE = 0.3008 | LCC = 0.8464 | SRCC = 0.8450 ]
+INFO:root:Average loss=0.10654697099175209
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2488 | LCC = 0.8468 | SRCC = 0.8454 ]
+INFO:root:Average loss=0.1012338790564965
+INFO:root:
+[Valid][20][UTT][ MSE = 0.3068 | LCC = 0.8422 | SRCC = 0.8417 ]
+INFO:root:Average loss=0.10265656675283726
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2414 | LCC = 0.8506 | SRCC = 0.8495 ]
+INFO:root:Average loss=0.0958816347978054
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2623 | LCC = 0.8450 | SRCC = 0.8436 ]
+INFO:root:Average loss=0.08844794638645954
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2644 | LCC = 0.8445 | SRCC = 0.8442 ]
+INFO:root:Average loss=0.08726239767976296
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2722 | LCC = 0.8456 | SRCC = 0.8444 ]
+INFO:root:Average loss=0.08903845958411694
+INFO:root:
+[Valid][25][UTT][ MSE = 0.3081 | LCC = 0.8481 | SRCC = 0.8483 ]
+INFO:root:Average loss=0.08786079831994496
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2508 | LCC = 0.8497 | SRCC = 0.8485 ]
+INFO:root:Average loss=0.08469467180279586
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2450 | LCC = 0.8485 | SRCC = 0.8476 ]
+INFO:root:Average loss=0.07920091355649325
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2528 | LCC = 0.8470 | SRCC = 0.8465 ]
+INFO:root:Average loss=0.07507136316062549
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2453 | LCC = 0.8480 | SRCC = 0.8474 ]
+INFO:root:Average loss=0.0728788711130619
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2559 | LCC = 0.8486 | SRCC = 0.8482 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2579 | LCC = 0.8554 | SRCC = 0.8532 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run3/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run3/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39e04901f0c7684dec0f71c23996942b5de1561e7b298e98635d26038f918778
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run3/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:817ef44e113973ca2620964878941835dd0cddb7070ab69523e6da0414097fc5
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run3/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.40541908622552186
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3292 | LCC = 0.8130 | SRCC = 0.8154 ]
+INFO:root:Average loss=0.27299155103854644
+INFO:root:
+[Valid][1][UTT][ MSE = 0.2932 | LCC = 0.8382 | SRCC = 0.8358 ]
+INFO:root:Average loss=0.2485059251387914
+INFO:root:
+[Valid][2][UTT][ MSE = 0.2642 | LCC = 0.8462 | SRCC = 0.8457 ]
+INFO:root:Average loss=0.2358058819021934
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2710 | LCC = 0.8525 | SRCC = 0.8533 ]
+INFO:root:Average loss=0.2178884758017002
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2632 | LCC = 0.8556 | SRCC = 0.8525 ]
+INFO:root:Average loss=0.21346342353484568
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2489 | LCC = 0.8528 | SRCC = 0.8503 ]
+INFO:root:Average loss=0.20036793786745805
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2706 | LCC = 0.8604 | SRCC = 0.8579 ]
+INFO:root:Average loss=0.18231598717662004
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2438 | LCC = 0.8625 | SRCC = 0.8604 ]
+INFO:root:Average loss=0.17464523705152366
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2510 | LCC = 0.8567 | SRCC = 0.8542 ]
+INFO:root:Average loss=0.16686499223877221
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2490 | LCC = 0.8642 | SRCC = 0.8619 ]
+INFO:root:Average loss=0.15523532529671988
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2472 | LCC = 0.8595 | SRCC = 0.8564 ]
+INFO:root:Average loss=0.14768393203998223
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2777 | LCC = 0.8592 | SRCC = 0.8579 ]
+INFO:root:Average loss=0.14827872946476325
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2755 | LCC = 0.8584 | SRCC = 0.8579 ]
+INFO:root:Average loss=0.1361596627304187
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2317 | LCC = 0.8554 | SRCC = 0.8531 ]
+INFO:root:Average loss=0.12682292954279825
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2496 | LCC = 0.8557 | SRCC = 0.8531 ]
+INFO:root:Average loss=0.12293366266366763
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2605 | LCC = 0.8584 | SRCC = 0.8576 ]
+INFO:root:Average loss=0.11543400318194659
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2605 | LCC = 0.8524 | SRCC = 0.8505 ]
+INFO:root:Average loss=0.11308515215149292
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2459 | LCC = 0.8544 | SRCC = 0.8535 ]
+INFO:root:Average loss=0.11106220088326015
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2462 | LCC = 0.8513 | SRCC = 0.8499 ]
+INFO:root:Average loss=0.10959643316574585
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2388 | LCC = 0.8552 | SRCC = 0.8537 ]
+INFO:root:Average loss=0.10280151264025615
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2367 | LCC = 0.8572 | SRCC = 0.8561 ]
+INFO:root:Average loss=0.09583375765345035
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2472 | LCC = 0.8525 | SRCC = 0.8525 ]
+INFO:root:Average loss=0.09108715327695394
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2474 | LCC = 0.8538 | SRCC = 0.8519 ]
+INFO:root:Average loss=0.08602023435135682
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2464 | LCC = 0.8517 | SRCC = 0.8498 ]
+INFO:root:Average loss=0.08819783736880009
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2689 | LCC = 0.8464 | SRCC = 0.8452 ]
+INFO:root:Average loss=0.0810032367037657
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2375 | LCC = 0.8509 | SRCC = 0.8496 ]
+INFO:root:Average loss=0.07612811587750912
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2506 | LCC = 0.8500 | SRCC = 0.8486 ]
+INFO:root:Average loss=0.08309972878450002
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2633 | LCC = 0.8494 | SRCC = 0.8484 ]
+INFO:root:Average loss=0.07546157108094448
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2501 | LCC = 0.8498 | SRCC = 0.8478 ]
+INFO:root:Average loss=0.0741928331553936
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2582 | LCC = 0.8511 | SRCC = 0.8502 ]
+INFO:root:Average loss=0.06476368224964692
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2410 | LCC = 0.8492 | SRCC = 0.8491 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2692 | LCC = 0.8569 | SRCC = 0.8593 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run4/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run4/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a7bbb8e0472ccf1fbd917353cb8eda0f47b45934cd5cfc6aeee445c8fe214b0
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run4/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b0f3f4e414f8b12d9d5dcda1231a614be423c83892f5772f68e636a1dd89c78
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer8/run4/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.43741939942806196
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3538 | LCC = 0.7955 | SRCC = 0.7995 ]
+INFO:root:Average loss=0.2892684185734162
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3225 | LCC = 0.8224 | SRCC = 0.8260 ]
+INFO:root:Average loss=0.24687125247258407
+INFO:root:
+[Valid][2][UTT][ MSE = 0.2957 | LCC = 0.8453 | SRCC = 0.8444 ]
+INFO:root:Average loss=0.22453046322633058
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2956 | LCC = 0.8495 | SRCC = 0.8497 ]
+INFO:root:Average loss=0.2161407819352089
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2601 | LCC = 0.8594 | SRCC = 0.8583 ]
+INFO:root:Average loss=0.2075838952874526
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2637 | LCC = 0.8630 | SRCC = 0.8620 ]
+INFO:root:Average loss=0.2106938322003071
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2440 | LCC = 0.8615 | SRCC = 0.8610 ]
+INFO:root:Average loss=0.1833732172082632
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2426 | LCC = 0.8628 | SRCC = 0.8626 ]
+INFO:root:Average loss=0.17070448857087356
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2519 | LCC = 0.8630 | SRCC = 0.8629 ]
+INFO:root:Average loss=0.16309963032985345
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2384 | LCC = 0.8641 | SRCC = 0.8645 ]
+INFO:root:Average loss=0.15788956072468024
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2264 | LCC = 0.8653 | SRCC = 0.8652 ]
+INFO:root:Average loss=0.15759127290967184
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3060 | LCC = 0.8598 | SRCC = 0.8622 ]
+INFO:root:Average loss=0.15851521339171973
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2240 | LCC = 0.8663 | SRCC = 0.8675 ]
+INFO:root:Average loss=0.1394150136755063
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2274 | LCC = 0.8664 | SRCC = 0.8670 ]
+INFO:root:Average loss=0.13565322651694983
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2362 | LCC = 0.8644 | SRCC = 0.8649 ]
+INFO:root:Average loss=0.12542991130015788
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2336 | LCC = 0.8641 | SRCC = 0.8647 ]
+INFO:root:Average loss=0.11608249178299537
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2440 | LCC = 0.8606 | SRCC = 0.8614 ]
+INFO:root:Average loss=0.11595186696220668
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2413 | LCC = 0.8592 | SRCC = 0.8591 ]
+INFO:root:Average loss=0.11024917996464631
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2382 | LCC = 0.8596 | SRCC = 0.8605 ]
+INFO:root:Average loss=0.10416477918624878
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2674 | LCC = 0.8542 | SRCC = 0.8535 ]
+INFO:root:Average loss=0.09718882345045224
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2379 | LCC = 0.8586 | SRCC = 0.8599 ]
+INFO:root:Average loss=0.09196937079421985
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2561 | LCC = 0.8534 | SRCC = 0.8541 ]
+INFO:root:Average loss=0.09043105338246395
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2331 | LCC = 0.8574 | SRCC = 0.8587 ]
+INFO:root:Average loss=0.08995892900304917
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2339 | LCC = 0.8598 | SRCC = 0.8597 ]
+INFO:root:Average loss=0.08405351543273681
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2510 | LCC = 0.8571 | SRCC = 0.8573 ]
+INFO:root:Average loss=0.07768201933075221
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2522 | LCC = 0.8562 | SRCC = 0.8572 ]
+INFO:root:Average loss=0.0780353932044445
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2251 | LCC = 0.8571 | SRCC = 0.8581 ]
+INFO:root:Average loss=0.07864247553814681
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2314 | LCC = 0.8559 | SRCC = 0.8565 ]
+INFO:root:Average loss=0.07323050661346851
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2243 | LCC = 0.8572 | SRCC = 0.8585 ]
+INFO:root:Average loss=0.07265138511474316
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2550 | LCC = 0.8526 | SRCC = 0.8534 ]
+INFO:root:Average loss=0.0646772900930582
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2387 | LCC = 0.8538 | SRCC = 0.8545 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2626 | LCC = 0.8533 | SRCC = 0.8526 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run0/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run0/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8712c99bb3cbfe5a0a5e91f3d2054ea82f6084c0e6ed874a46179e4a3bf6dca8
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run0/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87bc07c53d1a2118dfc2a1a194b372b56b536e961461d15da3d0c288e1571401
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run0/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.40266313346532673
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3359 | LCC = 0.7966 | SRCC = 0.7982 ]
+INFO:root:Average loss=0.29929162313540775
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3918 | LCC = 0.8220 | SRCC = 0.8244 ]
+INFO:root:Average loss=0.2818653656122012
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3207 | LCC = 0.8372 | SRCC = 0.8360 ]
+INFO:root:Average loss=0.263999010125796
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2658 | LCC = 0.8363 | SRCC = 0.8379 ]
+INFO:root:Average loss=0.2569745114216438
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3762 | LCC = 0.8436 | SRCC = 0.8459 ]
+INFO:root:Average loss=0.24703097190612403
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2841 | LCC = 0.8474 | SRCC = 0.8477 ]
+INFO:root:Average loss=0.2624498519759912
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3735 | LCC = 0.8507 | SRCC = 0.8502 ]
+INFO:root:Average loss=0.23869414455615556
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2764 | LCC = 0.8506 | SRCC = 0.8507 ]
+INFO:root:Average loss=0.21669101361662912
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2513 | LCC = 0.8581 | SRCC = 0.8578 ]
+INFO:root:Average loss=0.20718366289750123
+INFO:root:
+[Valid][9][UTT][ MSE = 0.3034 | LCC = 0.8592 | SRCC = 0.8612 ]
+INFO:root:Average loss=0.2161941912311774
+INFO:root:
+[Valid][10][UTT][ MSE = 0.3129 | LCC = 0.8583 | SRCC = 0.8575 ]
+INFO:root:Average loss=0.21269655743470559
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3040 | LCC = 0.8601 | SRCC = 0.8583 ]
+INFO:root:Average loss=0.20141950383400306
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2737 | LCC = 0.8605 | SRCC = 0.8580 ]
+INFO:root:Average loss=0.20354594595921346
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2812 | LCC = 0.8585 | SRCC = 0.8570 ]
+INFO:root:Average loss=0.19188268568653327
+INFO:root:
+[Valid][14][UTT][ MSE = 0.3086 | LCC = 0.8573 | SRCC = 0.8610 ]
+INFO:root:Average loss=0.1888957713276912
+INFO:root:
+[Valid][15][UTT][ MSE = 0.3459 | LCC = 0.8562 | SRCC = 0.8609 ]
+INFO:root:Average loss=0.18730529254445663
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2625 | LCC = 0.8589 | SRCC = 0.8570 ]
+INFO:root:Average loss=0.1779670580648459
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2522 | LCC = 0.8563 | SRCC = 0.8567 ]
+INFO:root:Average loss=0.1712119739789229
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2592 | LCC = 0.8588 | SRCC = 0.8585 ]
+INFO:root:Average loss=0.17276713022818932
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2630 | LCC = 0.8600 | SRCC = 0.8582 ]
+INFO:root:Average loss=0.1593610692100647
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2969 | LCC = 0.8575 | SRCC = 0.8592 ]
+INFO:root:Average loss=0.1659397159058314
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2753 | LCC = 0.8618 | SRCC = 0.8607 ]
+INFO:root:Average loss=0.1549930579196184
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2566 | LCC = 0.8619 | SRCC = 0.8616 ]
+INFO:root:Average loss=0.16745337662406456
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2867 | LCC = 0.8570 | SRCC = 0.8570 ]
+INFO:root:Average loss=0.15673001817403695
+INFO:root:
+[Valid][24][UTT][ MSE = 0.3009 | LCC = 0.8588 | SRCC = 0.8578 ]
+INFO:root:Average loss=0.14925719224489653
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2394 | LCC = 0.8603 | SRCC = 0.8587 ]
+INFO:root:Average loss=0.15450562527164435
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2645 | LCC = 0.8614 | SRCC = 0.8594 ]
+INFO:root:Average loss=0.1366595337406183
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2373 | LCC = 0.8575 | SRCC = 0.8565 ]
+INFO:root:Average loss=0.13738801101079354
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2639 | LCC = 0.8533 | SRCC = 0.8522 ]
+INFO:root:Average loss=0.1267202989413188
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2985 | LCC = 0.8510 | SRCC = 0.8510 ]
+INFO:root:Average loss=0.12687058336077592
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2258 | LCC = 0.8599 | SRCC = 0.8586 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2939 | LCC = 0.8493 | SRCC = 0.8522 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run1/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run1/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c236fa89aa193ebad1e250f365e8baee945fb63a83c1f20b9871743db6e32abb
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run1/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f35b472145f937f13071729146a86316d7127143bfacc1e8e7a56a24b259c011
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run1/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.4648682929766484
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3894 | LCC = 0.7893 | SRCC = 0.7919 ]
+INFO:root:Average loss=0.29751755621952886
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3217 | LCC = 0.8200 | SRCC = 0.8215 ]
+INFO:root:Average loss=0.2782701930174461
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3154 | LCC = 0.8149 | SRCC = 0.8217 ]
+INFO:root:Average loss=0.2735561869847469
+INFO:root:
+[Valid][3][UTT][ MSE = 0.3193 | LCC = 0.8401 | SRCC = 0.8377 ]
+INFO:root:Average loss=0.26198548556138307
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3044 | LCC = 0.8434 | SRCC = 0.8425 ]
+INFO:root:Average loss=0.24024102378350037
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2920 | LCC = 0.8517 | SRCC = 0.8515 ]
+INFO:root:Average loss=0.2412498067968931
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2637 | LCC = 0.8512 | SRCC = 0.8502 ]
+INFO:root:Average loss=0.24265832874255303
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3037 | LCC = 0.8526 | SRCC = 0.8512 ]
+INFO:root:Average loss=0.2187001244762005
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2693 | LCC = 0.8561 | SRCC = 0.8561 ]
+INFO:root:Average loss=0.2090743953982989
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2790 | LCC = 0.8521 | SRCC = 0.8535 ]
+INFO:root:Average loss=0.20778891501518396
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2772 | LCC = 0.8602 | SRCC = 0.8603 ]
+INFO:root:Average loss=0.21923466695424837
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2922 | LCC = 0.8600 | SRCC = 0.8575 ]
+INFO:root:Average loss=0.19139932201076776
+INFO:root:
+[Valid][12][UTT][ MSE = 0.3009 | LCC = 0.8639 | SRCC = 0.8625 ]
+INFO:root:Average loss=0.19710553341951126
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3082 | LCC = 0.8659 | SRCC = 0.8639 ]
+INFO:root:Average loss=0.18678924231193003
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2350 | LCC = 0.8624 | SRCC = 0.8617 ]
+INFO:root:Average loss=0.18032619930230653
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2665 | LCC = 0.8625 | SRCC = 0.8607 ]
+INFO:root:Average loss=0.1813369193711342
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2409 | LCC = 0.8636 | SRCC = 0.8624 ]
+INFO:root:Average loss=0.16888423808492148
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2555 | LCC = 0.8607 | SRCC = 0.8598 ]
+INFO:root:Average loss=0.17025642154308465
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2545 | LCC = 0.8646 | SRCC = 0.8632 ]
+INFO:root:Average loss=0.16329536577447867
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2634 | LCC = 0.8619 | SRCC = 0.8599 ]
+INFO:root:Average loss=0.16247353483086976
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2443 | LCC = 0.8630 | SRCC = 0.8624 ]
+INFO:root:Average loss=0.160555169177361
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2394 | LCC = 0.8619 | SRCC = 0.8615 ]
+INFO:root:Average loss=0.163032844185065
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2617 | LCC = 0.8571 | SRCC = 0.8571 ]
+INFO:root:Average loss=0.15858055183138603
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2323 | LCC = 0.8579 | SRCC = 0.8590 ]
+INFO:root:Average loss=0.14591626756084272
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2573 | LCC = 0.8587 | SRCC = 0.8566 ]
+INFO:root:Average loss=0.14349040169364366
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2601 | LCC = 0.8614 | SRCC = 0.8605 ]
+INFO:root:Average loss=0.13605844850341478
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2414 | LCC = 0.8572 | SRCC = 0.8570 ]
+INFO:root:Average loss=0.1352860598036876
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2451 | LCC = 0.8560 | SRCC = 0.8558 ]
+INFO:root:Average loss=0.13141561834475932
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2441 | LCC = 0.8581 | SRCC = 0.8571 ]
+INFO:root:Average loss=0.1243367326947359
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2485 | LCC = 0.8564 | SRCC = 0.8564 ]
+INFO:root:Average loss=0.12883145486315092
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2562 | LCC = 0.8565 | SRCC = 0.8565 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.3199 | LCC = 0.8519 | SRCC = 0.8508 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run2/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run2/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca08dd9b5e33413b8dfb4bfeb23a7d2e2aeab58e3e5e42db51e17a3a56114e08
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run2/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37a7c12a4e14f45e409ca9e3dea3ed4b67e48065b2e575c091175d872c1969bc
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run2/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.4380430716734666
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3651 | LCC = 0.7911 | SRCC = 0.7934 ]
+INFO:root:Average loss=0.3175906286789821
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3453 | LCC = 0.8137 | SRCC = 0.8168 ]
+INFO:root:Average loss=0.30156727077869266
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3076 | LCC = 0.8227 | SRCC = 0.8246 ]
+INFO:root:Average loss=0.2668433584845983
+INFO:root:
+[Valid][3][UTT][ MSE = 0.3279 | LCC = 0.8348 | SRCC = 0.8328 ]
+INFO:root:Average loss=0.2522684801847507
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3280 | LCC = 0.8371 | SRCC = 0.8353 ]
+INFO:root:Average loss=0.2482475391947306
+INFO:root:
+[Valid][5][UTT][ MSE = 0.3321 | LCC = 0.8457 | SRCC = 0.8466 ]
+INFO:root:Average loss=0.2267345585502111
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2629 | LCC = 0.8489 | SRCC = 0.8479 ]
+INFO:root:Average loss=0.23283594827621412
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2862 | LCC = 0.8523 | SRCC = 0.8527 ]
+INFO:root:Average loss=0.23233816027641296
+INFO:root:
+[Valid][8][UTT][ MSE = 0.3290 | LCC = 0.8566 | SRCC = 0.8573 ]
+INFO:root:Average loss=0.217007499283705
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2853 | LCC = 0.8593 | SRCC = 0.8587 ]
+INFO:root:Average loss=0.21657805335827363
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2816 | LCC = 0.8598 | SRCC = 0.8581 ]
+INFO:root:Average loss=0.21848731402021188
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2686 | LCC = 0.8609 | SRCC = 0.8591 ]
+INFO:root:Average loss=0.2011777832148931
+INFO:root:
+[Valid][12][UTT][ MSE = 0.3169 | LCC = 0.8572 | SRCC = 0.8586 ]
+INFO:root:Average loss=0.18928339083989462
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2622 | LCC = 0.8605 | SRCC = 0.8581 ]
+INFO:root:Average loss=0.19051613677770662
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2824 | LCC = 0.8590 | SRCC = 0.8566 ]
+INFO:root:Average loss=0.18298456187431628
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2672 | LCC = 0.8652 | SRCC = 0.8638 ]
+INFO:root:Average loss=0.17545409223590142
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2493 | LCC = 0.8627 | SRCC = 0.8610 ]
+INFO:root:Average loss=0.1711289882659912
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2479 | LCC = 0.8610 | SRCC = 0.8615 ]
+INFO:root:Average loss=0.1718349771048778
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2877 | LCC = 0.8571 | SRCC = 0.8563 ]
+INFO:root:Average loss=0.16342714390693566
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2460 | LCC = 0.8560 | SRCC = 0.8543 ]
+INFO:root:Average loss=0.17224891751240462
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2750 | LCC = 0.8583 | SRCC = 0.8568 ]
+INFO:root:Average loss=0.1667442812751501
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2911 | LCC = 0.8527 | SRCC = 0.8518 ]
+INFO:root:Average loss=0.15686444708934197
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2725 | LCC = 0.8543 | SRCC = 0.8517 ]
+INFO:root:Average loss=0.14840580427493805
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2502 | LCC = 0.8616 | SRCC = 0.8595 ]
+INFO:root:Average loss=0.15023532681740248
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2550 | LCC = 0.8574 | SRCC = 0.8574 ]
+INFO:root:Average loss=0.13739874433630553
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2442 | LCC = 0.8603 | SRCC = 0.8593 ]
+INFO:root:Average loss=0.13486138492440566
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2634 | LCC = 0.8524 | SRCC = 0.8510 ]
+INFO:root:Average loss=0.13031689574321112
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2684 | LCC = 0.8541 | SRCC = 0.8516 ]
+INFO:root:Average loss=0.13020408506958914
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2748 | LCC = 0.8530 | SRCC = 0.8532 ]
+INFO:root:Average loss=0.13747126618639016
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2592 | LCC = 0.8526 | SRCC = 0.8527 ]
+INFO:root:Average loss=0.12138828176718491
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2682 | LCC = 0.8466 | SRCC = 0.8427 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2706 | LCC = 0.8554 | SRCC = 0.8557 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run3/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run3/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b548a25e71148bd5546559aae6dd1ece4414ac7f19ff6949e0c017e983bbd436
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run3/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61376ce0e5e7483513dca887e6ef4f2198e93a888ed19f707b1657d6d4aae2f2
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run3/train.log

@@ -0,0 +1,99 @@
+INFO:root:Num train speech clips: 4974
+INFO:root:Num val speech clips: 1066
+INFO:root:Num test speech clips: 1066
+INFO:root:Device=cuda
+INFO:root:Average loss=0.5438558818438114
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3763 | LCC = 0.7776 | SRCC = 0.7818 ]
+INFO:root:Average loss=0.31684504716824263
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3293 | LCC = 0.8029 | SRCC = 0.8075 ]
+INFO:root:Average loss=0.29425239715820706
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3285 | LCC = 0.8216 | SRCC = 0.8232 ]
+INFO:root:Average loss=0.2700358713284517
+INFO:root:
+[Valid][3][UTT][ MSE = 0.3121 | LCC = 0.8352 | SRCC = 0.8358 ]
+INFO:root:Average loss=0.26226788205214036
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3116 | LCC = 0.8435 | SRCC = 0.8439 ]
+INFO:root:Average loss=0.25144613820772904
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2722 | LCC = 0.8406 | SRCC = 0.8421 ]
+INFO:root:Average loss=0.24430168592012846
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2788 | LCC = 0.8474 | SRCC = 0.8465 ]
+INFO:root:Average loss=0.23114579151838255
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3113 | LCC = 0.8413 | SRCC = 0.8448 ]
+INFO:root:Average loss=0.22467450759349725
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2778 | LCC = 0.8434 | SRCC = 0.8444 ]
+INFO:root:Average loss=0.23356003715441778
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2907 | LCC = 0.8512 | SRCC = 0.8503 ]
+INFO:root:Average loss=0.21789864087716126
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2630 | LCC = 0.8572 | SRCC = 0.8562 ]
+INFO:root:Average loss=0.2068679734873466
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2605 | LCC = 0.8578 | SRCC = 0.8578 ]
+INFO:root:Average loss=0.2001887424251972
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2562 | LCC = 0.8586 | SRCC = 0.8572 ]
+INFO:root:Average loss=0.1998200210241171
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2862 | LCC = 0.8608 | SRCC = 0.8604 ]
+INFO:root:Average loss=0.1897058802155348
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2510 | LCC = 0.8601 | SRCC = 0.8589 ]
+INFO:root:Average loss=0.18851592544561777
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2587 | LCC = 0.8573 | SRCC = 0.8566 ]
+INFO:root:Average loss=0.1853850024441878
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2748 | LCC = 0.8663 | SRCC = 0.8648 ]
+INFO:root:Average loss=0.17419022197524706
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2541 | LCC = 0.8611 | SRCC = 0.8600 ]
+INFO:root:Average loss=0.17481951577923235
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2906 | LCC = 0.8537 | SRCC = 0.8595 ]
+INFO:root:Average loss=0.16672851966741759
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2557 | LCC = 0.8599 | SRCC = 0.8582 ]
+INFO:root:Average loss=0.16473962548069465
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2713 | LCC = 0.8644 | SRCC = 0.8624 ]
+INFO:root:Average loss=0.1695888197192779
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2281 | LCC = 0.8591 | SRCC = 0.8583 ]
+INFO:root:Average loss=0.15933551925879258
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2526 | LCC = 0.8579 | SRCC = 0.8556 ]
+INFO:root:Average loss=0.14816596855719885
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2568 | LCC = 0.8595 | SRCC = 0.8589 ]
+INFO:root:Average loss=0.15036036523106772
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2616 | LCC = 0.8517 | SRCC = 0.8531 ]
+INFO:root:Average loss=0.1405003054592854
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2301 | LCC = 0.8589 | SRCC = 0.8591 ]
+INFO:root:Average loss=0.140032014499108
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2576 | LCC = 0.8561 | SRCC = 0.8565 ]
+INFO:root:Average loss=0.1408207094631134
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2308 | LCC = 0.8619 | SRCC = 0.8610 ]
+INFO:root:Average loss=0.1366943697898816
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2400 | LCC = 0.8593 | SRCC = 0.8582 ]
+INFO:root:Average loss=0.13014958646052924
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2415 | LCC = 0.8564 | SRCC = 0.8551 ]
+INFO:root:Average loss=0.12681995341793084
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2585 | LCC = 0.8520 | SRCC = 0.8501 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2955 | LCC = 0.8546 | SRCC = 0.8541 ]

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run4/config.gin

@@ -0,0 +1,31 @@
+# Training config on BVCC dataset.
+
+# ------------------------------
+# Architecture
+# ------------------------------
+ProjectionHead:
+    conv_channels = (32, 32)
+    dense_neurons = (64, 32, 1)
+    use_poolings = (True, True)
+    kernel_size = 5
+    pool_size = 5
+    
+# ------------------------------
+# Dataset
+# ------------------------------
+Bvcc:
+    debug = False
+
+get_dataloader:
+    num_workers = 12
+    shuffle = True
+
+# ------------------------------
+# Training loop
+# ------------------------------
+TrainingLoop:
+    num_epochs = 30
+    learning_rate = 1e-4
+    dataset_cls = @Bvcc
+    weight_decay = 0.0
+    batch_size_train = 64

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run4/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc97282c85157499e35d31ae8323f0ebf4c60e9b69d16a3f7b9cf28675c87eeb
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_2epoch/layer9/run4/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e48897cd2e4fd733ede66bbcaa4e114dbc7093a2593f07663faeb8f0f94226ed
+size 2749702