Upload SSL Layer MOS checkpoints (part 4)

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer1/run2/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4e4b0ad4d78054b6c88cc06c2a2d60197c98e996998b4b3e23e7ed4e6e9135f
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer1/run2/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:134a826c34e22c6a8fca0a5bf8bd2a487d03cc5cdfd3d50e848e6d54e7424eb2
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer1/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=0.6006987996590443
+INFO:root:
+[Valid][0][UTT][ MSE = 0.4035 | LCC = 0.7393 | SRCC = 0.7421 ]
+INFO:root:Average loss=0.3137065510337169
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3200 | LCC = 0.8046 | SRCC = 0.8113 ]
+INFO:root:Average loss=0.2671936918527652
+INFO:root:
+[Valid][2][UTT][ MSE = 0.2864 | LCC = 0.8191 | SRCC = 0.8263 ]
+INFO:root:Average loss=0.23879283322737768
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2656 | LCC = 0.8378 | SRCC = 0.8368 ]
+INFO:root:Average loss=0.2259096677104632
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2962 | LCC = 0.8344 | SRCC = 0.8401 ]
+INFO:root:Average loss=0.21637474974760643
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2990 | LCC = 0.8439 | SRCC = 0.8460 ]
+INFO:root:Average loss=0.19388003131517997
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2436 | LCC = 0.8568 | SRCC = 0.8550 ]
+INFO:root:Average loss=0.19029025418254045
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2536 | LCC = 0.8575 | SRCC = 0.8558 ]
+INFO:root:Average loss=0.18381341470357698
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2399 | LCC = 0.8521 | SRCC = 0.8534 ]
+INFO:root:Average loss=0.1676634604541155
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2438 | LCC = 0.8603 | SRCC = 0.8580 ]
+INFO:root:Average loss=0.15254133318861327
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2648 | LCC = 0.8552 | SRCC = 0.8573 ]
+INFO:root:Average loss=0.16495537557280981
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2461 | LCC = 0.8611 | SRCC = 0.8600 ]
+INFO:root:Average loss=0.1467983495348539
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2518 | LCC = 0.8587 | SRCC = 0.8578 ]
+INFO:root:Average loss=0.14043838798235625
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2561 | LCC = 0.8601 | SRCC = 0.8600 ]
+INFO:root:Average loss=0.1383707364782309
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2628 | LCC = 0.8569 | SRCC = 0.8559 ]
+INFO:root:Average loss=0.1251800062182622
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2177 | LCC = 0.8619 | SRCC = 0.8603 ]
+INFO:root:Average loss=0.11802030803683476
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3029 | LCC = 0.8581 | SRCC = 0.8580 ]
+INFO:root:Average loss=0.1248860084093534
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2571 | LCC = 0.8634 | SRCC = 0.8607 ]
+INFO:root:Average loss=0.11575037221878003
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2250 | LCC = 0.8592 | SRCC = 0.8571 ]
+INFO:root:Average loss=0.11726610811474995
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2842 | LCC = 0.8518 | SRCC = 0.8499 ]
+INFO:root:Average loss=0.10158162525831124
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2816 | LCC = 0.8557 | SRCC = 0.8544 ]
+INFO:root:Average loss=0.10218472902973492
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2462 | LCC = 0.8570 | SRCC = 0.8523 ]
+INFO:root:Average loss=0.09679935595546013
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2275 | LCC = 0.8571 | SRCC = 0.8536 ]
+INFO:root:Average loss=0.08840126854677995
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2408 | LCC = 0.8610 | SRCC = 0.8589 ]
+INFO:root:Average loss=0.08986751415217534
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2866 | LCC = 0.8540 | SRCC = 0.8524 ]
+INFO:root:Average loss=0.09172978113668087
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2293 | LCC = 0.8573 | SRCC = 0.8543 ]
+INFO:root:Average loss=0.08408539570294894
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2987 | LCC = 0.8526 | SRCC = 0.8531 ]
+INFO:root:Average loss=0.07780154694158298
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2972 | LCC = 0.8521 | SRCC = 0.8531 ]
+INFO:root:Average loss=0.0770007719596227
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2529 | LCC = 0.8601 | SRCC = 0.8591 ]
+INFO:root:Average loss=0.08227077680520523
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2347 | LCC = 0.8619 | SRCC = 0.8597 ]
+INFO:root:Average loss=0.07572061382234097
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2232 | LCC = 0.8599 | SRCC = 0.8564 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2355 | LCC = 0.8661 | SRCC = 0.8644 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer1/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_1epoch/layer1/run3/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c636624e53d1c0afa3183a58996ded79e592b348187fcd15deda685d927d29a0
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer1/run3/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e51b115bb9b02bb48305274413ef23770ee01eae941ba4e752dbb16ad019d3c9
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer1/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=0.5577032099931668
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3960 | LCC = 0.7473 | SRCC = 0.7570 ]
+INFO:root:Average loss=0.3139839286987598
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3711 | LCC = 0.7898 | SRCC = 0.8000 ]
+INFO:root:Average loss=0.26633496028490555
+INFO:root:
+[Valid][2][UTT][ MSE = 0.2800 | LCC = 0.8172 | SRCC = 0.8215 ]
+INFO:root:Average loss=0.23809264447444525
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2644 | LCC = 0.8336 | SRCC = 0.8349 ]
+INFO:root:Average loss=0.22117552180320788
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2862 | LCC = 0.8385 | SRCC = 0.8441 ]
+INFO:root:Average loss=0.22384615643666342
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2635 | LCC = 0.8399 | SRCC = 0.8414 ]
+INFO:root:Average loss=0.196821590169118
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2531 | LCC = 0.8515 | SRCC = 0.8517 ]
+INFO:root:Average loss=0.1905044674491271
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2389 | LCC = 0.8539 | SRCC = 0.8538 ]
+INFO:root:Average loss=0.1764051616191864
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2429 | LCC = 0.8567 | SRCC = 0.8578 ]
+INFO:root:Average loss=0.1762571218304145
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2372 | LCC = 0.8569 | SRCC = 0.8569 ]
+INFO:root:Average loss=0.1647373527670518
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2341 | LCC = 0.8606 | SRCC = 0.8609 ]
+INFO:root:Average loss=0.14897103540790388
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2563 | LCC = 0.8568 | SRCC = 0.8567 ]
+INFO:root:Average loss=0.14665505738976675
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2752 | LCC = 0.8621 | SRCC = 0.8635 ]
+INFO:root:Average loss=0.14233640380776846
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2302 | LCC = 0.8609 | SRCC = 0.8600 ]
+INFO:root:Average loss=0.13174179072181383
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2480 | LCC = 0.8584 | SRCC = 0.8587 ]
+INFO:root:Average loss=0.12609843747356
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2196 | LCC = 0.8611 | SRCC = 0.8603 ]
+INFO:root:Average loss=0.12385736807034566
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2273 | LCC = 0.8609 | SRCC = 0.8597 ]
+INFO:root:Average loss=0.12338605494453357
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2211 | LCC = 0.8622 | SRCC = 0.8624 ]
+INFO:root:Average loss=0.11610402061771123
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2434 | LCC = 0.8593 | SRCC = 0.8601 ]
+INFO:root:Average loss=0.10993184798803085
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2175 | LCC = 0.8614 | SRCC = 0.8616 ]
+INFO:root:Average loss=0.10942104134040001
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2827 | LCC = 0.8608 | SRCC = 0.8638 ]
+INFO:root:Average loss=0.10266295400185463
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2215 | LCC = 0.8581 | SRCC = 0.8576 ]
+INFO:root:Average loss=0.09171454837689033
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2152 | LCC = 0.8604 | SRCC = 0.8597 ]
+INFO:root:Average loss=0.09755968875609912
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2246 | LCC = 0.8598 | SRCC = 0.8579 ]
+INFO:root:Average loss=0.08820604752653684
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2275 | LCC = 0.8575 | SRCC = 0.8584 ]
+INFO:root:Average loss=0.09225147413328671
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2402 | LCC = 0.8590 | SRCC = 0.8582 ]
+INFO:root:Average loss=0.08601127798931721
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2248 | LCC = 0.8610 | SRCC = 0.8600 ]
+INFO:root:Average loss=0.08617788347869347
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2310 | LCC = 0.8594 | SRCC = 0.8588 ]
+INFO:root:Average loss=0.08935599688154
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2203 | LCC = 0.8642 | SRCC = 0.8635 ]
+INFO:root:Average loss=0.08155740830951776
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2393 | LCC = 0.8626 | SRCC = 0.8622 ]
+INFO:root:Average loss=0.0741566596313929
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2308 | LCC = 0.8627 | SRCC = 0.8617 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2517 | LCC = 0.8589 | SRCC = 0.8608 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer1/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_1epoch/layer1/run4/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9998b5a682cc919f1674f782e52a2a8c8441c3bee308a80ce7615d5310d59aab
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer1/run4/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29e10179f34e76d38dccebd4a19ea7ec650289e330f4368acd0419332be11647
+size 2749702

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer1/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=0.5525378086245977
+INFO:root:
+[Valid][0][UTT][ MSE = 0.3801 | LCC = 0.7480 | SRCC = 0.7594 ]
+INFO:root:Average loss=0.3283146917819977
+INFO:root:
+[Valid][1][UTT][ MSE = 0.3379 | LCC = 0.7939 | SRCC = 0.7984 ]
+INFO:root:Average loss=0.2786462990901409
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3004 | LCC = 0.8176 | SRCC = 0.8213 ]
+INFO:root:Average loss=0.24937959588491
+INFO:root:
+[Valid][3][UTT][ MSE = 0.2951 | LCC = 0.8254 | SRCC = 0.8301 ]
+INFO:root:Average loss=0.22587873032077765
+INFO:root:
+[Valid][4][UTT][ MSE = 0.2642 | LCC = 0.8376 | SRCC = 0.8403 ]
+INFO:root:Average loss=0.22421716019893304
+INFO:root:
+[Valid][5][UTT][ MSE = 0.2549 | LCC = 0.8432 | SRCC = 0.8443 ]
+INFO:root:Average loss=0.1979722100763749
+INFO:root:
+[Valid][6][UTT][ MSE = 0.2470 | LCC = 0.8482 | SRCC = 0.8487 ]
+INFO:root:Average loss=0.1896534298474972
+INFO:root:
+[Valid][7][UTT][ MSE = 0.2490 | LCC = 0.8540 | SRCC = 0.8532 ]
+INFO:root:Average loss=0.18755720479366106
+INFO:root:
+[Valid][8][UTT][ MSE = 0.2494 | LCC = 0.8522 | SRCC = 0.8527 ]
+INFO:root:Average loss=0.1769320450914212
+INFO:root:
+[Valid][9][UTT][ MSE = 0.2816 | LCC = 0.8599 | SRCC = 0.8602 ]
+INFO:root:Average loss=0.1606948836109577
+INFO:root:
+[Valid][10][UTT][ MSE = 0.2416 | LCC = 0.8544 | SRCC = 0.8563 ]
+INFO:root:Average loss=0.15922243253160745
+INFO:root:
+[Valid][11][UTT][ MSE = 0.2432 | LCC = 0.8520 | SRCC = 0.8570 ]
+INFO:root:Average loss=0.15271418493909714
+INFO:root:
+[Valid][12][UTT][ MSE = 0.2739 | LCC = 0.8622 | SRCC = 0.8626 ]
+INFO:root:Average loss=0.13986208490454233
+INFO:root:
+[Valid][13][UTT][ MSE = 0.2309 | LCC = 0.8565 | SRCC = 0.8570 ]
+INFO:root:Average loss=0.13474046658629027
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2371 | LCC = 0.8566 | SRCC = 0.8574 ]
+INFO:root:Average loss=0.13050641730809823
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2291 | LCC = 0.8577 | SRCC = 0.8600 ]
+INFO:root:Average loss=0.13568030297756195
+INFO:root:
+[Valid][16][UTT][ MSE = 0.2652 | LCC = 0.8479 | SRCC = 0.8506 ]
+INFO:root:Average loss=0.1308771126354352
+INFO:root:
+[Valid][17][UTT][ MSE = 0.2555 | LCC = 0.8565 | SRCC = 0.8586 ]
+INFO:root:Average loss=0.11732197944552471
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2207 | LCC = 0.8625 | SRCC = 0.8624 ]
+INFO:root:Average loss=0.11015717618358441
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2351 | LCC = 0.8544 | SRCC = 0.8545 ]
+INFO:root:Average loss=0.11742200769293003
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2320 | LCC = 0.8560 | SRCC = 0.8550 ]
+INFO:root:Average loss=0.10351483776974373
+INFO:root:
+[Valid][21][UTT][ MSE = 0.2205 | LCC = 0.8598 | SRCC = 0.8599 ]
+INFO:root:Average loss=0.09843618785723662
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2249 | LCC = 0.8631 | SRCC = 0.8629 ]
+INFO:root:Average loss=0.09560239171752563
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2195 | LCC = 0.8619 | SRCC = 0.8623 ]
+INFO:root:Average loss=0.09047879804021274
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2188 | LCC = 0.8632 | SRCC = 0.8631 ]
+INFO:root:Average loss=0.08644061974990062
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2262 | LCC = 0.8628 | SRCC = 0.8638 ]
+INFO:root:Average loss=0.08795356654968017
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2242 | LCC = 0.8632 | SRCC = 0.8626 ]
+INFO:root:Average loss=0.08254784541443372
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2226 | LCC = 0.8636 | SRCC = 0.8642 ]
+INFO:root:Average loss=0.07980456217550315
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2352 | LCC = 0.8621 | SRCC = 0.8624 ]
+INFO:root:Average loss=0.08143908043320362
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2126 | LCC = 0.8656 | SRCC = 0.8648 ]
+INFO:root:Average loss=0.07162522601011471
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2210 | LCC = 0.8644 | SRCC = 0.8631 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2354 | LCC = 0.8562 | SRCC = 0.8566 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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_1epoch/layer10/run0/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c01b85956dd2c265e049fd027abc53232b8fcac639ae292c57d90a2c30d3e84c
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/run0/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7191ad77c6b4b6e91368e9805e5e5a4605c975fdbb33f86101ff06ad90581abe
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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=0.5794957517049252
+INFO:root:
+[Valid][0][UTT][ MSE = 0.4068 | LCC = 0.7381 | SRCC = 0.7421 ]
+INFO:root:Average loss=0.4189493446013866
+INFO:root:
+[Valid][1][UTT][ MSE = 0.4210 | LCC = 0.7564 | SRCC = 0.7648 ]
+INFO:root:Average loss=0.36668664350723607
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3434 | LCC = 0.7733 | SRCC = 0.7771 ]
+INFO:root:Average loss=0.3401531764330008
+INFO:root:
+[Valid][3][UTT][ MSE = 0.4237 | LCC = 0.7830 | SRCC = 0.7853 ]
+INFO:root:Average loss=0.37762189331727153
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3745 | LCC = 0.7919 | SRCC = 0.7934 ]
+INFO:root:Average loss=0.34444289979262227
+INFO:root:
+[Valid][5][UTT][ MSE = 0.3363 | LCC = 0.7900 | SRCC = 0.7927 ]
+INFO:root:Average loss=0.3259281334586633
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3513 | LCC = 0.7941 | SRCC = 0.7969 ]
+INFO:root:Average loss=0.3640765360532663
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3682 | LCC = 0.7982 | SRCC = 0.8001 ]
+INFO:root:Average loss=0.3193130762531207
+INFO:root:
+[Valid][8][UTT][ MSE = 0.3944 | LCC = 0.8038 | SRCC = 0.8028 ]
+INFO:root:Average loss=0.31694377423861086
+INFO:root:
+[Valid][9][UTT][ MSE = 0.3548 | LCC = 0.8045 | SRCC = 0.8080 ]
+INFO:root:Average loss=0.2900777955849965
+INFO:root:
+[Valid][10][UTT][ MSE = 0.3224 | LCC = 0.8106 | SRCC = 0.8107 ]
+INFO:root:Average loss=0.30088877563293165
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3669 | LCC = 0.8102 | SRCC = 0.8106 ]
+INFO:root:Average loss=0.31550150078076583
+INFO:root:
+[Valid][12][UTT][ MSE = 0.3481 | LCC = 0.8067 | SRCC = 0.8133 ]
+INFO:root:Average loss=0.32858273272330946
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3460 | LCC = 0.8098 | SRCC = 0.8102 ]
+INFO:root:Average loss=0.29860056879428715
+INFO:root:
+[Valid][14][UTT][ MSE = 0.3120 | LCC = 0.8110 | SRCC = 0.8130 ]
+INFO:root:Average loss=0.28414352238178253
+INFO:root:
+[Valid][15][UTT][ MSE = 0.3954 | LCC = 0.8114 | SRCC = 0.8155 ]
+INFO:root:Average loss=0.3009450370684648
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3076 | LCC = 0.8176 | SRCC = 0.8183 ]
+INFO:root:Average loss=0.28250264758482957
+INFO:root:
+[Valid][17][UTT][ MSE = 0.3151 | LCC = 0.8181 | SRCC = 0.8192 ]
+INFO:root:Average loss=0.28335456511913204
+INFO:root:
+[Valid][18][UTT][ MSE = 0.2960 | LCC = 0.8216 | SRCC = 0.8225 ]
+INFO:root:Average loss=0.27502379814783734
+INFO:root:
+[Valid][19][UTT][ MSE = 0.3184 | LCC = 0.8235 | SRCC = 0.8223 ]
+INFO:root:Average loss=0.28518662066795886
+INFO:root:
+[Valid][20][UTT][ MSE = 0.3113 | LCC = 0.8196 | SRCC = 0.8211 ]
+INFO:root:Average loss=0.2678028616385582
+INFO:root:
+[Valid][21][UTT][ MSE = 0.3142 | LCC = 0.8194 | SRCC = 0.8222 ]
+INFO:root:Average loss=0.2691647553673157
+INFO:root:
+[Valid][22][UTT][ MSE = 0.3001 | LCC = 0.8212 | SRCC = 0.8246 ]
+INFO:root:Average loss=0.26632089358873856
+INFO:root:
+[Valid][23][UTT][ MSE = 0.3196 | LCC = 0.8226 | SRCC = 0.8241 ]
+INFO:root:Average loss=0.25906851028020567
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2901 | LCC = 0.8264 | SRCC = 0.8262 ]
+INFO:root:Average loss=0.27062089913166487
+INFO:root:
+[Valid][25][UTT][ MSE = 0.3355 | LCC = 0.8223 | SRCC = 0.8220 ]
+INFO:root:Average loss=0.2600910098124773
+INFO:root:
+[Valid][26][UTT][ MSE = 0.3299 | LCC = 0.8237 | SRCC = 0.8261 ]
+INFO:root:Average loss=0.2629608867260126
+INFO:root:
+[Valid][27][UTT][ MSE = 0.3656 | LCC = 0.8260 | SRCC = 0.8273 ]
+INFO:root:Average loss=0.26037159600319004
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2837 | LCC = 0.8263 | SRCC = 0.8271 ]
+INFO:root:Average loss=0.2623340986095942
+INFO:root:
+[Valid][29][UTT][ MSE = 0.3047 | LCC = 0.8281 | SRCC = 0.8288 ]
+INFO:root:Average loss=0.2514979585241049
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2898 | LCC = 0.8320 | SRCC = 0.8312 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2730 | LCC = 0.8443 | SRCC = 0.8449 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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_1epoch/layer10/run1/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0b794925a03f4cc7a3177521458677cee751004db5a8778837ebcad1347ea42
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/run1/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:308cf8b65f03eac70da9b4a0502d70ca898dd9e9afe1f2f0fd14b80c8d558a4a
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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=0.5769437398665991
+INFO:root:
+[Valid][0][UTT][ MSE = 0.5157 | LCC = 0.7236 | SRCC = 0.7292 ]
+INFO:root:Average loss=0.39619894574085873
+INFO:root:
+[Valid][1][UTT][ MSE = 0.6369 | LCC = 0.7354 | SRCC = 0.7573 ]
+INFO:root:Average loss=0.4234802210942293
+INFO:root:
+[Valid][2][UTT][ MSE = 0.4320 | LCC = 0.7649 | SRCC = 0.7706 ]
+INFO:root:Average loss=0.34623727737328946
+INFO:root:
+[Valid][3][UTT][ MSE = 0.4378 | LCC = 0.7664 | SRCC = 0.7802 ]
+INFO:root:Average loss=0.35806382734041947
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3903 | LCC = 0.7893 | SRCC = 0.7910 ]
+INFO:root:Average loss=0.34265081469829267
+INFO:root:
+[Valid][5][UTT][ MSE = 0.3588 | LCC = 0.7943 | SRCC = 0.7971 ]
+INFO:root:Average loss=0.3497260153675691
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3834 | LCC = 0.7989 | SRCC = 0.7980 ]
+INFO:root:Average loss=0.3079837965659606
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3546 | LCC = 0.7959 | SRCC = 0.8022 ]
+INFO:root:Average loss=0.34615337886871433
+INFO:root:
+[Valid][8][UTT][ MSE = 0.3828 | LCC = 0.8016 | SRCC = 0.8015 ]
+INFO:root:Average loss=0.32590545389132625
+INFO:root:
+[Valid][9][UTT][ MSE = 0.3786 | LCC = 0.8042 | SRCC = 0.8030 ]
+INFO:root:Average loss=0.31244672987705624
+INFO:root:
+[Valid][10][UTT][ MSE = 0.3602 | LCC = 0.8035 | SRCC = 0.8060 ]
+INFO:root:Average loss=0.30032417598443156
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3104 | LCC = 0.8084 | SRCC = 0.8090 ]
+INFO:root:Average loss=0.2947550729299203
+INFO:root:
+[Valid][12][UTT][ MSE = 0.3908 | LCC = 0.8032 | SRCC = 0.8076 ]
+INFO:root:Average loss=0.3025146457247245
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3462 | LCC = 0.8093 | SRCC = 0.8115 ]
+INFO:root:Average loss=0.3008393139029161
+INFO:root:
+[Valid][14][UTT][ MSE = 0.3523 | LCC = 0.8120 | SRCC = 0.8133 ]
+INFO:root:Average loss=0.28342798753426623
+INFO:root:
+[Valid][15][UTT][ MSE = 0.2944 | LCC = 0.8133 | SRCC = 0.8164 ]
+INFO:root:Average loss=0.2859558656047552
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3266 | LCC = 0.8179 | SRCC = 0.8187 ]
+INFO:root:Average loss=0.2910258477697006
+INFO:root:
+[Valid][17][UTT][ MSE = 0.3861 | LCC = 0.8145 | SRCC = 0.8193 ]
+INFO:root:Average loss=0.2872061393199823
+INFO:root:
+[Valid][18][UTT][ MSE = 0.3239 | LCC = 0.8148 | SRCC = 0.8149 ]
+INFO:root:Average loss=0.27876897098926395
+INFO:root:
+[Valid][19][UTT][ MSE = 0.2842 | LCC = 0.8175 | SRCC = 0.8201 ]
+INFO:root:Average loss=0.3017055675005301
+INFO:root:
+[Valid][20][UTT][ MSE = 0.3043 | LCC = 0.8213 | SRCC = 0.8217 ]
+INFO:root:Average loss=0.2770625746402985
+INFO:root:
+[Valid][21][UTT][ MSE = 0.3750 | LCC = 0.8132 | SRCC = 0.8160 ]
+INFO:root:Average loss=0.28191186850651717
+INFO:root:
+[Valid][22][UTT][ MSE = 0.3439 | LCC = 0.8216 | SRCC = 0.8208 ]
+INFO:root:Average loss=0.2829067158775452
+INFO:root:
+[Valid][23][UTT][ MSE = 0.3105 | LCC = 0.8190 | SRCC = 0.8208 ]
+INFO:root:Average loss=0.257286178568999
+INFO:root:
+[Valid][24][UTT][ MSE = 0.3301 | LCC = 0.8209 | SRCC = 0.8244 ]
+INFO:root:Average loss=0.2666809998261623
+INFO:root:
+[Valid][25][UTT][ MSE = 0.3797 | LCC = 0.8139 | SRCC = 0.8200 ]
+INFO:root:Average loss=0.26847826402920943
+INFO:root:
+[Valid][26][UTT][ MSE = 0.3254 | LCC = 0.8207 | SRCC = 0.8233 ]
+INFO:root:Average loss=0.26848746855289507
+INFO:root:
+[Valid][27][UTT][ MSE = 0.3341 | LCC = 0.8252 | SRCC = 0.8263 ]
+INFO:root:Average loss=0.25932069772329086
+INFO:root:
+[Valid][28][UTT][ MSE = 0.4292 | LCC = 0.8149 | SRCC = 0.8217 ]
+INFO:root:Average loss=0.25409679133922625
+INFO:root:
+[Valid][29][UTT][ MSE = 0.3159 | LCC = 0.8272 | SRCC = 0.8283 ]
+INFO:root:Average loss=0.278778137304844
+INFO:root:
+[Valid][30][UTT][ MSE = 0.3522 | LCC = 0.8238 | SRCC = 0.8233 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2762 | LCC = 0.8442 | SRCC = 0.8439 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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_1epoch/layer10/run2/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14ea434751c8c3bfa0f3e31df79a3bce90d9313a5b8fb091f8d53eb36667666e
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/run2/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dafcfe76d0c533f3fc655b9b0c9189a9b110efdd7afe0b1d5388544304b6aa05
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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=0.5942672498715229
+INFO:root:
+[Valid][0][UTT][ MSE = 0.4659 | LCC = 0.7239 | SRCC = 0.7251 ]
+INFO:root:Average loss=0.382832509966997
+INFO:root:
+[Valid][1][UTT][ MSE = 0.4325 | LCC = 0.7512 | SRCC = 0.7651 ]
+INFO:root:Average loss=0.3669925102820763
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3678 | LCC = 0.7743 | SRCC = 0.7784 ]
+INFO:root:Average loss=0.3417179158482796
+INFO:root:
+[Valid][3][UTT][ MSE = 0.4020 | LCC = 0.7845 | SRCC = 0.7916 ]
+INFO:root:Average loss=0.34532631360567534
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3430 | LCC = 0.7916 | SRCC = 0.7938 ]
+INFO:root:Average loss=0.33108560951092303
+INFO:root:
+[Valid][5][UTT][ MSE = 0.3680 | LCC = 0.8033 | SRCC = 0.8022 ]
+INFO:root:Average loss=0.3267887274806316
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3214 | LCC = 0.8031 | SRCC = 0.8039 ]
+INFO:root:Average loss=0.3265558108687401
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3813 | LCC = 0.7969 | SRCC = 0.8028 ]
+INFO:root:Average loss=0.32854782083095646
+INFO:root:
+[Valid][8][UTT][ MSE = 0.3276 | LCC = 0.8058 | SRCC = 0.8069 ]
+INFO:root:Average loss=0.3254155149826637
+INFO:root:
+[Valid][9][UTT][ MSE = 0.3376 | LCC = 0.8090 | SRCC = 0.8091 ]
+INFO:root:Average loss=0.35408404450386
+INFO:root:
+[Valid][10][UTT][ MSE = 0.4920 | LCC = 0.8031 | SRCC = 0.8057 ]
+INFO:root:Average loss=0.312020586278194
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3196 | LCC = 0.8076 | SRCC = 0.8113 ]
+INFO:root:Average loss=0.3042628608452968
+INFO:root:
+[Valid][12][UTT][ MSE = 0.3685 | LCC = 0.8150 | SRCC = 0.8135 ]
+INFO:root:Average loss=0.2875899960979437
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3601 | LCC = 0.8030 | SRCC = 0.8116 ]
+INFO:root:Average loss=0.33370670542503017
+INFO:root:
+[Valid][14][UTT][ MSE = 0.3509 | LCC = 0.8135 | SRCC = 0.8145 ]
+INFO:root:Average loss=0.2957675413061411
+INFO:root:
+[Valid][15][UTT][ MSE = 0.4166 | LCC = 0.8062 | SRCC = 0.8131 ]
+INFO:root:Average loss=0.29341954948046267
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3046 | LCC = 0.8207 | SRCC = 0.8188 ]
+INFO:root:Average loss=0.28014899484622174
+INFO:root:
+[Valid][17][UTT][ MSE = 0.3088 | LCC = 0.8125 | SRCC = 0.8148 ]
+INFO:root:Average loss=0.27201277628923076
+INFO:root:
+[Valid][18][UTT][ MSE = 0.3136 | LCC = 0.8228 | SRCC = 0.8208 ]
+INFO:root:Average loss=0.2936048278441796
+INFO:root:
+[Valid][19][UTT][ MSE = 0.3057 | LCC = 0.8176 | SRCC = 0.8207 ]
+INFO:root:Average loss=0.2729914942995096
+INFO:root:
+[Valid][20][UTT][ MSE = 0.2892 | LCC = 0.8234 | SRCC = 0.8236 ]
+INFO:root:Average loss=0.27233778494290817
+INFO:root:
+[Valid][21][UTT][ MSE = 0.3265 | LCC = 0.8279 | SRCC = 0.8260 ]
+INFO:root:Average loss=0.2711257313688596
+INFO:root:
+[Valid][22][UTT][ MSE = 0.3274 | LCC = 0.8245 | SRCC = 0.8237 ]
+INFO:root:Average loss=0.2759468169548573
+INFO:root:
+[Valid][23][UTT][ MSE = 0.2849 | LCC = 0.8237 | SRCC = 0.8256 ]
+INFO:root:Average loss=0.26103865756438327
+INFO:root:
+[Valid][24][UTT][ MSE = 0.3207 | LCC = 0.8210 | SRCC = 0.8245 ]
+INFO:root:Average loss=0.25575430328265214
+INFO:root:
+[Valid][25][UTT][ MSE = 0.2863 | LCC = 0.8270 | SRCC = 0.8277 ]
+INFO:root:Average loss=0.2723673280232992
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2872 | LCC = 0.8306 | SRCC = 0.8307 ]
+INFO:root:Average loss=0.2772990440328916
+INFO:root:
+[Valid][27][UTT][ MSE = 0.3110 | LCC = 0.8282 | SRCC = 0.8293 ]
+INFO:root:Average loss=0.2672784169897055
+INFO:root:
+[Valid][28][UTT][ MSE = 0.3508 | LCC = 0.8232 | SRCC = 0.8269 ]
+INFO:root:Average loss=0.26590201277763414
+INFO:root:
+[Valid][29][UTT][ MSE = 0.3083 | LCC = 0.8316 | SRCC = 0.8308 ]
+INFO:root:Average loss=0.26222832921223765
+INFO:root:
+[Valid][30][UTT][ MSE = 0.3052 | LCC = 0.8275 | SRCC = 0.8314 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2806 | LCC = 0.8491 | SRCC = 0.8478 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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_1epoch/layer10/run3/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d860d9f3c403d48670dbd17f4af8b7e058e0f4b0886c2ae97eca1eab96414b5
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/run3/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89da6ce2a9ba23d2a7e0d6f4421fe94296eacccb5f70ae333d4d5c2f56a3c620
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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=0.6074026456246009
+INFO:root:
+[Valid][0][UTT][ MSE = 0.5345 | LCC = 0.7236 | SRCC = 0.7236 ]
+INFO:root:Average loss=0.40795303384462994
+INFO:root:
+[Valid][1][UTT][ MSE = 0.6051 | LCC = 0.7298 | SRCC = 0.7651 ]
+INFO:root:Average loss=0.4146918862676009
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3776 | LCC = 0.7675 | SRCC = 0.7758 ]
+INFO:root:Average loss=0.3678717937989113
+INFO:root:
+[Valid][3][UTT][ MSE = 0.3681 | LCC = 0.7764 | SRCC = 0.7816 ]
+INFO:root:Average loss=0.35989341120689344
+INFO:root:
+[Valid][4][UTT][ MSE = 0.4203 | LCC = 0.7881 | SRCC = 0.7889 ]
+INFO:root:Average loss=0.3471070033235428
+INFO:root:
+[Valid][5][UTT][ MSE = 0.4267 | LCC = 0.7826 | SRCC = 0.7925 ]
+INFO:root:Average loss=0.3475740718153807
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3774 | LCC = 0.7944 | SRCC = 0.7979 ]
+INFO:root:Average loss=0.3508263612404848
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3217 | LCC = 0.7878 | SRCC = 0.7992 ]
+INFO:root:Average loss=0.29930401880007523
+INFO:root:
+[Valid][8][UTT][ MSE = 0.3527 | LCC = 0.8042 | SRCC = 0.8045 ]
+INFO:root:Average loss=0.3052817672873155
+INFO:root:
+[Valid][9][UTT][ MSE = 0.3255 | LCC = 0.8051 | SRCC = 0.8069 ]
+INFO:root:Average loss=0.31258848481453383
+INFO:root:
+[Valid][10][UTT][ MSE = 0.3281 | LCC = 0.8078 | SRCC = 0.8107 ]
+INFO:root:Average loss=0.3130040663557175
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3402 | LCC = 0.8110 | SRCC = 0.8119 ]
+INFO:root:Average loss=0.3051484076258464
+INFO:root:
+[Valid][12][UTT][ MSE = 0.3238 | LCC = 0.8137 | SRCC = 0.8136 ]
+INFO:root:Average loss=0.2829606403142978
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3048 | LCC = 0.8171 | SRCC = 0.8172 ]
+INFO:root:Average loss=0.29059502940911514
+INFO:root:
+[Valid][14][UTT][ MSE = 0.2993 | LCC = 0.8183 | SRCC = 0.8187 ]
+INFO:root:Average loss=0.2894028942936506
+INFO:root:
+[Valid][15][UTT][ MSE = 0.3329 | LCC = 0.8189 | SRCC = 0.8209 ]
+INFO:root:Average loss=0.28993573975868714
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3131 | LCC = 0.8155 | SRCC = 0.8177 ]
+INFO:root:Average loss=0.3032311447537862
+INFO:root:
+[Valid][17][UTT][ MSE = 0.3310 | LCC = 0.8186 | SRCC = 0.8225 ]
+INFO:root:Average loss=0.2717400250526575
+INFO:root:
+[Valid][18][UTT][ MSE = 0.3201 | LCC = 0.8210 | SRCC = 0.8209 ]
+INFO:root:Average loss=0.2961716319505985
+INFO:root:
+[Valid][19][UTT][ MSE = 0.3170 | LCC = 0.8213 | SRCC = 0.8207 ]
+INFO:root:Average loss=0.30499528692318845
+INFO:root:
+[Valid][20][UTT][ MSE = 0.3340 | LCC = 0.8228 | SRCC = 0.8212 ]
+INFO:root:Average loss=0.27432053631696945
+INFO:root:
+[Valid][21][UTT][ MSE = 0.3545 | LCC = 0.8208 | SRCC = 0.8224 ]
+INFO:root:Average loss=0.27994152655204135
+INFO:root:
+[Valid][22][UTT][ MSE = 0.3241 | LCC = 0.8251 | SRCC = 0.8243 ]
+INFO:root:Average loss=0.26268342767770475
+INFO:root:
+[Valid][23][UTT][ MSE = 0.3406 | LCC = 0.8240 | SRCC = 0.8251 ]
+INFO:root:Average loss=0.2586102655682808
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2932 | LCC = 0.8276 | SRCC = 0.8269 ]
+INFO:root:Average loss=0.2711377218365669
+INFO:root:
+[Valid][25][UTT][ MSE = 0.3267 | LCC = 0.8289 | SRCC = 0.8272 ]
+INFO:root:Average loss=0.25820626681431746
+INFO:root:
+[Valid][26][UTT][ MSE = 0.2837 | LCC = 0.8290 | SRCC = 0.8287 ]
+INFO:root:Average loss=0.26998501958755344
+INFO:root:
+[Valid][27][UTT][ MSE = 0.3382 | LCC = 0.8252 | SRCC = 0.8273 ]
+INFO:root:Average loss=0.25168586503236723
+INFO:root:
+[Valid][28][UTT][ MSE = 0.2823 | LCC = 0.8258 | SRCC = 0.8267 ]
+INFO:root:Average loss=0.25991330525049794
+INFO:root:
+[Valid][29][UTT][ MSE = 0.3235 | LCC = 0.8293 | SRCC = 0.8282 ]
+INFO:root:Average loss=0.26805608776899487
+INFO:root:
+[Valid][30][UTT][ MSE = 0.3083 | LCC = 0.8309 | SRCC = 0.8311 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2685 | LCC = 0.8476 | SRCC = 0.8455 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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_1epoch/layer10/run4/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6586bbec66a08b780361caee691b4106158753070c36826aefbf55ef00af1e3b
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/run4/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b307556514ede366dfa7056d83cb3bd5e71a3a69cac0b088e08f7eb624482378
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer10/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=0.5919525225957235
+INFO:root:
+[Valid][0][UTT][ MSE = 0.4319 | LCC = 0.7196 | SRCC = 0.7269 ]
+INFO:root:Average loss=0.3931235648118533
+INFO:root:
+[Valid][1][UTT][ MSE = 0.4207 | LCC = 0.7632 | SRCC = 0.7690 ]
+INFO:root:Average loss=0.3702609174144574
+INFO:root:
+[Valid][2][UTT][ MSE = 0.4174 | LCC = 0.7676 | SRCC = 0.7700 ]
+INFO:root:Average loss=0.39132164132136565
+INFO:root:
+[Valid][3][UTT][ MSE = 0.3678 | LCC = 0.7762 | SRCC = 0.7778 ]
+INFO:root:Average loss=0.3711143535299179
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3688 | LCC = 0.7859 | SRCC = 0.7861 ]
+INFO:root:Average loss=0.32915401248595655
+INFO:root:
+[Valid][5][UTT][ MSE = 0.3608 | LCC = 0.7899 | SRCC = 0.7915 ]
+INFO:root:Average loss=0.33503451981605625
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3533 | LCC = 0.7919 | SRCC = 0.7947 ]
+INFO:root:Average loss=0.35452189296483994
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3331 | LCC = 0.8011 | SRCC = 0.8017 ]
+INFO:root:Average loss=0.3177190146003014
+INFO:root:
+[Valid][8][UTT][ MSE = 0.4371 | LCC = 0.7977 | SRCC = 0.8012 ]
+INFO:root:Average loss=0.3301977371940246
+INFO:root:
+[Valid][9][UTT][ MSE = 0.3284 | LCC = 0.8029 | SRCC = 0.8029 ]
+INFO:root:Average loss=0.32799746287174714
+INFO:root:
+[Valid][10][UTT][ MSE = 0.3176 | LCC = 0.8065 | SRCC = 0.8078 ]
+INFO:root:Average loss=0.3009518109835111
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3393 | LCC = 0.8084 | SRCC = 0.8070 ]
+INFO:root:Average loss=0.30156356688493335
+INFO:root:
+[Valid][12][UTT][ MSE = 0.3293 | LCC = 0.8052 | SRCC = 0.8070 ]
+INFO:root:Average loss=0.31853661972742814
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3287 | LCC = 0.8120 | SRCC = 0.8117 ]
+INFO:root:Average loss=0.29982112500912106
+INFO:root:
+[Valid][14][UTT][ MSE = 0.3996 | LCC = 0.8077 | SRCC = 0.8099 ]
+INFO:root:Average loss=0.3019583529004684
+INFO:root:
+[Valid][15][UTT][ MSE = 0.3620 | LCC = 0.8151 | SRCC = 0.8158 ]
+INFO:root:Average loss=0.2804290268283624
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3148 | LCC = 0.8142 | SRCC = 0.8143 ]
+INFO:root:Average loss=0.27405899189985716
+INFO:root:
+[Valid][17][UTT][ MSE = 0.3193 | LCC = 0.8188 | SRCC = 0.8171 ]
+INFO:root:Average loss=0.2996655435134203
+INFO:root:
+[Valid][18][UTT][ MSE = 0.3444 | LCC = 0.8211 | SRCC = 0.8193 ]
+INFO:root:Average loss=0.2868417201515956
+INFO:root:
+[Valid][19][UTT][ MSE = 0.3405 | LCC = 0.8231 | SRCC = 0.8210 ]
+INFO:root:Average loss=0.2829140644425001
+INFO:root:
+[Valid][20][UTT][ MSE = 0.3079 | LCC = 0.8099 | SRCC = 0.8210 ]
+INFO:root:Average loss=0.2662664563992085
+INFO:root:
+[Valid][21][UTT][ MSE = 0.3098 | LCC = 0.8201 | SRCC = 0.8206 ]
+INFO:root:Average loss=0.277060608451183
+INFO:root:
+[Valid][22][UTT][ MSE = 0.2927 | LCC = 0.8234 | SRCC = 0.8237 ]
+INFO:root:Average loss=0.275130726205997
+INFO:root:
+[Valid][23][UTT][ MSE = 0.3379 | LCC = 0.8210 | SRCC = 0.8213 ]
+INFO:root:Average loss=0.27564884340151763
+INFO:root:
+[Valid][24][UTT][ MSE = 0.2925 | LCC = 0.8256 | SRCC = 0.8246 ]
+INFO:root:Average loss=0.264566234480112
+INFO:root:
+[Valid][25][UTT][ MSE = 0.3564 | LCC = 0.8263 | SRCC = 0.8251 ]
+INFO:root:Average loss=0.2591870858883246
+INFO:root:
+[Valid][26][UTT][ MSE = 0.3351 | LCC = 0.8287 | SRCC = 0.8251 ]
+INFO:root:Average loss=0.2626095484846678
+INFO:root:
+[Valid][27][UTT][ MSE = 0.3118 | LCC = 0.8280 | SRCC = 0.8270 ]
+INFO:root:Average loss=0.26458298644194234
+INFO:root:
+[Valid][28][UTT][ MSE = 0.3027 | LCC = 0.8279 | SRCC = 0.8266 ]
+INFO:root:Average loss=0.25217374796286607
+INFO:root:
+[Valid][29][UTT][ MSE = 0.2870 | LCC = 0.8287 | SRCC = 0.8282 ]
+INFO:root:Average loss=0.2601147655111093
+INFO:root:
+[Valid][30][UTT][ MSE = 0.2859 | LCC = 0.8285 | SRCC = 0.8283 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2592 | LCC = 0.8478 | SRCC = 0.8481 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/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_1epoch/layer11/run0/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:156f69c399c4893e577902557dd134bccd75a0faa5d90e4ddcf1a2be9f90e06b
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/run0/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e75a3ff80515a5e9fa48cee3a99b9bf56301b72ac7628a5df52ca6757111f15
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/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=0.6729580377921079
+INFO:root:
+[Valid][0][UTT][ MSE = 0.6144 | LCC = 0.6437 | SRCC = 0.6433 ]
+INFO:root:Average loss=0.4392205820633815
+INFO:root:
+[Valid][1][UTT][ MSE = 0.5890 | LCC = 0.6986 | SRCC = 0.7416 ]
+INFO:root:Average loss=0.42539280347335034
+INFO:root:
+[Valid][2][UTT][ MSE = 0.4591 | LCC = 0.7428 | SRCC = 0.7544 ]
+INFO:root:Average loss=0.4404584815104802
+INFO:root:
+[Valid][3][UTT][ MSE = 0.4064 | LCC = 0.7589 | SRCC = 0.7627 ]
+INFO:root:Average loss=0.39679989505272645
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3677 | LCC = 0.7607 | SRCC = 0.7660 ]
+INFO:root:Average loss=0.35056606641946697
+INFO:root:
+[Valid][5][UTT][ MSE = 0.4482 | LCC = 0.7583 | SRCC = 0.7716 ]
+INFO:root:Average loss=0.3767879463923283
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3750 | LCC = 0.7693 | SRCC = 0.7766 ]
+INFO:root:Average loss=0.3796848373917433
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3820 | LCC = 0.7779 | SRCC = 0.7826 ]
+INFO:root:Average loss=0.3379638451032149
+INFO:root:
+[Valid][8][UTT][ MSE = 0.3881 | LCC = 0.7753 | SRCC = 0.7808 ]
+INFO:root:Average loss=0.3764944424231847
+INFO:root:
+[Valid][9][UTT][ MSE = 0.4350 | LCC = 0.7812 | SRCC = 0.7878 ]
+INFO:root:Average loss=0.33237109180444324
+INFO:root:
+[Valid][10][UTT][ MSE = 0.4639 | LCC = 0.7711 | SRCC = 0.7868 ]
+INFO:root:Average loss=0.32199583603785586
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3349 | LCC = 0.7930 | SRCC = 0.7950 ]
+INFO:root:Average loss=0.32168945345358974
+INFO:root:
+[Valid][12][UTT][ MSE = 0.3347 | LCC = 0.7957 | SRCC = 0.7978 ]
+INFO:root:Average loss=0.33441059310466814
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3615 | LCC = 0.7974 | SRCC = 0.7990 ]
+INFO:root:Average loss=0.32213599998981524
+INFO:root:
+[Valid][14][UTT][ MSE = 0.3278 | LCC = 0.7976 | SRCC = 0.8003 ]
+INFO:root:Average loss=0.3144732795846768
+INFO:root:
+[Valid][15][UTT][ MSE = 0.3469 | LCC = 0.8020 | SRCC = 0.8024 ]
+INFO:root:Average loss=0.30663015311345077
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3637 | LCC = 0.7976 | SRCC = 0.8008 ]
+INFO:root:Average loss=0.29915269101277375
+INFO:root:
+[Valid][17][UTT][ MSE = 0.3612 | LCC = 0.8025 | SRCC = 0.8045 ]
+INFO:root:Average loss=0.2893740036166631
+INFO:root:
+[Valid][18][UTT][ MSE = 0.3567 | LCC = 0.8006 | SRCC = 0.8061 ]
+INFO:root:Average loss=0.311808311404326
+INFO:root:
+[Valid][19][UTT][ MSE = 0.3223 | LCC = 0.8037 | SRCC = 0.8075 ]
+INFO:root:Average loss=0.29704856891662645
+INFO:root:
+[Valid][20][UTT][ MSE = 0.3461 | LCC = 0.8105 | SRCC = 0.8098 ]
+INFO:root:Average loss=0.2921784422718562
+INFO:root:
+[Valid][21][UTT][ MSE = 0.3170 | LCC = 0.8082 | SRCC = 0.8084 ]
+INFO:root:Average loss=0.30228491605092317
+INFO:root:
+[Valid][22][UTT][ MSE = 0.4087 | LCC = 0.8081 | SRCC = 0.8082 ]
+INFO:root:Average loss=0.2978213614760301
+INFO:root:
+[Valid][23][UTT][ MSE = 0.4036 | LCC = 0.7952 | SRCC = 0.8056 ]
+INFO:root:Average loss=0.33107572125318724
+INFO:root:
+[Valid][24][UTT][ MSE = 0.3237 | LCC = 0.8100 | SRCC = 0.8101 ]
+INFO:root:Average loss=0.29745176358100694
+INFO:root:
+[Valid][25][UTT][ MSE = 0.3308 | LCC = 0.8117 | SRCC = 0.8119 ]
+INFO:root:Average loss=0.28526496486021924
+INFO:root:
+[Valid][26][UTT][ MSE = 0.3792 | LCC = 0.8086 | SRCC = 0.8102 ]
+INFO:root:Average loss=0.30388311869823015
+INFO:root:
+[Valid][27][UTT][ MSE = 0.3872 | LCC = 0.8080 | SRCC = 0.8119 ]
+INFO:root:Average loss=0.2939731801549594
+INFO:root:
+[Valid][28][UTT][ MSE = 0.3466 | LCC = 0.8124 | SRCC = 0.8131 ]
+INFO:root:Average loss=0.2995765264599751
+INFO:root:
+[Valid][29][UTT][ MSE = 0.3417 | LCC = 0.8034 | SRCC = 0.8077 ]
+INFO:root:Average loss=0.3072073830243869
+INFO:root:
+[Valid][30][UTT][ MSE = 0.3827 | LCC = 0.8101 | SRCC = 0.8120 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.3347 | LCC = 0.8324 | SRCC = 0.8372 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/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_1epoch/layer11/run1/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:929c145cc02debf9c0538911e79d5d5bde07d8f418d1e5ee745398b2ed48e2c2
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/run1/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab70190b6c141dc788dcc38b53d068e6e90332a665c68b428cca83f8487cbb30
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/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=0.6581237021929178
+INFO:root:
+[Valid][0][UTT][ MSE = 0.6564 | LCC = 0.6689 | SRCC = 0.6788 ]
+INFO:root:Average loss=0.42959490991555727
+INFO:root:
+[Valid][1][UTT][ MSE = 0.4299 | LCC = 0.7417 | SRCC = 0.7464 ]
+INFO:root:Average loss=0.37951474617689085
+INFO:root:
+[Valid][2][UTT][ MSE = 0.4261 | LCC = 0.7583 | SRCC = 0.7603 ]
+INFO:root:Average loss=0.39183056889436185
+INFO:root:
+[Valid][3][UTT][ MSE = 0.3669 | LCC = 0.7610 | SRCC = 0.7696 ]
+INFO:root:Average loss=0.4422425774809642
+INFO:root:
+[Valid][4][UTT][ MSE = 0.4534 | LCC = 0.7616 | SRCC = 0.7717 ]
+INFO:root:Average loss=0.3932936424628282
+INFO:root:
+[Valid][5][UTT][ MSE = 0.6588 | LCC = 0.7383 | SRCC = 0.7689 ]
+INFO:root:Average loss=0.3543844696802971
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3613 | LCC = 0.7733 | SRCC = 0.7796 ]
+INFO:root:Average loss=0.36080430161494476
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3836 | LCC = 0.7860 | SRCC = 0.7873 ]
+INFO:root:Average loss=0.38030731773529297
+INFO:root:
+[Valid][8][UTT][ MSE = 0.4231 | LCC = 0.7775 | SRCC = 0.7821 ]
+INFO:root:Average loss=0.3319390423786946
+INFO:root:
+[Valid][9][UTT][ MSE = 0.3978 | LCC = 0.7770 | SRCC = 0.7874 ]
+INFO:root:Average loss=0.3409799784421921
+INFO:root:
+[Valid][10][UTT][ MSE = 0.4475 | LCC = 0.7822 | SRCC = 0.7921 ]
+INFO:root:Average loss=0.3385501202100363
+INFO:root:
+[Valid][11][UTT][ MSE = 0.5145 | LCC = 0.7766 | SRCC = 0.7912 ]
+INFO:root:Average loss=0.34339119092776227
+INFO:root:
+[Valid][12][UTT][ MSE = 0.4081 | LCC = 0.7830 | SRCC = 0.7937 ]
+INFO:root:Average loss=0.3401635623513124
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3651 | LCC = 0.7977 | SRCC = 0.7993 ]
+INFO:root:Average loss=0.31414138640348727
+INFO:root:
+[Valid][14][UTT][ MSE = 0.4214 | LCC = 0.7905 | SRCC = 0.7952 ]
+INFO:root:Average loss=0.3529744694630305
+INFO:root:
+[Valid][15][UTT][ MSE = 0.3782 | LCC = 0.8015 | SRCC = 0.8014 ]
+INFO:root:Average loss=0.31871499369541806
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3447 | LCC = 0.7906 | SRCC = 0.7999 ]
+INFO:root:Average loss=0.29900134335725737
+INFO:root:
+[Valid][17][UTT][ MSE = 0.3306 | LCC = 0.8037 | SRCC = 0.8041 ]
+INFO:root:Average loss=0.3000396258938007
+INFO:root:
+[Valid][18][UTT][ MSE = 0.3486 | LCC = 0.8065 | SRCC = 0.8069 ]
+INFO:root:Average loss=0.30084676047166187
+INFO:root:
+[Valid][19][UTT][ MSE = 0.3331 | LCC = 0.8064 | SRCC = 0.8087 ]
+INFO:root:Average loss=0.3170011144800064
+INFO:root:
+[Valid][20][UTT][ MSE = 0.3987 | LCC = 0.7968 | SRCC = 0.8060 ]
+INFO:root:Average loss=0.3147989662411885
+INFO:root:
+[Valid][21][UTT][ MSE = 0.3197 | LCC = 0.8078 | SRCC = 0.8096 ]
+INFO:root:Average loss=0.2965874480895507
+INFO:root:
+[Valid][22][UTT][ MSE = 0.3131 | LCC = 0.8099 | SRCC = 0.8112 ]
+INFO:root:Average loss=0.32708879101734895
+INFO:root:
+[Valid][23][UTT][ MSE = 0.3991 | LCC = 0.8095 | SRCC = 0.8090 ]
+INFO:root:Average loss=0.3183735649173076
+INFO:root:
+[Valid][24][UTT][ MSE = 0.3126 | LCC = 0.8075 | SRCC = 0.8103 ]
+INFO:root:Average loss=0.31792674079919475
+INFO:root:
+[Valid][25][UTT][ MSE = 0.3612 | LCC = 0.8132 | SRCC = 0.8122 ]
+INFO:root:Average loss=0.2930086820553511
+INFO:root:
+[Valid][26][UTT][ MSE = 0.4721 | LCC = 0.7961 | SRCC = 0.8079 ]
+INFO:root:Average loss=0.28792792768814623
+INFO:root:
+[Valid][27][UTT][ MSE = 0.2961 | LCC = 0.8119 | SRCC = 0.8141 ]
+INFO:root:Average loss=0.30539400722735965
+INFO:root:
+[Valid][28][UTT][ MSE = 0.3561 | LCC = 0.8113 | SRCC = 0.8112 ]
+INFO:root:Average loss=0.30708794162059444
+INFO:root:
+[Valid][29][UTT][ MSE = 0.4237 | LCC = 0.8134 | SRCC = 0.8139 ]
+INFO:root:Average loss=0.3381939790187738
+INFO:root:
+[Valid][30][UTT][ MSE = 0.3813 | LCC = 0.8095 | SRCC = 0.8093 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.4106 | LCC = 0.8301 | SRCC = 0.8366 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/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_1epoch/layer11/run2/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:822f4fa1222f31160b4d2d84bc98404560288f94f7045582125a5bacfd9a26bb
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/run2/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b181cd74d514ee4d068470f01e23bfce0051fe5c22da7dcee1501ef313356b8
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/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=0.6874579221774371
+INFO:root:
+[Valid][0][UTT][ MSE = 0.5044 | LCC = 0.6915 | SRCC = 0.6913 ]
+INFO:root:Average loss=0.4680313211985124
+INFO:root:
+[Valid][1][UTT][ MSE = 0.4324 | LCC = 0.7318 | SRCC = 0.7362 ]
+INFO:root:Average loss=0.37523861649708873
+INFO:root:
+[Valid][2][UTT][ MSE = 0.4084 | LCC = 0.7505 | SRCC = 0.7559 ]
+INFO:root:Average loss=0.3899931791119086
+INFO:root:
+[Valid][3][UTT][ MSE = 0.3897 | LCC = 0.7627 | SRCC = 0.7641 ]
+INFO:root:Average loss=0.3557039323525551
+INFO:root:
+[Valid][4][UTT][ MSE = 0.4442 | LCC = 0.7627 | SRCC = 0.7732 ]
+INFO:root:Average loss=0.3772805075232799
+INFO:root:
+[Valid][5][UTT][ MSE = 0.3816 | LCC = 0.7775 | SRCC = 0.7786 ]
+INFO:root:Average loss=0.3755926938775258
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3889 | LCC = 0.7840 | SRCC = 0.7837 ]
+INFO:root:Average loss=0.33382165279143894
+INFO:root:
+[Valid][7][UTT][ MSE = 0.4150 | LCC = 0.7837 | SRCC = 0.7876 ]
+INFO:root:Average loss=0.35614499105856967
+INFO:root:
+[Valid][8][UTT][ MSE = 0.3999 | LCC = 0.7898 | SRCC = 0.7896 ]
+INFO:root:Average loss=0.31742671514168763
+INFO:root:
+[Valid][9][UTT][ MSE = 0.3383 | LCC = 0.7871 | SRCC = 0.7906 ]
+INFO:root:Average loss=0.3147903044636433
+INFO:root:
+[Valid][10][UTT][ MSE = 0.3937 | LCC = 0.7951 | SRCC = 0.7972 ]
+INFO:root:Average loss=0.32551616258346117
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3781 | LCC = 0.7992 | SRCC = 0.7993 ]
+INFO:root:Average loss=0.34049213257355565
+INFO:root:
+[Valid][12][UTT][ MSE = 0.4585 | LCC = 0.7973 | SRCC = 0.8017 ]
+INFO:root:Average loss=0.3466266693595128
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3762 | LCC = 0.7966 | SRCC = 0.7947 ]
+INFO:root:Average loss=0.30925640158164197
+INFO:root:
+[Valid][14][UTT][ MSE = 0.4061 | LCC = 0.8015 | SRCC = 0.8014 ]
+INFO:root:Average loss=0.30770480995758986
+INFO:root:
+[Valid][15][UTT][ MSE = 0.3715 | LCC = 0.8040 | SRCC = 0.8031 ]
+INFO:root:Average loss=0.29292874401196456
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3240 | LCC = 0.8037 | SRCC = 0.8033 ]
+INFO:root:Average loss=0.2952551343120061
+INFO:root:
+[Valid][17][UTT][ MSE = 0.3670 | LCC = 0.8071 | SRCC = 0.8074 ]
+INFO:root:Average loss=0.33644014539626926
+INFO:root:
+[Valid][18][UTT][ MSE = 0.3722 | LCC = 0.8059 | SRCC = 0.8057 ]
+INFO:root:Average loss=0.32253950394881076
+INFO:root:
+[Valid][19][UTT][ MSE = 0.3672 | LCC = 0.8078 | SRCC = 0.8087 ]
+INFO:root:Average loss=0.30008579389407086
+INFO:root:
+[Valid][20][UTT][ MSE = 0.3328 | LCC = 0.8028 | SRCC = 0.8031 ]
+INFO:root:Average loss=0.29782283707306934
+INFO:root:
+[Valid][21][UTT][ MSE = 0.3807 | LCC = 0.8116 | SRCC = 0.8104 ]
+INFO:root:Average loss=0.29301549819035405
+INFO:root:
+[Valid][22][UTT][ MSE = 0.4340 | LCC = 0.8114 | SRCC = 0.8106 ]
+INFO:root:Average loss=0.3072131071717311
+INFO:root:
+[Valid][23][UTT][ MSE = 0.3356 | LCC = 0.8097 | SRCC = 0.8093 ]
+INFO:root:Average loss=0.28786266098419827
+INFO:root:
+[Valid][24][UTT][ MSE = 0.3029 | LCC = 0.8053 | SRCC = 0.8075 ]
+INFO:root:Average loss=0.29310321769653225
+INFO:root:
+[Valid][25][UTT][ MSE = 0.3536 | LCC = 0.8119 | SRCC = 0.8122 ]
+INFO:root:Average loss=0.28970870776818347
+INFO:root:
+[Valid][26][UTT][ MSE = 0.3620 | LCC = 0.8040 | SRCC = 0.8074 ]
+INFO:root:Average loss=0.27608862633888537
+INFO:root:
+[Valid][27][UTT][ MSE = 0.3451 | LCC = 0.8156 | SRCC = 0.8139 ]
+INFO:root:Average loss=0.27924639750749636
+INFO:root:
+[Valid][28][UTT][ MSE = 0.3395 | LCC = 0.8153 | SRCC = 0.8148 ]
+INFO:root:Average loss=0.2864933799092586
+INFO:root:
+[Valid][29][UTT][ MSE = 0.3340 | LCC = 0.8154 | SRCC = 0.8129 ]
+INFO:root:Average loss=0.2967428853496527
+INFO:root:
+[Valid][30][UTT][ MSE = 0.3296 | LCC = 0.8172 | SRCC = 0.8164 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.3042 | LCC = 0.8362 | SRCC = 0.8360 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/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_1epoch/layer11/run3/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31ebfbf93784f981fb56f1903b11bf73d163febc6411afda3ebd1b5c42b83dfd
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/run3/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b8c4057199440db0745706bdbed3a1989bac63d381646c6ff5be876468bdf76
+size 2749638

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/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=0.696067886474805
+INFO:root:
+[Valid][0][UTT][ MSE = 0.5179 | LCC = 0.6556 | SRCC = 0.6590 ]
+INFO:root:Average loss=0.44739482761957705
+INFO:root:
+[Valid][1][UTT][ MSE = 0.5517 | LCC = 0.7096 | SRCC = 0.7462 ]
+INFO:root:Average loss=0.3830511042704949
+INFO:root:
+[Valid][2][UTT][ MSE = 0.3875 | LCC = 0.7497 | SRCC = 0.7602 ]
+INFO:root:Average loss=0.38545174476427907
+INFO:root:
+[Valid][3][UTT][ MSE = 0.3973 | LCC = 0.7579 | SRCC = 0.7660 ]
+INFO:root:Average loss=0.38210188731169087
+INFO:root:
+[Valid][4][UTT][ MSE = 0.3735 | LCC = 0.7675 | SRCC = 0.7702 ]
+INFO:root:Average loss=0.4005351104797461
+INFO:root:
+[Valid][5][UTT][ MSE = 0.3601 | LCC = 0.7682 | SRCC = 0.7728 ]
+INFO:root:Average loss=0.33564635003224397
+INFO:root:
+[Valid][6][UTT][ MSE = 0.3727 | LCC = 0.7756 | SRCC = 0.7815 ]
+INFO:root:Average loss=0.35139848330082035
+INFO:root:
+[Valid][7][UTT][ MSE = 0.3497 | LCC = 0.7824 | SRCC = 0.7838 ]
+INFO:root:Average loss=0.3481192625103853
+INFO:root:
+[Valid][8][UTT][ MSE = 0.3655 | LCC = 0.7852 | SRCC = 0.7878 ]
+INFO:root:Average loss=0.34001534183820087
+INFO:root:
+[Valid][9][UTT][ MSE = 0.3633 | LCC = 0.7868 | SRCC = 0.7874 ]
+INFO:root:Average loss=0.3134442233504393
+INFO:root:
+[Valid][10][UTT][ MSE = 0.3377 | LCC = 0.7893 | SRCC = 0.7922 ]
+INFO:root:Average loss=0.3428114320223148
+INFO:root:
+[Valid][11][UTT][ MSE = 0.3650 | LCC = 0.7949 | SRCC = 0.7955 ]
+INFO:root:Average loss=0.3338470262212631
+INFO:root:
+[Valid][12][UTT][ MSE = 0.4090 | LCC = 0.7799 | SRCC = 0.7894 ]
+INFO:root:Average loss=0.3234367775611388
+INFO:root:
+[Valid][13][UTT][ MSE = 0.3352 | LCC = 0.7934 | SRCC = 0.7966 ]
+INFO:root:Average loss=0.32435524520965725
+INFO:root:
+[Valid][14][UTT][ MSE = 0.3531 | LCC = 0.7989 | SRCC = 0.8006 ]
+INFO:root:Average loss=0.36163984678494626
+INFO:root:
+[Valid][15][UTT][ MSE = 0.4671 | LCC = 0.7902 | SRCC = 0.7992 ]
+INFO:root:Average loss=0.32743537655243504
+INFO:root:
+[Valid][16][UTT][ MSE = 0.3657 | LCC = 0.7930 | SRCC = 0.7970 ]
+INFO:root:Average loss=0.320269026244298
+INFO:root:
+[Valid][17][UTT][ MSE = 0.3388 | LCC = 0.8004 | SRCC = 0.8021 ]
+INFO:root:Average loss=0.3269584895326541
+INFO:root:
+[Valid][18][UTT][ MSE = 0.3997 | LCC = 0.8004 | SRCC = 0.8007 ]
+INFO:root:Average loss=0.3281559944152832
+INFO:root:
+[Valid][19][UTT][ MSE = 0.3254 | LCC = 0.7997 | SRCC = 0.8033 ]
+INFO:root:Average loss=0.31543307598584736
+INFO:root:
+[Valid][20][UTT][ MSE = 0.3537 | LCC = 0.8034 | SRCC = 0.8037 ]
+INFO:root:Average loss=0.2931396451133948
+INFO:root:
+[Valid][21][UTT][ MSE = 0.3467 | LCC = 0.7998 | SRCC = 0.8057 ]
+INFO:root:Average loss=0.3020174054381175
+INFO:root:
+[Valid][22][UTT][ MSE = 0.3369 | LCC = 0.8035 | SRCC = 0.8027 ]
+INFO:root:Average loss=0.2944817732159908
+INFO:root:
+[Valid][23][UTT][ MSE = 0.3415 | LCC = 0.8067 | SRCC = 0.8068 ]
+INFO:root:Average loss=0.2979569958570676
+INFO:root:
+[Valid][24][UTT][ MSE = 0.3795 | LCC = 0.7986 | SRCC = 0.8057 ]
+INFO:root:Average loss=0.2853924439121515
+INFO:root:
+[Valid][25][UTT][ MSE = 0.3471 | LCC = 0.8087 | SRCC = 0.8097 ]
+INFO:root:Average loss=0.292974207454767
+INFO:root:
+[Valid][26][UTT][ MSE = 0.3312 | LCC = 0.8116 | SRCC = 0.8094 ]
+INFO:root:Average loss=0.3221439260702867
+INFO:root:
+[Valid][27][UTT][ MSE = 0.3133 | LCC = 0.8106 | SRCC = 0.8108 ]
+INFO:root:Average loss=0.29090672654983324
+INFO:root:
+[Valid][28][UTT][ MSE = 0.3578 | LCC = 0.8103 | SRCC = 0.8110 ]
+INFO:root:Average loss=0.2899026366380545
+INFO:root:
+[Valid][29][UTT][ MSE = 0.3106 | LCC = 0.8123 | SRCC = 0.8120 ]
+INFO:root:Average loss=0.27947496011471135
+INFO:root:
+[Valid][30][UTT][ MSE = 0.3250 | LCC = 0.8126 | SRCC = 0.8104 ]
+INFO:root:
+[Test][31][UTT][ MSE = 0.2931 | LCC = 0.8325 | SRCC = 0.8324 ]

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/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_1epoch/layer11/run4/model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae7fabacf8e288a5b8d4bac4475ba91aee64a11b3e63d26d81b6b1e67e9a297d
+size 2747979

results_bvcc/w2v2_base_bvcc_finetuned_1epoch/layer11/run4/model_best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dc7dcb0f4f70ce33911cd87e0d22e8a006d91d0c765689e1557ed97f2549b2b
+size 2749638