CoolFace
Modelpublic

minhtien2405/wav2vec2-base-north-vi

sourceHugging Facecc-by-nc-4.0updated 1y agoView on Hugging Face
0likes6downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

wav2vec2-base-north-vi

This model is a fine-tuned version of nguyenvulebinh/wav2vec2-base-vietnamese-250h on the nguyendv02/ViMD_Dataset dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.3268
  • —Wer: 0.1288

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • —learning_rate: 0.0003
  • —trainbatchsize: 4
  • —evalbatchsize: 4
  • —seed: 42
  • —gradientaccumulationsteps: 8
  • —totaltrainbatch_size: 32
  • —optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 20
  • —num_epochs: 30
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossWer
1.21710.2164400.35310.2336
0.5270.4327800.36740.1767
0.52450.64911200.34670.1967
0.51890.86541600.36350.1812
0.45891.08112000.34100.1807
0.4421.29752400.33820.1764
0.45281.51392800.34040.1713
1.13441.73023200.34030.1843
0.47261.94663600.33650.1762
0.4752.16234000.34420.1729
0.43452.37864400.33170.1706
0.42492.59504800.31490.1769
0.43852.81145200.32810.1646
1.1193.02705600.34220.1613
0.40823.24346000.34490.1680
1.02623.45986400.34590.1630
0.4113.67616800.31570.1744
0.39223.89257200.33470.1672
0.4084.10827600.32600.1619
0.39224.32458000.32120.1718
0.40024.54098400.32120.2031
0.3994.75738800.32070.1677
0.4184.97369200.33920.1605
0.3795.18939600.31450.1718
0.37295.405710000.32340.1665
0.36755.622010400.32620.1663
0.39415.838410800.33750.1580
0.37636.054111200.31990.1701
0.35676.270511600.32670.1651
0.35216.486812000.31840.1572
0.34646.703212400.33570.1621
0.34136.919512800.30940.1590
0.35637.135213200.33430.1600
0.3567.351613600.32850.1561
0.35997.568014000.32990.1573
0.34977.784314400.32990.1540
0.84698.014800.31990.1549
0.33528.216415200.31730.1670
0.33628.432715600.32560.1550
0.34688.649116000.32410.1615
0.33528.865416400.32190.1580
0.35889.081116800.32560.1534
0.30779.297517200.33840.1564
0.31699.513917600.32720.1495
0.34069.730218000.32490.1553
0.33419.946618400.32500.1531
0.307110.162318800.35220.1493
0.292410.378619200.32010.1553
0.337810.595019600.32380.1528
0.323410.811420000.33440.1555
0.314311.027020400.32690.1558
0.302311.243420800.32200.1565
0.296111.459821200.31870.1721
0.299511.676121600.34640.1527
0.325111.892522000.32250.1539
0.339512.108222400.33600.1539
0.315812.324522800.31730.1496
0.31112.540923200.33020.1473
0.28412.757323600.33990.1500
0.309212.973624000.32450.1509
0.324513.189324400.32810.1503
0.388913.405724800.34190.1495
0.260913.622025200.34030.1480
0.276913.838425600.32640.1483
0.264314.054126000.33150.1574
0.280414.270526400.33570.1489
0.266814.486826800.31860.1456
0.273914.703227200.34070.1492
0.26314.919527600.33060.1491
0.258215.135228000.33070.1473
0.278715.351628400.33100.1520
0.27615.568028800.32700.1486
0.275815.784329200.33700.1471
0.29216.029600.34560.1451
0.264316.216430000.33840.1499
0.270716.432730400.34600.1444
0.260616.649130800.33550.1462
0.255416.865431200.35340.1441
0.248417.081131600.34660.1517
0.23117.297532000.33530.1454
0.250217.513932400.34060.1464
0.257417.730232800.33470.1451
0.233917.946633200.34300.1490
0.230518.162333600.34720.1476
0.241518.378634000.33930.1455
0.257918.595034400.33960.1466
0.25418.811434800.34430.1436
0.229219.027035200.35030.1454
0.235819.243435600.35470.1447
0.23119.459836000.35450.1436
0.254219.676136400.34320.1426
0.246619.892536800.35390.1403
0.236720.108237200.34580.1453
0.219620.324537600.34600.1412
0.212620.540938000.35390.1466
0.225420.757338400.35610.1400
0.230120.973638800.34460.1428
0.215721.189339200.35420.1432
0.215721.405739600.35570.1400
0.217221.622040000.34380.1408
0.196921.838440400.35380.1451
0.200122.054140800.35780.1415
0.2322.270541200.35010.1414
0.228522.486841600.36220.1403
0.204922.703242000.36490.1397
0.222822.919542400.36020.1391
0.239323.135242800.36240.1386
0.211623.351643200.35480.1374
0.25623.568043600.35360.1399
0.215723.784344000.36700.1380
0.215524.044400.35960.1399
0.193824.216444800.36370.1407
0.197224.432745200.37330.1372
0.214224.649145600.35790.1399
0.209224.865446000.36470.1361
0.305925.081146400.37070.1387
0.201425.297546800.37230.1352
0.211625.513947200.36290.1374
0.185425.730247600.36240.1371
0.207425.946648000.38730.1345
0.203426.162348400.36030.1376
0.189326.378648800.37610.1369
0.185926.595049200.37370.1354
0.207626.811449600.35280.1372
0.187927.027050000.36570.1356
0.192727.243450400.36370.1351
0.205927.459850800.37890.1341
0.175127.676151200.36710.1355
0.186427.892551600.36570.1348
0.182228.108252000.36530.1358
0.195528.324552400.37190.1356
0.19428.540952800.37060.1360
0.188828.757353200.37000.1358
0.195428.973653600.36640.1347
0.189729.189354000.36870.1350
0.185129.405754400.36640.1356
0.18229.622054800.36740.1354
0.18729.838455200.36510.1348

Framework versions

  • —Transformers 4.53.0
  • —Pytorch 2.7.1+cu126
  • —Datasets 3.6.0
  • —Tokenizers 0.21.2