marczenko's picture
SpeechResearch/wtimit-base-normal-all-nofreeze
692eb53 verified
metadata
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
  - generated_from_trainer
datasets:
  - wtimit_asr
metrics:
  - wer
model-index:
  - name: wtimit-base-normal-all-nofreeze
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: wtimit_asr
          type: wtimit_asr
          config: clean
          split: None
          args: clean
        metrics:
          - name: Wer
            type: wer
            value: 0.09987953700309014

wtimit-base-normal-all-nofreeze

This model is a fine-tuned version of facebook/wav2vec2-base on the wtimit_asr dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3190
  • Wer: 0.0999

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: 0.0001
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 1000
  • num_epochs: 50
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
1.5076 0.4 1000 1.1220 0.6793
0.4102 0.81 2000 0.7851 0.4338
0.2278 1.21 3000 0.6897 0.3203
0.1723 1.61 4000 0.5668 0.2890
0.1407 2.02 5000 0.4399 0.2362
0.117 2.42 6000 0.4853 0.2508
0.098 2.83 7000 0.6732 0.2871
0.0862 3.23 8000 0.5802 0.2680
0.0806 3.63 9000 0.4730 0.2488
0.0706 4.04 10000 0.4001 0.1953
0.061 4.44 11000 0.4108 0.1971
0.063 4.84 12000 0.4544 0.2056
0.0527 5.25 13000 0.4235 0.1938
0.049 5.65 14000 0.4375 0.2054
0.0489 6.06 15000 0.5451 0.2522
0.0473 6.46 16000 0.3939 0.1868
0.0442 6.86 17000 0.5662 0.2548
0.0428 7.27 18000 0.6695 0.2755
0.0379 7.67 19000 0.3929 0.1947
0.0398 8.07 20000 0.4446 0.2066
0.0336 8.48 21000 0.5409 0.2260
0.0316 8.88 22000 0.3819 0.1715
0.0322 9.29 23000 0.3861 0.1711
0.0352 9.69 24000 0.4063 0.1728
0.0315 10.09 25000 0.4992 0.2146
0.0254 10.5 26000 0.5838 0.2158
0.0243 10.9 27000 0.3458 0.1523
0.0245 11.3 28000 0.5121 0.1953
0.0231 11.71 29000 0.3773 0.1616
0.0202 12.11 30000 0.4110 0.1715
0.0261 12.52 31000 0.5376 0.2116
0.0243 12.92 32000 0.4066 0.1569
0.0201 13.32 33000 0.5944 0.2276
0.0211 13.73 34000 0.4670 0.1997
0.0249 14.13 35000 0.5521 0.2254
0.021 14.53 36000 0.4602 0.2061
0.0169 14.94 37000 0.4870 0.1690
0.0184 15.34 38000 0.6038 0.2208
0.0207 15.74 39000 0.5266 0.2068
0.0209 16.15 40000 0.5197 0.2083
0.0175 16.55 41000 0.5074 0.1927
0.0164 16.96 42000 0.4594 0.1615
0.0164 17.36 43000 0.2956 0.1151
0.0142 17.76 44000 0.3834 0.1580
0.0139 18.17 45000 0.5316 0.2175
0.0181 18.57 46000 0.5226 0.1890
0.0159 18.97 47000 0.4914 0.1689
0.0127 19.38 48000 0.5454 0.1957
0.0136 19.78 49000 0.5530 0.2172
0.0129 20.19 50000 0.6980 0.2636
0.0131 20.59 51000 0.3984 0.1379
0.0123 20.99 52000 0.4925 0.1843
0.0095 21.4 53000 0.5367 0.1931
0.0124 21.8 54000 0.4299 0.1763
0.0115 22.2 55000 0.4797 0.1803
0.0136 22.61 56000 0.6638 0.2300
0.0121 23.01 57000 0.4292 0.1530
0.0097 23.42 58000 0.4064 0.1520
0.0143 23.82 59000 0.4691 0.1771
0.0092 24.22 60000 0.5134 0.2009
0.0097 24.63 61000 0.6165 0.2281
0.0078 25.03 62000 0.4828 0.1863
0.0114 25.43 63000 0.4817 0.1868
0.0089 25.84 64000 0.5137 0.2003
0.0083 26.24 65000 0.4194 0.1524
0.01 26.65 66000 0.3416 0.1332
0.0102 27.05 67000 0.3834 0.1475
0.0076 27.45 68000 0.3390 0.1277
0.0085 27.86 69000 0.4708 0.1843
0.0074 28.26 70000 0.4434 0.1530
0.0078 28.66 71000 0.2942 0.1104
0.0075 29.07 72000 0.3623 0.1442
0.0066 29.47 73000 0.4709 0.1547
0.0073 29.87 74000 0.5198 0.1750
0.0056 30.28 75000 0.3083 0.1211
0.0066 30.68 76000 0.3204 0.1243
0.0048 31.09 77000 0.3713 0.1326
0.0047 31.49 78000 0.3121 0.1018
0.0066 31.89 79000 0.4510 0.1473
0.0053 32.3 80000 0.3599 0.1130
0.0058 32.7 81000 0.4256 0.1463
0.0056 33.1 82000 0.4393 0.1605
0.0046 33.51 83000 0.6327 0.2056
0.0049 33.91 84000 0.4069 0.1360
0.0031 34.32 85000 0.4359 0.1458
0.0052 34.72 86000 0.2825 0.1032
0.0039 35.12 87000 0.3545 0.1256
0.003 35.53 88000 0.3674 0.1252
0.004 35.93 89000 0.3849 0.1288
0.0029 36.33 90000 0.3465 0.1130
0.003 36.74 91000 0.4034 0.1294
0.0036 37.14 92000 0.3456 0.1209
0.0033 37.55 93000 0.3882 0.1407
0.0037 37.95 94000 0.3372 0.1094
0.0025 38.35 95000 0.3601 0.1137
0.0037 38.76 96000 0.2804 0.1027
0.0022 39.16 97000 0.4160 0.1354
0.0027 39.56 98000 0.3379 0.1202
0.002 39.97 99000 0.3462 0.1171
0.0021 40.37 100000 0.3694 0.1272
0.0014 40.78 101000 0.3315 0.1048
0.0025 41.18 102000 0.3316 0.1088
0.002 41.58 103000 0.3776 0.1319
0.0028 41.99 104000 0.3024 0.1028
0.0015 42.39 105000 0.3087 0.1102
0.0018 42.79 106000 0.3254 0.1067
0.0028 43.2 107000 0.3305 0.1081
0.002 43.6 108000 0.3445 0.1120
0.0019 44.0 109000 0.3264 0.1082
0.0019 44.41 110000 0.3650 0.1202
0.001 44.81 111000 0.3415 0.1133
0.0015 45.22 112000 0.3194 0.1044
0.0011 45.62 113000 0.3302 0.1085
0.0013 46.02 114000 0.3083 0.1053
0.0008 46.43 115000 0.2976 0.0982
0.0019 46.83 116000 0.3212 0.1057
0.0006 47.23 117000 0.3415 0.1089
0.0025 47.64 118000 0.3188 0.1043
0.0009 48.04 119000 0.3136 0.1025
0.0015 48.45 120000 0.3180 0.1050
0.0013 48.85 121000 0.3439 0.1110
0.0007 49.25 122000 0.3286 0.1048
0.0014 49.66 123000 0.3190 0.0999

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.0.1+cu117
  • Datasets 2.18.0
  • Tokenizers 0.15.2