Hugging Face's logo

DewiBrynJones
/
wav2vec2-xlsr-53-ft-btb-ccv-cy

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
wav2vec2-xlsr-53-ft-btb-ccv-cy / README.md
DewiBrynJones's picture
DewiBrynJones
End of training
b88adb3 verified
preview code
|
raw
history blame
14.2 kB
metadata
license: apache-2.0
base_model: facebook/wav2vec2-large-xlsr-53
tags:
  - automatic-speech-recognition
  - DewiBrynJones/banc-trawsgrifiadau-bangor-clean-with-ccv
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: wav2vec2-xlsr-53-ft-btb-ccv-cy
    results: []

wav2vec2-xlsr-53-ft-btb-ccv-cy

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-CLEAN-WITH-CCV - DEFAULT dataset. It achieves the following results on the evaluation set:

  • Loss: inf
  • Wer: 0.2962

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
  • 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: 500
  • training_steps: 200000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
1.059 0.0383 1000 inf 0.7913
0.7527 0.0765 2000 inf 0.7347
0.6861 0.1148 3000 inf 0.6767
0.651 0.1531 4000 inf 0.6462
0.6372 0.1913 5000 inf 0.6245
0.6078 0.2296 6000 inf 0.5932
0.6006 0.2679 7000 inf 0.6101
0.6008 0.3061 8000 inf 0.5832
0.592 0.3444 9000 inf 0.5834
0.5638 0.3827 10000 inf 0.5573
0.5585 0.4209 11000 inf 0.5664
0.5569 0.4592 12000 inf 0.5488
0.5293 0.4975 13000 inf 0.5435
0.5388 0.5357 14000 inf 0.5418
0.5163 0.5740 15000 inf 0.5408
0.5226 0.6123 16000 inf 0.5311
0.4952 0.6505 17000 inf 0.5289
0.524 0.6888 18000 inf 0.5215
0.5076 0.7271 19000 inf 0.5187
0.492 0.7653 20000 inf 0.5095
0.4934 0.8036 21000 inf 0.5061
0.4985 0.8419 22000 inf 0.5129
0.4887 0.8801 23000 inf 0.4928
0.484 0.9184 24000 inf 0.4949
0.4741 0.9567 25000 inf 0.4865
0.4816 0.9949 26000 inf 0.5055
0.44 1.0332 27000 inf 0.4797
0.4359 1.0715 28000 inf 0.4912
0.411 1.1098 29000 inf 0.4774
0.4298 1.1480 30000 inf 0.4773
0.4305 1.1863 31000 inf 0.4898
0.4126 1.2246 32000 inf 0.4739
0.4234 1.2628 33000 inf 0.4844
0.4252 1.3011 34000 inf 0.4763
0.4106 1.3394 35000 inf 0.4709
0.4254 1.3776 36000 inf 0.4737
0.4245 1.4159 37000 inf 0.4534
0.4154 1.4542 38000 inf 0.4566
0.4071 1.4924 39000 inf 0.4635
0.4065 1.5307 40000 inf 0.4668
0.4086 1.5690 41000 inf 0.4607
0.4037 1.6072 42000 inf 0.4616
0.4071 1.6455 43000 inf 0.4607
0.394 1.6838 44000 inf 0.4431
0.4103 1.7220 45000 inf 0.4398
0.3909 1.7603 46000 inf 0.4455
0.3909 1.7986 47000 inf 0.4421
0.3982 1.8368 48000 inf 0.4371
0.3896 1.8751 49000 inf 0.4418
0.3986 1.9134 50000 inf 0.4382
0.3968 1.9516 51000 inf 0.4275
0.4025 1.9899 52000 inf 0.4204
0.3404 2.0282 53000 inf 0.4272
0.3354 2.0664 54000 inf 0.4298
0.3352 2.1047 55000 inf 0.4224
0.3384 2.1430 56000 inf 0.4267
0.3342 2.1812 57000 inf 0.4187
0.3425 2.2195 58000 inf 0.4199
0.3417 2.2578 59000 inf 0.4174
0.3355 2.2960 60000 inf 0.4158
0.3501 2.3343 61000 inf 0.4128
0.3358 2.3726 62000 inf 0.4116
0.3343 2.4108 63000 inf 0.4164
0.3343 2.4491 64000 inf 0.4179
0.3367 2.4874 65000 inf 0.4117
0.3237 2.5256 66000 inf 0.4068
0.3335 2.5639 67000 inf 0.4080
0.3254 2.6022 68000 inf 0.3982
0.3295 2.6404 69000 inf 0.4136
0.3326 2.6787 70000 inf 0.4045
0.3167 2.7170 71000 inf 0.4044
0.3376 2.7552 72000 inf 0.3942
0.3245 2.7935 73000 inf 0.3958
0.315 2.8318 74000 inf 0.4065
0.327 2.8700 75000 inf 0.4010
0.3211 2.9083 76000 inf 0.3926
0.323 2.9466 77000 inf 0.4005
0.323 2.9848 78000 inf 0.3864
0.2747 3.0231 79000 inf 0.3988
0.2706 3.0614 80000 inf 0.3861
0.2696 3.0996 81000 inf 0.3878
0.2792 3.1379 82000 inf 0.3945
0.2809 3.1762 83000 inf 0.3949
0.2709 3.2144 84000 inf 0.3852
0.2808 3.2527 85000 inf 0.3913
0.2746 3.2910 86000 inf 0.3856
0.2633 3.3293 87000 inf 0.3885
0.2745 3.3675 88000 inf 0.3849
0.2832 3.4058 89000 inf 0.3821
0.2806 3.4441 90000 inf 0.3857
0.2756 3.4823 91000 inf 0.3810
0.2733 3.5206 92000 inf 0.3738
0.2807 3.5589 93000 inf 0.3857
0.2773 3.5971 94000 inf 0.3720
0.2725 3.6354 95000 inf 0.3690
0.2614 3.6737 96000 inf 0.3753
0.2674 3.7119 97000 inf 0.3826
0.2605 3.7502 98000 inf 0.3733
0.2649 3.7885 99000 inf 0.3691
0.2638 3.8267 100000 inf 0.3753
0.2749 3.8650 101000 inf 0.3675
0.2635 3.9033 102000 inf 0.3667
0.2639 3.9415 103000 inf 0.3673
0.2602 3.9798 104000 inf 0.3629
0.2217 4.0181 105000 inf 0.3645
0.2226 4.0563 106000 inf 0.3569
0.2209 4.0946 107000 inf 0.3550
0.2326 4.1329 108000 inf 0.3595
0.2203 4.1711 109000 inf 0.3556
0.2267 4.2094 110000 inf 0.3509
0.223 4.2477 111000 inf 0.3581
0.2273 4.2859 112000 inf 0.3548
0.2278 4.3242 113000 inf 0.3493
0.2372 4.3625 114000 inf 0.3601
0.22 4.4007 115000 inf 0.3549
0.228 4.4390 116000 inf 0.3499
0.2291 4.4773 117000 inf 0.3485
0.2301 4.5155 118000 inf 0.3488
0.2084 4.5538 119000 inf 0.3515
0.2251 4.5921 120000 inf 0.3509
0.2205 4.6303 121000 inf 0.3446
0.2174 4.6686 122000 inf 0.3459
0.2136 4.7069 123000 inf 0.3499
0.2142 4.7451 124000 inf 0.3449
0.2152 4.7834 125000 inf 0.3466
0.2216 4.8217 126000 inf 0.3443
0.2209 4.8599 127000 inf 0.3455
0.2183 4.8982 128000 inf 0.3404
0.2174 4.9365 129000 inf 0.3403
0.2165 4.9747 130000 inf 0.3420
0.1806 5.0130 131000 inf 0.3381
0.1821 5.0513 132000 inf 0.3426
0.1825 5.0895 133000 inf 0.3400
0.1876 5.1278 134000 inf 0.3381
0.1858 5.1661 135000 inf 0.3342
0.1729 5.2043 136000 inf 0.3325
0.1843 5.2426 137000 inf 0.3314
0.1828 5.2809 138000 inf 0.3338
0.1878 5.3191 139000 inf 0.3299
0.1784 5.3574 140000 inf 0.3305
0.1791 5.3957 141000 inf 0.3263
0.1861 5.4340 142000 inf 0.3238
0.176 5.4722 143000 inf 0.3245
0.1821 5.5105 144000 inf 0.3216
0.176 5.5488 145000 inf 0.3245
0.1799 5.5870 146000 inf 0.3251
0.1696 5.6253 147000 inf 0.3222
0.1711 5.6636 148000 inf 0.3243
0.1794 5.7018 149000 inf 0.3212
0.1806 5.7401 150000 inf 0.3201
0.1736 5.7784 151000 inf 0.3236
0.1664 5.8166 152000 inf 0.3222
0.1704 5.8549 153000 inf 0.3200
0.1713 5.8932 154000 inf 0.3300
0.1701 5.9314 155000 inf 0.3172
0.1687 5.9697 156000 inf 0.3186
0.1543 6.0080 157000 inf 0.3141
0.142 6.0462 158000 inf 0.3166
0.1438 6.0845 159000 inf 0.3156
0.1433 6.1228 160000 inf 0.3159
0.1442 6.1610 161000 inf 0.3143
0.1494 6.1993 162000 inf 0.3107
0.1355 6.2376 163000 inf 0.3166
0.1403 6.2758 164000 inf 0.3117
0.1435 6.3141 165000 inf 0.3124
0.1446 6.3524 166000 inf 0.3123
0.1385 6.3906 167000 inf 0.3140
0.1437 6.4289 168000 inf 0.3103
0.1328 6.4672 169000 inf 0.3102
0.1354 6.5054 170000 inf 0.3112
0.1394 6.5437 171000 inf 0.3094
0.1385 6.5820 172000 inf 0.3055
0.138 6.6202 173000 inf 0.3055
0.138 6.6585 174000 inf 0.3061
0.1313 6.6968 175000 inf 0.3061
0.1427 6.7350 176000 inf 0.3083
0.1432 6.7733 177000 inf 0.3048
0.136 6.8116 178000 inf 0.3039
0.1424 6.8498 179000 inf 0.3016
0.1347 6.8881 180000 inf 0.3039
0.1307 6.9264 181000 inf 0.3029
0.1293 6.9646 182000 inf 0.3026
0.1259 7.0029 183000 inf 0.3025
0.1151 7.0412 184000 inf 0.3034
0.1143 7.0794 185000 inf 0.3025
0.1105 7.1177 186000 inf 0.3006
0.1126 7.1560 187000 inf 0.3006
0.1139 7.1942 188000 inf 0.2996
0.1101 7.2325 189000 inf 0.2982
0.1187 7.2708 190000 inf 0.2988
0.1174 7.3090 191000 inf 0.2993
0.1132 7.3473 192000 inf 0.2996
0.1108 7.3856 193000 inf 0.2995
0.1119 7.4238 194000 inf 0.2991
0.1098 7.4621 195000 inf 0.2985
0.1053 7.5004 196000 inf 0.2977
0.11 7.5386 197000 inf 0.2975
0.1091 7.5769 198000 inf 0.2959
0.108 7.6152 199000 inf 0.2963
0.1077 7.6535 200000 inf 0.2962

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.3.1+cu121
  • Datasets 2.20.0
  • Tokenizers 0.19.1