Hugging Face's logo

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

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
DewiBrynJones/banc-trawsgrifiadau-bangor-clean-with-ccv
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
b30b16e verified
preview code
|
raw
history blame contribute delete
No virus
17.1 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: 0.4212
  • Wer: 0.3394

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: 50000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
No log 0.0079 200 3.1755 1.0
No log 0.0157 400 2.8797 1.0
4.8076 0.0236 600 1.4754 0.9040
4.8076 0.0314 800 1.2526 0.8548
1.1153 0.0393 1000 1.1312 0.7888
1.1153 0.0472 1200 1.0896 0.7735
1.1153 0.0550 1400 1.0288 0.7571
0.8282 0.0629 1600 0.9748 0.7254
0.8282 0.0707 1800 0.9748 0.7195
0.7335 0.0786 2000 0.9883 0.7144
0.7335 0.0864 2200 0.9365 0.7062
0.7335 0.0943 2400 0.9165 0.6802
0.6931 0.1022 2600 0.9170 0.6774
0.6931 0.1100 2800 0.9080 0.6692
0.67 0.1179 3000 0.8609 0.6622
0.67 0.1257 3200 0.8863 0.6659
0.67 0.1336 3400 0.8670 0.6611
0.6282 0.1415 3600 0.8718 0.6820
0.6282 0.1493 3800 0.8617 0.6482
0.6311 0.1572 4000 0.8505 0.6597
0.6311 0.1650 4200 0.8290 0.6292
0.6311 0.1729 4400 0.8300 0.6568
0.615 0.1808 4600 0.8008 0.6109
0.615 0.1886 4800 0.8039 0.6045
0.5785 0.1965 5000 0.7908 0.6072
0.5785 0.2043 5200 0.7868 0.6037
0.5785 0.2122 5400 0.7710 0.5988
0.5928 0.2200 5600 0.7662 0.5747
0.5928 0.2279 5800 0.7673 0.5946
0.5799 0.2358 6000 0.7804 0.5990
0.5799 0.2436 6200 0.7587 0.5781
0.5799 0.2515 6400 0.7495 0.5729
0.5534 0.2593 6600 0.7537 0.5769
0.5534 0.2672 6800 0.7662 0.5812
0.5592 0.2751 7000 0.7571 0.5608
0.5592 0.2829 7200 0.7475 0.5635
0.5592 0.2908 7400 0.7267 0.5592
0.5512 0.2986 7600 0.7362 0.5588
0.5512 0.3065 7800 0.7624 0.5811
0.54 0.3144 8000 0.7657 0.5622
0.54 0.3222 8200 0.7301 0.5454
0.54 0.3301 8400 0.7118 0.5382
0.531 0.3379 8600 0.7253 0.5482
0.531 0.3458 8800 0.7305 0.5583
0.5406 0.3536 9000 0.7098 0.5520
0.5406 0.3615 9200 0.6987 0.5372
0.5406 0.3694 9400 0.7045 0.5473
0.5252 0.3772 9600 0.7025 0.5333
0.5252 0.3851 9800 0.7077 0.5462
0.5156 0.3929 10000 0.7007 0.5383
0.5156 0.4008 10200 0.6947 0.5426
0.5156 0.4087 10400 0.7128 0.5361
0.5181 0.4165 10600 0.6945 0.5276
0.5181 0.4244 10800 0.6986 0.5311
0.5096 0.4322 11000 0.6910 0.5293
0.5096 0.4401 11200 0.6855 0.5281
0.5096 0.4480 11400 0.6890 0.5262
0.5099 0.4558 11600 0.6776 0.5298
0.5099 0.4637 11800 0.6817 0.5142
0.481 0.4715 12000 0.6749 0.5318
0.481 0.4794 12200 0.6648 0.5132
0.481 0.4872 12400 0.6659 0.5151
0.4899 0.4951 12600 0.6744 0.5207
0.4899 0.5030 12800 0.6733 0.5229
0.492 0.5108 13000 0.6457 0.5042
0.492 0.5187 13200 0.6671 0.5259
0.492 0.5265 13400 0.6544 0.5179
0.4782 0.5344 13600 0.6561 0.5054
0.4782 0.5423 13800 0.6382 0.4992
0.507 0.5501 14000 0.6555 0.5044
0.507 0.5580 14200 0.6400 0.4955
0.507 0.5658 14400 0.6468 0.5014
0.4899 0.5737 14600 0.6371 0.4972
0.4899 0.5816 14800 0.6356 0.5026
0.4677 0.5894 15000 0.6386 0.5021
0.4677 0.5973 15200 0.6653 0.5190
0.4677 0.6051 15400 0.6443 0.4998
0.461 0.6130 15600 0.6210 0.4897
0.461 0.6208 15800 0.6396 0.5012
0.4528 0.6287 16000 0.6226 0.4933
0.4528 0.6366 16200 0.6254 0.4937
0.4528 0.6444 16400 0.6289 0.5013
0.451 0.6523 16600 0.6230 0.4972
0.451 0.6601 16800 0.6153 0.4957
0.4444 0.6680 17000 0.6033 0.4748
0.4444 0.6759 17200 0.6154 0.4771
0.4444 0.6837 17400 0.6170 0.4859
0.4357 0.6916 17600 0.6021 0.4815
0.4357 0.6994 17800 0.6071 0.4730
0.4413 0.7073 18000 0.6042 0.4766
0.4413 0.7152 18200 0.6119 0.4838
0.4413 0.7230 18400 0.6046 0.4757
0.4375 0.7309 18600 0.6081 0.4833
0.4375 0.7387 18800 0.6008 0.4728
0.4329 0.7466 19000 0.6008 0.4692
0.4329 0.7545 19200 0.6007 0.4822
0.4329 0.7623 19400 0.5838 0.4658
0.4318 0.7702 19600 0.6008 0.4652
0.4318 0.7780 19800 0.5919 0.4665
0.4265 0.7859 20000 0.5904 0.4722
0.4265 0.7937 20200 0.5923 0.4815
0.4265 0.8016 20400 0.5979 0.4661
0.4321 0.8095 20600 0.5838 0.4561
0.4321 0.8173 20800 0.5825 0.4524
0.4192 0.8252 21000 0.5839 0.4552
0.4192 0.8330 21200 0.5804 0.4594
0.4192 0.8409 21400 0.5891 0.4722
0.4151 0.8488 21600 0.5831 0.4525
0.4151 0.8566 21800 0.5677 0.4543
0.417 0.8645 22000 0.5605 0.4468
0.417 0.8723 22200 0.5705 0.4442
0.417 0.8802 22400 0.5686 0.4551
0.4014 0.8881 22600 0.5752 0.4602
0.4014 0.8959 22800 0.5623 0.4453
0.4024 0.9038 23000 0.5632 0.4424
0.4024 0.9116 23200 0.5681 0.4471
0.4024 0.9195 23400 0.5659 0.4511
0.3899 0.9273 23600 0.5654 0.4417
0.3899 0.9352 23800 0.5691 0.4542
0.3977 0.9431 24000 0.5613 0.4434
0.3977 0.9509 24200 0.5688 0.4433
0.3977 0.9588 24400 0.5749 0.4455
0.3889 0.9666 24600 0.5500 0.4318
0.3889 0.9745 24800 0.5436 0.4372
0.39 0.9824 25000 0.5475 0.4388
0.39 0.9902 25200 0.5532 0.4424
0.39 0.9981 25400 0.5450 0.4281
0.3853 1.0059 25600 0.5463 0.4308
0.3853 1.0138 25800 0.5458 0.4278
0.3413 1.0217 26000 0.5470 0.4344
0.3413 1.0295 26200 0.5358 0.4226
0.3413 1.0374 26400 0.5404 0.4231
0.339 1.0452 26600 0.5345 0.4243
0.339 1.0531 26800 0.5397 0.4200
0.3235 1.0609 27000 0.5379 0.4183
0.3235 1.0688 27200 0.5305 0.4275
0.3235 1.0767 27400 0.5441 0.4248
0.3252 1.0845 27600 0.5362 0.4178
0.3252 1.0924 27800 0.5305 0.4202
0.3301 1.1002 28000 0.5307 0.4185
0.3301 1.1081 28200 0.5402 0.4311
0.3301 1.1160 28400 0.5309 0.4179
0.3087 1.1238 28600 0.5298 0.4214
0.3087 1.1317 28800 0.5331 0.4215
0.3222 1.1395 29000 0.5273 0.4145
0.3222 1.1474 29200 0.5283 0.4131
0.3222 1.1553 29400 0.5257 0.4116
0.3227 1.1631 29600 0.5169 0.4084
0.3227 1.1710 29800 0.5185 0.4107
0.309 1.1788 30000 0.5076 0.4028
0.309 1.1867 30200 0.5178 0.4054
0.309 1.1945 30400 0.5226 0.4122
0.3138 1.2024 30600 0.5227 0.4073
0.3138 1.2103 30800 0.5130 0.4050
0.3083 1.2181 31000 0.5168 0.4113
0.3083 1.2260 31200 0.5054 0.4004
0.3083 1.2338 31400 0.5144 0.4067
0.2981 1.2417 31600 0.5082 0.3992
0.2981 1.2496 31800 0.5134 0.3961
0.2952 1.2574 32000 0.4970 0.3999
0.2952 1.2653 32200 0.5029 0.4006
0.2952 1.2731 32400 0.4980 0.4002
0.2995 1.2810 32600 0.4992 0.4046
0.2995 1.2889 32800 0.4969 0.3912
0.3046 1.2967 33000 0.4943 0.3933
0.3046 1.3046 33200 0.4883 0.3932
0.3046 1.3124 33400 0.4965 0.3935
0.2972 1.3203 33600 0.4910 0.3942
0.2972 1.3281 33800 0.5008 0.4097
0.3093 1.3360 34000 0.4958 0.3957
0.3093 1.3439 34200 0.5045 0.4018
0.3093 1.3517 34400 0.4925 0.3970
0.2947 1.3596 34600 0.4829 0.3905
0.2947 1.3674 34800 0.4870 0.3952
0.2801 1.3753 35000 0.4897 0.3937
0.2801 1.3832 35200 0.5007 0.3997
0.2801 1.3910 35400 0.4823 0.3849
0.2772 1.3989 35600 0.4849 0.3912
0.2772 1.4067 35800 0.4845 0.3882
0.281 1.4146 36000 0.4829 0.3842
0.281 1.4225 36200 0.4815 0.3859
0.281 1.4303 36400 0.4772 0.3808
0.2697 1.4382 36600 0.4870 0.3914
0.2697 1.4460 36800 0.4770 0.3866
0.2766 1.4539 37000 0.4787 0.3821
0.2766 1.4617 37200 0.4793 0.3810
0.2766 1.4696 37400 0.4739 0.3803
0.2905 1.4775 37600 0.4725 0.3811
0.2905 1.4853 37800 0.4727 0.3783
0.2799 1.4932 38000 0.4705 0.3777
0.2799 1.5010 38200 0.4659 0.3751
0.2799 1.5089 38400 0.4691 0.3743
0.267 1.5168 38600 0.4690 0.3664
0.267 1.5246 38800 0.4633 0.3681
0.2632 1.5325 39000 0.4651 0.3726
0.2632 1.5403 39200 0.4690 0.3674
0.2632 1.5482 39400 0.4613 0.3715
0.2716 1.5561 39600 0.4655 0.3697
0.2716 1.5639 39800 0.4597 0.3648
0.2651 1.5718 40000 0.4550 0.3662
0.2651 1.5796 40200 0.4539 0.3676
0.2651 1.5875 40400 0.4543 0.3675
0.2659 1.5953 40600 0.4556 0.3623
0.2659 1.6032 40800 0.4633 0.3685
0.2559 1.6111 41000 0.4529 0.3608
0.2559 1.6189 41200 0.4535 0.3639
0.2559 1.6268 41400 0.4511 0.3637
0.2629 1.6346 41600 0.4556 0.3605
0.2629 1.6425 41800 0.4571 0.3639
0.259 1.6504 42000 0.4620 0.3690
0.259 1.6582 42200 0.4550 0.3635
0.259 1.6661 42400 0.4522 0.3584
0.2594 1.6739 42600 0.4495 0.3589
0.2594 1.6818 42800 0.4453 0.3562
0.2538 1.6897 43000 0.4438 0.3555
0.2538 1.6975 43200 0.4494 0.3567
0.2538 1.7054 43400 0.4444 0.3538
0.2512 1.7132 43600 0.4455 0.3530
0.2512 1.7211 43800 0.4454 0.3522
0.2358 1.7289 44000 0.4445 0.3520
0.2358 1.7368 44200 0.4416 0.3500
0.2358 1.7447 44400 0.4420 0.3490
0.2418 1.7525 44600 0.4386 0.3479
0.2418 1.7604 44800 0.4355 0.3461
0.2421 1.7682 45000 0.4386 0.3437
0.2421 1.7761 45200 0.4348 0.3458
0.2421 1.7840 45400 0.4335 0.3435
0.2418 1.7918 45600 0.4309 0.3444
0.2418 1.7997 45800 0.4321 0.3425
0.2424 1.8075 46000 0.4300 0.3408
0.2424 1.8154 46200 0.4301 0.3423
0.2424 1.8233 46400 0.4339 0.3407
0.228 1.8311 46600 0.4317 0.3429
0.228 1.8390 46800 0.4300 0.3433
0.2532 1.8468 47000 0.4249 0.3439
0.2532 1.8547 47200 0.4257 0.3430
0.2532 1.8625 47400 0.4264 0.3408
0.2347 1.8704 47600 0.4254 0.3409
0.2347 1.8783 47800 0.4237 0.3391
0.2265 1.8861 48000 0.4247 0.3395
0.2265 1.8940 48200 0.4253 0.3389
0.2265 1.9018 48400 0.4246 0.3390
0.2262 1.9097 48600 0.4227 0.3379
0.2262 1.9176 48800 0.4228 0.3389
0.2358 1.9254 49000 0.4225 0.3391
0.2358 1.9333 49200 0.4224 0.3390
0.2358 1.9411 49400 0.4215 0.3390
0.231 1.9490 49600 0.4215 0.3400
0.231 1.9569 49800 0.4212 0.3393
0.2331 1.9647 50000 0.4212 0.3394

Framework versions

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