--- license: apache-2.0 base_model: facebook/wav2vec2-large-xlsr-53 tags: - 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](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.4118 - Wer: 0.3219 ## 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: 10000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:------:|:-----:|:---------------:|:------:| | No log | 0.0194 | 100 | 3.5728 | 1.0 | | No log | 0.0387 | 200 | 3.0768 | 1.0 | | No log | 0.0581 | 300 | 3.5010 | 1.0 | | No log | 0.0774 | 400 | 2.0594 | 0.9900 | | 4.06 | 0.0968 | 500 | 1.4703 | 0.8800 | | 4.06 | 0.1161 | 600 | 1.2464 | 0.8297 | | 4.06 | 0.1355 | 700 | 1.0686 | 0.7493 | | 4.06 | 0.1549 | 800 | 1.0069 | 0.7116 | | 4.06 | 0.1742 | 900 | 0.9367 | 0.6888 | | 1.0399 | 0.1936 | 1000 | 0.8961 | 0.6742 | | 1.0399 | 0.2129 | 1100 | 0.8967 | 0.6413 | | 1.0399 | 0.2323 | 1200 | 0.8311 | 0.6153 | | 1.0399 | 0.2516 | 1300 | 0.8019 | 0.5965 | | 1.0399 | 0.2710 | 1400 | 0.7925 | 0.5927 | | 0.8395 | 0.2904 | 1500 | 0.8165 | 0.5987 | | 0.8395 | 0.3097 | 1600 | 0.7696 | 0.6150 | | 0.8395 | 0.3291 | 1700 | 0.7455 | 0.5624 | | 0.8395 | 0.3484 | 1800 | 0.7681 | 0.5684 | | 0.8395 | 0.3678 | 1900 | 0.7292 | 0.5609 | | 0.7574 | 0.3871 | 2000 | 0.7305 | 0.5534 | | 0.7574 | 0.4065 | 2100 | 0.7096 | 0.5363 | | 0.7574 | 0.4259 | 2200 | 0.7108 | 0.5572 | | 0.7574 | 0.4452 | 2300 | 0.6703 | 0.5175 | | 0.7574 | 0.4646 | 2400 | 0.6596 | 0.5149 | | 0.6864 | 0.4839 | 2500 | 0.6846 | 0.5336 | | 0.6864 | 0.5033 | 2600 | 0.6666 | 0.5286 | | 0.6864 | 0.5226 | 2700 | 0.6391 | 0.4949 | | 0.6864 | 0.5420 | 2800 | 0.6296 | 0.4990 | | 0.6864 | 0.5614 | 2900 | 0.6292 | 0.4957 | | 0.6734 | 0.5807 | 3000 | 0.6164 | 0.4765 | | 0.6734 | 0.6001 | 3100 | 0.6180 | 0.4778 | | 0.6734 | 0.6194 | 3200 | 0.6132 | 0.4909 | | 0.6734 | 0.6388 | 3300 | 0.6107 | 0.4683 | | 0.6734 | 0.6581 | 3400 | 0.6068 | 0.4749 | | 0.6433 | 0.6775 | 3500 | 0.6008 | 0.4773 | | 0.6433 | 0.6969 | 3600 | 0.5917 | 0.4656 | | 0.6433 | 0.7162 | 3700 | 0.5885 | 0.4601 | | 0.6433 | 0.7356 | 3800 | 0.5848 | 0.4482 | | 0.6433 | 0.7549 | 3900 | 0.5852 | 0.4496 | | 0.6217 | 0.7743 | 4000 | 0.5772 | 0.4416 | | 0.6217 | 0.7937 | 4100 | 0.5671 | 0.4469 | | 0.6217 | 0.8130 | 4200 | 0.5668 | 0.4463 | | 0.6217 | 0.8324 | 4300 | 0.5558 | 0.4401 | | 0.6217 | 0.8517 | 4400 | 0.5652 | 0.4307 | | 0.5954 | 0.8711 | 4500 | 0.5561 | 0.4307 | | 0.5954 | 0.8904 | 4600 | 0.5432 | 0.4206 | | 0.5954 | 0.9098 | 4700 | 0.5294 | 0.4137 | | 0.5954 | 0.9292 | 4800 | 0.5444 | 0.4210 | | 0.5954 | 0.9485 | 4900 | 0.5291 | 0.4157 | | 0.5663 | 0.9679 | 5000 | 0.5429 | 0.4140 | | 0.5663 | 0.9872 | 5100 | 0.5209 | 0.4116 | | 0.5663 | 1.0066 | 5200 | 0.5282 | 0.4042 | | 0.5663 | 1.0259 | 5300 | 0.5118 | 0.3918 | | 0.5663 | 1.0453 | 5400 | 0.5089 | 0.3993 | | 0.4941 | 1.0647 | 5500 | 0.5011 | 0.3921 | | 0.4941 | 1.0840 | 5600 | 0.5022 | 0.3887 | | 0.4941 | 1.1034 | 5700 | 0.5066 | 0.3853 | | 0.4941 | 1.1227 | 5800 | 0.4907 | 0.3815 | | 0.4941 | 1.1421 | 5900 | 0.4982 | 0.3809 | | 0.4628 | 1.1614 | 6000 | 0.4913 | 0.3896 | | 0.4628 | 1.1808 | 6100 | 0.4826 | 0.3734 | | 0.4628 | 1.2002 | 6200 | 0.4884 | 0.3740 | | 0.4628 | 1.2195 | 6300 | 0.4841 | 0.3700 | | 0.4628 | 1.2389 | 6400 | 0.4828 | 0.3697 | | 0.4435 | 1.2582 | 6500 | 0.4816 | 0.3739 | | 0.4435 | 1.2776 | 6600 | 0.4793 | 0.3674 | | 0.4435 | 1.2969 | 6700 | 0.4744 | 0.3669 | | 0.4435 | 1.3163 | 6800 | 0.4682 | 0.3609 | | 0.4435 | 1.3357 | 6900 | 0.4628 | 0.3594 | | 0.4298 | 1.3550 | 7000 | 0.4663 | 0.3554 | | 0.4298 | 1.3744 | 7100 | 0.4656 | 0.3584 | | 0.4298 | 1.3937 | 7200 | 0.4593 | 0.3565 | | 0.4298 | 1.4131 | 7300 | 0.4599 | 0.3566 | | 0.4298 | 1.4324 | 7400 | 0.4613 | 0.3521 | | 0.4292 | 1.4518 | 7500 | 0.4521 | 0.3475 | | 0.4292 | 1.4712 | 7600 | 0.4512 | 0.3491 | | 0.4292 | 1.4905 | 7700 | 0.4478 | 0.3518 | | 0.4292 | 1.5099 | 7800 | 0.4416 | 0.3421 | | 0.4292 | 1.5292 | 7900 | 0.4427 | 0.3459 | | 0.4072 | 1.5486 | 8000 | 0.4388 | 0.3457 | | 0.4072 | 1.5679 | 8100 | 0.4401 | 0.3453 | | 0.4072 | 1.5873 | 8200 | 0.4365 | 0.3434 | | 0.4072 | 1.6067 | 8300 | 0.4346 | 0.3397 | | 0.4072 | 1.6260 | 8400 | 0.4325 | 0.3360 | | 0.3991 | 1.6454 | 8500 | 0.4320 | 0.3358 | | 0.3991 | 1.6647 | 8600 | 0.4287 | 0.3355 | | 0.3991 | 1.6841 | 8700 | 0.4293 | 0.3334 | | 0.3991 | 1.7034 | 8800 | 0.4272 | 0.3333 | | 0.3991 | 1.7228 | 8900 | 0.4220 | 0.3303 | | 0.3916 | 1.7422 | 9000 | 0.4238 | 0.3292 | | 0.3916 | 1.7615 | 9100 | 0.4215 | 0.3281 | | 0.3916 | 1.7809 | 9200 | 0.4177 | 0.3266 | | 0.3916 | 1.8002 | 9300 | 0.4188 | 0.3257 | | 0.3916 | 1.8196 | 9400 | 0.4164 | 0.3247 | | 0.3687 | 1.8389 | 9500 | 0.4163 | 0.3243 | | 0.3687 | 1.8583 | 9600 | 0.4140 | 0.3239 | | 0.3687 | 1.8777 | 9700 | 0.4132 | 0.3247 | | 0.3687 | 1.8970 | 9800 | 0.4122 | 0.3224 | | 0.3687 | 1.9164 | 9900 | 0.4117 | 0.3219 | | 0.3707 | 1.9357 | 10000 | 0.4118 | 0.3219 | ### Framework versions - Transformers 4.41.2 - Pytorch 2.3.1+cu121 - Datasets 2.19.2 - Tokenizers 0.19.1