--- 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](https://huggingface.co/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: 0.4122 - Wer: 0.3223 ## 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 | 3.5545 | 1.0 | | No log | 0.0387 | 200 | 3.0260 | 1.0 | | No log | 0.0581 | 300 | 2.9066 | 1.0 | | No log | 0.0774 | 400 | 2.0133 | 0.9847 | | 4.0489 | 0.0968 | 500 | 1.4598 | 0.9004 | | 4.0489 | 0.1161 | 600 | 1.1772 | 0.8042 | | 4.0489 | 0.1355 | 700 | 1.0787 | 0.7590 | | 4.0489 | 0.1549 | 800 | 1.0144 | 0.7212 | | 4.0489 | 0.1742 | 900 | 0.9339 | 0.6932 | | 1.0454 | 0.1936 | 1000 | 0.8806 | 0.6597 | | 1.0454 | 0.2129 | 1100 | 0.8644 | 0.6554 | | 1.0454 | 0.2323 | 1200 | 0.8454 | 0.6314 | | 1.0454 | 0.2516 | 1300 | 0.8093 | 0.5919 | | 1.0454 | 0.2710 | 1400 | 0.8076 | 0.6072 | | 0.842 | 0.2904 | 1500 | 0.7783 | 0.5857 | | 0.842 | 0.3097 | 1600 | 0.7965 | 0.5941 | | 0.842 | 0.3291 | 1700 | 0.7415 | 0.5505 | | 0.842 | 0.3484 | 1800 | 0.7440 | 0.5637 | | 0.842 | 0.3678 | 1900 | 0.7361 | 0.5865 | | 0.755 | 0.3871 | 2000 | 0.7314 | 0.5427 | | 0.755 | 0.4065 | 2100 | 0.6866 | 0.5181 | | 0.755 | 0.4259 | 2200 | 0.6948 | 0.5426 | | 0.755 | 0.4452 | 2300 | 0.6796 | 0.5159 | | 0.755 | 0.4646 | 2400 | 0.6899 | 0.5305 | | 0.6884 | 0.4839 | 2500 | 0.6736 | 0.5103 | | 0.6884 | 0.5033 | 2600 | 0.6728 | 0.5257 | | 0.6884 | 0.5226 | 2700 | 0.6537 | 0.5027 | | 0.6884 | 0.5420 | 2800 | 0.6314 | 0.4823 | | 0.6884 | 0.5614 | 2900 | 0.6317 | 0.4830 | | 0.6756 | 0.5807 | 3000 | 0.6204 | 0.4761 | | 0.6756 | 0.6001 | 3100 | 0.6311 | 0.4811 | | 0.6756 | 0.6194 | 3200 | 0.6236 | 0.4863 | | 0.6756 | 0.6388 | 3300 | 0.6224 | 0.4629 | | 0.6756 | 0.6581 | 3400 | 0.5973 | 0.4623 | | 0.6435 | 0.6775 | 3500 | 0.5913 | 0.4708 | | 0.6435 | 0.6969 | 3600 | 0.6087 | 0.4744 | | 0.6435 | 0.7162 | 3700 | 0.5827 | 0.4521 | | 0.6435 | 0.7356 | 3800 | 0.5875 | 0.4608 | | 0.6435 | 0.7549 | 3900 | 0.5925 | 0.4557 | | 0.6282 | 0.7743 | 4000 | 0.5799 | 0.4494 | | 0.6282 | 0.7937 | 4100 | 0.5679 | 0.4526 | | 0.6282 | 0.8130 | 4200 | 0.5700 | 0.4550 | | 0.6282 | 0.8324 | 4300 | 0.5610 | 0.4343 | | 0.6282 | 0.8517 | 4400 | 0.5616 | 0.4273 | | 0.5937 | 0.8711 | 4500 | 0.5464 | 0.4221 | | 0.5937 | 0.8904 | 4600 | 0.5486 | 0.4288 | | 0.5937 | 0.9098 | 4700 | 0.5308 | 0.4167 | | 0.5937 | 0.9292 | 4800 | 0.5520 | 0.4200 | | 0.5937 | 0.9485 | 4900 | 0.5321 | 0.4180 | | 0.5659 | 0.9679 | 5000 | 0.5333 | 0.4176 | | 0.5659 | 0.9872 | 5100 | 0.5260 | 0.4111 | | 0.5659 | 1.0066 | 5200 | 0.5185 | 0.3974 | | 0.5659 | 1.0259 | 5300 | 0.5147 | 0.3918 | | 0.5659 | 1.0453 | 5400 | 0.5155 | 0.3976 | | 0.4928 | 1.0647 | 5500 | 0.5058 | 0.3936 | | 0.4928 | 1.0840 | 5600 | 0.5048 | 0.3965 | | 0.4928 | 1.1034 | 5700 | 0.5011 | 0.3818 | | 0.4928 | 1.1227 | 5800 | 0.4965 | 0.3830 | | 0.4928 | 1.1421 | 5900 | 0.4969 | 0.3840 | | 0.4619 | 1.1614 | 6000 | 0.4863 | 0.3800 | | 0.4619 | 1.1808 | 6100 | 0.4908 | 0.3800 | | 0.4619 | 1.2002 | 6200 | 0.4835 | 0.3712 | | 0.4619 | 1.2195 | 6300 | 0.4927 | 0.3767 | | 0.4619 | 1.2389 | 6400 | 0.4942 | 0.3683 | | 0.4421 | 1.2582 | 6500 | 0.4834 | 0.3739 | | 0.4421 | 1.2776 | 6600 | 0.4751 | 0.3634 | | 0.4421 | 1.2969 | 6700 | 0.4734 | 0.3633 | | 0.4421 | 1.3163 | 6800 | 0.4685 | 0.3645 | | 0.4421 | 1.3357 | 6900 | 0.4654 | 0.3625 | | 0.4304 | 1.3550 | 7000 | 0.4742 | 0.3615 | | 0.4304 | 1.3744 | 7100 | 0.4645 | 0.3596 | | 0.4304 | 1.3937 | 7200 | 0.4599 | 0.3594 | | 0.4304 | 1.4131 | 7300 | 0.4554 | 0.3555 | | 0.4304 | 1.4324 | 7400 | 0.4578 | 0.3578 | | 0.4275 | 1.4518 | 7500 | 0.4518 | 0.3522 | | 0.4275 | 1.4712 | 7600 | 0.4480 | 0.3511 | | 0.4275 | 1.4905 | 7700 | 0.4465 | 0.3501 | | 0.4275 | 1.5099 | 7800 | 0.4454 | 0.3428 | | 0.4275 | 1.5292 | 7900 | 0.4427 | 0.3439 | | 0.4089 | 1.5486 | 8000 | 0.4376 | 0.3407 | | 0.4089 | 1.5679 | 8100 | 0.4396 | 0.3415 | | 0.4089 | 1.5873 | 8200 | 0.4343 | 0.3422 | | 0.4089 | 1.6067 | 8300 | 0.4359 | 0.3406 | | 0.4089 | 1.6260 | 8400 | 0.4358 | 0.3373 | | 0.4005 | 1.6454 | 8500 | 0.4331 | 0.3365 | | 0.4005 | 1.6647 | 8600 | 0.4302 | 0.3353 | | 0.4005 | 1.6841 | 8700 | 0.4308 | 0.3355 | | 0.4005 | 1.7034 | 8800 | 0.4258 | 0.3351 | | 0.4005 | 1.7228 | 8900 | 0.4222 | 0.3353 | | 0.3879 | 1.7422 | 9000 | 0.4238 | 0.3312 | | 0.3879 | 1.7615 | 9100 | 0.4245 | 0.3288 | | 0.3879 | 1.7809 | 9200 | 0.4206 | 0.3264 | | 0.3879 | 1.8002 | 9300 | 0.4201 | 0.3284 | | 0.3879 | 1.8196 | 9400 | 0.4189 | 0.3246 | | 0.369 | 1.8389 | 9500 | 0.4160 | 0.3258 | | 0.369 | 1.8583 | 9600 | 0.4142 | 0.3248 | | 0.369 | 1.8777 | 9700 | 0.4131 | 0.3252 | | 0.369 | 1.8970 | 9800 | 0.4128 | 0.3228 | | 0.369 | 1.9164 | 9900 | 0.4122 | 0.3221 | | 0.3738 | 1.9357 | 10000 | 0.4122 | 0.3223 | ### Framework versions - Transformers 4.41.2 - Pytorch 2.3.1+cu121 - Datasets 2.20.0 - Tokenizers 0.19.1