--- license: apache-2.0 base_model: facebook/wav2vec2-xls-r-2b tags: - automatic-speech-recognition - DewiBrynJones/banc-trawsgrifiadau-bangor-clean-with-ccv - generated_from_trainer metrics: - wer model-index: - name: wav2vec2-xls-r-2b-ft-btb-ccv-cy results: [] --- # wav2vec2-xls-r-2b-ft-btb-ccv-cy This model is a fine-tuned version of [facebook/wav2vec2-xls-r-2b](https://huggingface.co/facebook/wav2vec2-xls-r-2b) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-CLEAN-WITH-CCV - DEFAULT dataset. It achieves the following results on the evaluation set: - Loss: inf - Wer: 0.3671 ## 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 | |:-------------:|:------:|:------:|:---------------:|:------:| | 1.5221 | 0.0383 | 1000 | inf | 0.9140 | | 1.4078 | 0.0765 | 2000 | inf | 0.9066 | | 1.336 | 0.1148 | 3000 | inf | 0.8949 | | 1.2887 | 0.1531 | 4000 | inf | 0.8746 | | 1.26 | 0.1913 | 5000 | inf | 0.8671 | | 1.2188 | 0.2296 | 6000 | inf | 0.8700 | | 1.1992 | 0.2679 | 7000 | inf | 0.8538 | | 1.1773 | 0.3061 | 8000 | inf | 0.8329 | | 1.1602 | 0.3444 | 9000 | inf | 0.8200 | | 1.1128 | 0.3827 | 10000 | inf | 0.8059 | | 1.0893 | 0.4209 | 11000 | inf | 0.8277 | | 1.0809 | 0.4592 | 12000 | inf | 0.8028 | | 1.0426 | 0.4975 | 13000 | inf | 0.8004 | | 1.0202 | 0.5357 | 14000 | inf | 0.7825 | | 1.0005 | 0.5740 | 15000 | inf | 0.7786 | | 0.9987 | 0.6123 | 16000 | inf | 0.7521 | | 0.9705 | 0.6505 | 17000 | inf | 0.7593 | | 0.9884 | 0.6888 | 18000 | inf | 0.7381 | | 0.9618 | 0.7271 | 19000 | inf | 0.7365 | | 0.9374 | 0.7653 | 20000 | inf | 0.7416 | | 0.9175 | 0.8036 | 21000 | inf | 0.7299 | | 0.9247 | 0.8419 | 22000 | inf | 0.7260 | | 0.9001 | 0.8801 | 23000 | inf | 0.7235 | | 0.8836 | 0.9184 | 24000 | inf | 0.7086 | | 0.8789 | 0.9567 | 25000 | inf | 0.7145 | | 0.8734 | 0.9949 | 26000 | inf | 0.7195 | | 0.8398 | 1.0332 | 27000 | inf | 0.6838 | | 0.8268 | 1.0715 | 28000 | inf | 0.6793 | | 0.8196 | 1.1098 | 29000 | inf | 0.6639 | | 0.8124 | 1.1480 | 30000 | inf | 0.6615 | | 0.7935 | 1.1863 | 31000 | inf | 0.6608 | | 0.817 | 1.2246 | 32000 | inf | 0.6710 | | 0.7975 | 1.2628 | 33000 | inf | 0.6695 | | 0.7746 | 1.3011 | 34000 | inf | 0.6674 | | 0.8013 | 1.3394 | 35000 | inf | 0.6587 | | 0.7703 | 1.3776 | 36000 | inf | 0.6388 | | 0.7581 | 1.4159 | 37000 | inf | 0.6461 | | 0.7468 | 1.4542 | 38000 | inf | 0.6334 | | 0.7534 | 1.4924 | 39000 | inf | 0.6301 | | 0.752 | 1.5307 | 40000 | inf | 0.6222 | | 0.736 | 1.5690 | 41000 | inf | 0.6203 | | 0.7188 | 1.6072 | 42000 | inf | 0.6208 | | 0.7308 | 1.6455 | 43000 | inf | 0.6057 | | 0.7179 | 1.6838 | 44000 | inf | 0.6292 | | 0.7341 | 1.7220 | 45000 | inf | 0.6034 | | 0.7061 | 1.7603 | 46000 | inf | 0.6137 | | 0.7081 | 1.7986 | 47000 | inf | 0.6123 | | 0.6957 | 1.8368 | 48000 | inf | 0.6054 | | 0.7052 | 1.8751 | 49000 | inf | 0.6165 | | 0.6833 | 1.9134 | 50000 | inf | 0.5887 | | 0.6995 | 1.9516 | 51000 | inf | 0.5871 | | 0.6703 | 1.9899 | 52000 | inf | 0.5954 | | 0.6265 | 2.0282 | 53000 | inf | 0.5792 | | 0.633 | 2.0664 | 54000 | inf | 0.5696 | | 0.6399 | 2.1047 | 55000 | inf | 0.5718 | | 0.6165 | 2.1430 | 56000 | inf | 0.5837 | | 0.6148 | 2.1812 | 57000 | inf | 0.5597 | | 0.6228 | 2.2195 | 58000 | inf | 0.5707 | | 0.6302 | 2.2578 | 59000 | inf | 0.5718 | | 0.6035 | 2.2960 | 60000 | inf | 0.5639 | | 0.602 | 2.3343 | 61000 | inf | 0.5633 | | 0.6023 | 2.3726 | 62000 | inf | 0.5581 | | 0.5924 | 2.4108 | 63000 | inf | 0.5512 | | 0.5969 | 2.4491 | 64000 | inf | 0.5490 | | 0.6029 | 2.4874 | 65000 | inf | 0.5444 | | 0.6046 | 2.5256 | 66000 | inf | 0.5461 | | 0.6095 | 2.5639 | 67000 | inf | 0.5476 | | 0.598 | 2.6022 | 68000 | inf | 0.5321 | | 0.5812 | 2.6404 | 69000 | inf | 0.5357 | | 0.5957 | 2.6787 | 70000 | inf | 0.5368 | | 0.5909 | 2.7170 | 71000 | inf | 0.5239 | | 0.5953 | 2.7552 | 72000 | inf | 0.5422 | | 0.5702 | 2.7935 | 73000 | inf | 0.5226 | | 0.5755 | 2.8318 | 74000 | inf | 0.5319 | | 0.5659 | 2.8700 | 75000 | inf | 0.5287 | | 0.5581 | 2.9083 | 76000 | inf | 0.5278 | | 0.5786 | 2.9466 | 77000 | inf | 0.5195 | | 0.5485 | 2.9848 | 78000 | inf | 0.5256 | | 0.5113 | 3.0231 | 79000 | inf | 0.5220 | | 0.4973 | 3.0614 | 80000 | inf | 0.5146 | | 0.5085 | 3.0996 | 81000 | inf | 0.5241 | | 0.5111 | 3.1379 | 82000 | inf | 0.5105 | | 0.5047 | 3.1762 | 83000 | inf | 0.5118 | | 0.4994 | 3.2144 | 84000 | inf | 0.4993 | | 0.5077 | 3.2527 | 85000 | inf | 0.5100 | | 0.5035 | 3.2910 | 86000 | inf | 0.4929 | | 0.5045 | 3.3293 | 87000 | inf | 0.5026 | | 0.4951 | 3.3675 | 88000 | inf | 0.4971 | | 0.4915 | 3.4058 | 89000 | inf | 0.4984 | | 0.4875 | 3.4441 | 90000 | inf | 0.4968 | | 0.4964 | 3.4823 | 91000 | inf | 0.4989 | | 0.4767 | 3.5206 | 92000 | inf | 0.4922 | | 0.4765 | 3.5589 | 93000 | inf | 0.4869 | | 0.4967 | 3.5971 | 94000 | inf | 0.4981 | | 0.4941 | 3.6354 | 95000 | inf | 0.4962 | | 0.4808 | 3.6737 | 96000 | inf | 0.4856 | | 0.4838 | 3.7119 | 97000 | inf | 0.4749 | | 0.4644 | 3.7502 | 98000 | inf | 0.4738 | | 0.4818 | 3.7885 | 99000 | inf | 0.4737 | | 0.4741 | 3.8267 | 100000 | inf | 0.4812 | | 0.4734 | 3.8650 | 101000 | inf | 0.4772 | | 0.4733 | 3.9033 | 102000 | inf | 0.4736 | | 0.4937 | 3.9415 | 103000 | inf | 0.4695 | | 0.4864 | 3.9798 | 104000 | inf | 0.4749 | | 0.4126 | 4.0181 | 105000 | inf | 0.4635 | | 0.4228 | 4.0563 | 106000 | inf | 0.4701 | | 0.4098 | 4.0946 | 107000 | inf | 0.4589 | | 0.4193 | 4.1329 | 108000 | inf | 0.4615 | | 0.4083 | 4.1711 | 109000 | inf | 0.4640 | | 0.406 | 4.2094 | 110000 | inf | 0.4614 | | 0.4125 | 4.2477 | 111000 | inf | 0.4608 | | 0.4104 | 4.2859 | 112000 | inf | 0.4487 | | 0.3988 | 4.3242 | 113000 | inf | 0.4599 | | 0.4034 | 4.3625 | 114000 | inf | 0.4539 | | 0.4023 | 4.4007 | 115000 | inf | 0.4480 | | 0.4026 | 4.4390 | 116000 | inf | 0.4524 | | 0.4182 | 4.4773 | 117000 | inf | 0.4473 | | 0.4046 | 4.5155 | 118000 | inf | 0.4456 | | 0.4126 | 4.5538 | 119000 | inf | 0.4406 | | 0.4144 | 4.5921 | 120000 | inf | 0.4449 | | 0.4074 | 4.6303 | 121000 | inf | 0.4475 | | 0.3922 | 4.6686 | 122000 | inf | 0.4388 | | 0.3866 | 4.7069 | 123000 | inf | 0.4474 | | 0.3873 | 4.7451 | 124000 | inf | 0.4345 | | 0.3917 | 4.7834 | 125000 | inf | 0.4338 | | 0.3864 | 4.8217 | 126000 | inf | 0.4351 | | 0.3826 | 4.8599 | 127000 | inf | 0.4308 | | 0.391 | 4.8982 | 128000 | inf | 0.4315 | | 0.394 | 4.9365 | 129000 | inf | 0.4279 | | 0.3957 | 4.9747 | 130000 | inf | 0.4235 | | 0.3515 | 5.0130 | 131000 | inf | 0.4216 | | 0.3389 | 5.0513 | 132000 | inf | 0.4255 | | 0.333 | 5.0895 | 133000 | inf | 0.4253 | | 0.3313 | 5.1278 | 134000 | inf | 0.4178 | | 0.3351 | 5.1661 | 135000 | inf | 0.4223 | | 0.3262 | 5.2043 | 136000 | inf | 0.4163 | | 0.3333 | 5.2426 | 137000 | inf | 0.4216 | | 0.3229 | 5.2809 | 138000 | inf | 0.4133 | | 0.3345 | 5.3191 | 139000 | inf | 0.4136 | | 0.3365 | 5.3574 | 140000 | inf | 0.4193 | | 0.3165 | 5.3957 | 141000 | inf | 0.4112 | | 0.3224 | 5.4340 | 142000 | inf | 0.4075 | | 0.335 | 5.4722 | 143000 | inf | 0.4113 | | 0.3377 | 5.5105 | 144000 | inf | 0.4176 | | 0.3411 | 5.5488 | 145000 | inf | 0.4091 | | 0.3247 | 5.5870 | 146000 | inf | 0.4096 | | 0.3304 | 5.6253 | 147000 | inf | 0.4084 | | 0.3267 | 5.6636 | 148000 | inf | 0.4042 | | 0.3193 | 5.7018 | 149000 | inf | 0.4026 | | 0.3155 | 5.7401 | 150000 | inf | 0.4048 | | 0.3238 | 5.7784 | 151000 | inf | 0.4033 | | 0.3172 | 5.8166 | 152000 | inf | 0.4049 | | 0.3148 | 5.8549 | 153000 | inf | 0.3989 | | 0.3217 | 5.8932 | 154000 | inf | 0.3978 | | 0.3145 | 5.9314 | 155000 | inf | 0.3930 | | 0.3178 | 5.9697 | 156000 | inf | 0.3995 | | 0.2895 | 6.0080 | 157000 | inf | 0.3998 | | 0.269 | 6.0462 | 158000 | inf | 0.3926 | | 0.2757 | 6.0845 | 159000 | inf | 0.3923 | | 0.2573 | 6.1228 | 160000 | inf | 0.3906 | | 0.2666 | 6.1610 | 161000 | inf | 0.3883 | | 0.2691 | 6.1993 | 162000 | inf | 0.3920 | | 0.2699 | 6.2376 | 163000 | inf | 0.3962 | | 0.259 | 6.2758 | 164000 | inf | 0.3902 | | 0.2707 | 6.3141 | 165000 | inf | 0.3878 | | 0.265 | 6.3524 | 166000 | inf | 0.3856 | | 0.2657 | 6.3906 | 167000 | inf | 0.3851 | | 0.2625 | 6.4289 | 168000 | inf | 0.3841 | | 0.2615 | 6.4672 | 169000 | inf | 0.3832 | | 0.2629 | 6.5054 | 170000 | inf | 0.3834 | | 0.276 | 6.5437 | 171000 | inf | 0.3831 | | 0.2623 | 6.5820 | 172000 | inf | 0.3813 | | 0.27 | 6.6202 | 173000 | inf | 0.3815 | | 0.2712 | 6.6585 | 174000 | inf | 0.3812 | | 0.263 | 6.6968 | 175000 | inf | 0.3816 | | 0.2616 | 6.7350 | 176000 | inf | 0.3796 | | 0.253 | 6.7733 | 177000 | inf | 0.3794 | | 0.2572 | 6.8116 | 178000 | inf | 0.3829 | | 0.2517 | 6.8498 | 179000 | inf | 0.3773 | | 0.2471 | 6.8881 | 180000 | inf | 0.3783 | | 0.2441 | 6.9264 | 181000 | inf | 0.3763 | | 0.245 | 6.9646 | 182000 | inf | 0.3749 | | 0.235 | 7.0029 | 183000 | inf | 0.3724 | | 0.2281 | 7.0412 | 184000 | inf | 0.3743 | | 0.2155 | 7.0794 | 185000 | inf | 0.3742 | | 0.2177 | 7.1177 | 186000 | inf | 0.3737 | | 0.2107 | 7.1560 | 187000 | inf | 0.3708 | | 0.2129 | 7.1942 | 188000 | inf | 0.3716 | | 0.2173 | 7.2325 | 189000 | inf | 0.3695 | | 0.2145 | 7.2708 | 190000 | inf | 0.3722 | | 0.2116 | 7.3090 | 191000 | inf | 0.3702 | | 0.212 | 7.3473 | 192000 | inf | 0.3704 | | 0.2116 | 7.3856 | 193000 | inf | 0.3701 | | 0.2124 | 7.4238 | 194000 | inf | 0.3687 | | 0.2078 | 7.4621 | 195000 | inf | 0.3681 | | 0.2158 | 7.5004 | 196000 | inf | 0.3682 | | 0.2157 | 7.5386 | 197000 | inf | 0.3673 | | 0.2045 | 7.5769 | 198000 | inf | 0.3667 | | 0.2188 | 7.6152 | 199000 | inf | 0.3675 | | 0.2041 | 7.6535 | 200000 | inf | 0.3671 | ### Framework versions - Transformers 4.41.2 - Pytorch 2.3.1+cu121 - Datasets 2.20.0 - Tokenizers 0.19.1