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: 0.4064
- Wer: 0.2984
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 |
---|---|---|---|---|
1.092 | 0.0393 | 1000 | 1.1752 | 0.8087 |
0.7307 | 0.0786 | 2000 | 0.9363 | 0.6985 |
0.6633 | 0.1179 | 3000 | 0.8888 | 0.6624 |
0.6273 | 0.1572 | 4000 | 0.8303 | 0.6404 |
0.6031 | 0.1965 | 5000 | 0.7928 | 0.6134 |
0.6007 | 0.2358 | 6000 | 0.7720 | 0.5832 |
0.5739 | 0.2751 | 7000 | 0.7533 | 0.5686 |
0.5655 | 0.3144 | 8000 | 0.7523 | 0.5595 |
0.5584 | 0.3536 | 9000 | 0.7174 | 0.5668 |
0.5454 | 0.3929 | 10000 | 0.7537 | 0.5799 |
0.5322 | 0.4322 | 11000 | 0.7155 | 0.5614 |
0.5206 | 0.4715 | 12000 | 0.7130 | 0.5746 |
0.5304 | 0.5108 | 13000 | 0.6817 | 0.5390 |
0.55 | 0.5501 | 14000 | 0.6903 | 0.5340 |
0.5115 | 0.5894 | 15000 | 0.6974 | 0.5437 |
0.5097 | 0.6287 | 16000 | 0.6786 | 0.5198 |
0.504 | 0.6680 | 17000 | 0.6680 | 0.5067 |
0.4951 | 0.7073 | 18000 | 0.6600 | 0.5222 |
0.4982 | 0.7466 | 19000 | 0.6372 | 0.5011 |
0.493 | 0.7859 | 20000 | 0.6563 | 0.5236 |
0.4928 | 0.8252 | 21000 | 0.6478 | 0.5030 |
0.4964 | 0.8645 | 22000 | 0.6432 | 0.5103 |
0.4818 | 0.9038 | 23000 | 0.6236 | 0.4896 |
0.4752 | 0.9431 | 24000 | 0.6326 | 0.5008 |
0.4736 | 0.9824 | 25000 | 0.6310 | 0.5081 |
0.4241 | 1.0217 | 26000 | 0.6127 | 0.4713 |
0.4196 | 1.0609 | 27000 | 0.6066 | 0.4683 |
0.4177 | 1.1002 | 28000 | 0.5959 | 0.4774 |
0.4204 | 1.1395 | 29000 | 0.6071 | 0.4894 |
0.4238 | 1.1788 | 30000 | 0.6006 | 0.4764 |
0.4253 | 1.2181 | 31000 | 0.5803 | 0.4623 |
0.4156 | 1.2574 | 32000 | 0.5940 | 0.4573 |
0.4058 | 1.2967 | 33000 | 0.5802 | 0.4615 |
0.404 | 1.3360 | 34000 | 0.5882 | 0.4602 |
0.3995 | 1.3753 | 35000 | 0.5841 | 0.4615 |
0.4049 | 1.4146 | 36000 | 0.5853 | 0.4636 |
0.4018 | 1.4539 | 37000 | 0.5737 | 0.4533 |
0.3906 | 1.4932 | 38000 | 0.5848 | 0.4637 |
0.3932 | 1.5325 | 39000 | 0.5516 | 0.4400 |
0.4026 | 1.5718 | 40000 | 0.5642 | 0.4484 |
0.396 | 1.6111 | 41000 | 0.5584 | 0.4512 |
0.3976 | 1.6504 | 42000 | 0.5538 | 0.4436 |
0.3936 | 1.6897 | 43000 | 0.5518 | 0.4412 |
0.3879 | 1.7289 | 44000 | 0.5469 | 0.4297 |
0.3939 | 1.7682 | 45000 | 0.5502 | 0.4402 |
0.386 | 1.8075 | 46000 | 0.5627 | 0.4409 |
0.3823 | 1.8468 | 47000 | 0.5603 | 0.4372 |
0.3955 | 1.8861 | 48000 | 0.5350 | 0.4308 |
0.3808 | 1.9254 | 49000 | 0.5508 | 0.4448 |
0.3871 | 1.9647 | 50000 | 0.5387 | 0.4320 |
0.3668 | 2.0040 | 51000 | 0.5477 | 0.4207 |
0.3324 | 2.0433 | 52000 | 0.5283 | 0.4228 |
0.3327 | 2.0826 | 53000 | 0.5218 | 0.4156 |
0.3251 | 2.1219 | 54000 | 0.5331 | 0.4136 |
0.3466 | 2.1612 | 55000 | 0.5277 | 0.4141 |
0.3259 | 2.2005 | 56000 | 0.5228 | 0.4088 |
0.3292 | 2.2398 | 57000 | 0.5119 | 0.4133 |
0.3323 | 2.2791 | 58000 | 0.5191 | 0.4074 |
0.3228 | 2.3184 | 59000 | 0.5073 | 0.3956 |
0.3172 | 2.3577 | 60000 | 0.5084 | 0.4045 |
0.332 | 2.3970 | 61000 | 0.5130 | 0.4015 |
0.3218 | 2.4362 | 62000 | 0.5103 | 0.3997 |
0.3317 | 2.4755 | 63000 | 0.5020 | 0.4050 |
0.3222 | 2.5148 | 64000 | 0.5072 | 0.3996 |
0.3138 | 2.5541 | 65000 | 0.5098 | 0.4036 |
0.3074 | 2.5934 | 66000 | 0.5026 | 0.3981 |
0.3261 | 2.6327 | 67000 | 0.5030 | 0.3935 |
0.3257 | 2.6720 | 68000 | 0.5003 | 0.3903 |
0.3179 | 2.7113 | 69000 | 0.5139 | 0.4004 |
0.3154 | 2.7506 | 70000 | 0.5041 | 0.3946 |
0.3119 | 2.7899 | 71000 | 0.4914 | 0.3941 |
0.3128 | 2.8292 | 72000 | 0.4867 | 0.3831 |
0.3105 | 2.8685 | 73000 | 0.4871 | 0.3818 |
0.309 | 2.9078 | 74000 | 0.4887 | 0.3989 |
0.3073 | 2.9471 | 75000 | 0.4839 | 0.3904 |
0.3023 | 2.9864 | 76000 | 0.4839 | 0.3805 |
0.2715 | 3.0257 | 77000 | 0.4816 | 0.3877 |
0.2565 | 3.0650 | 78000 | 0.4994 | 0.3811 |
0.2697 | 3.1042 | 79000 | 0.4803 | 0.3813 |
0.2717 | 3.1435 | 80000 | 0.4843 | 0.3799 |
0.2738 | 3.1828 | 81000 | 0.4905 | 0.3797 |
0.2617 | 3.2221 | 82000 | 0.4754 | 0.3728 |
0.2634 | 3.2614 | 83000 | 0.4730 | 0.3668 |
0.2648 | 3.3007 | 84000 | 0.4768 | 0.3691 |
0.2567 | 3.3400 | 85000 | 0.4812 | 0.3741 |
0.2687 | 3.3793 | 86000 | 0.4683 | 0.3716 |
0.2757 | 3.4186 | 87000 | 0.4690 | 0.3732 |
0.2596 | 3.4579 | 88000 | 0.4753 | 0.3782 |
0.2589 | 3.4972 | 89000 | 0.4645 | 0.3691 |
0.2627 | 3.5365 | 90000 | 0.4690 | 0.3675 |
0.2804 | 3.5758 | 91000 | 0.4675 | 0.3742 |
0.2587 | 3.6151 | 92000 | 0.4674 | 0.3594 |
0.2615 | 3.6544 | 93000 | 0.4657 | 0.3632 |
0.2531 | 3.6937 | 94000 | 0.4589 | 0.3668 |
0.2466 | 3.7330 | 95000 | 0.4618 | 0.3691 |
0.2653 | 3.7723 | 96000 | 0.4614 | 0.3775 |
0.2542 | 3.8115 | 97000 | 0.4600 | 0.3726 |
0.2616 | 3.8508 | 98000 | 0.4511 | 0.3660 |
0.2625 | 3.8901 | 99000 | 0.4607 | 0.3644 |
0.2627 | 3.9294 | 100000 | 0.4456 | 0.3626 |
0.252 | 3.9687 | 101000 | 0.4579 | 0.3665 |
0.2489 | 4.0080 | 102000 | 0.4510 | 0.3620 |
0.2218 | 4.0473 | 103000 | 0.4419 | 0.3535 |
0.2211 | 4.0866 | 104000 | 0.4499 | 0.3771 |
0.2186 | 4.1259 | 105000 | 0.4546 | 0.3660 |
0.2199 | 4.1652 | 106000 | 0.4396 | 0.3542 |
0.2227 | 4.2045 | 107000 | 0.4469 | 0.3575 |
0.2212 | 4.2438 | 108000 | 0.4403 | 0.3501 |
0.2182 | 4.2831 | 109000 | 0.4507 | 0.3599 |
0.2212 | 4.3224 | 110000 | 0.4435 | 0.3576 |
0.2211 | 4.3617 | 111000 | 0.4514 | 0.3689 |
0.2116 | 4.4010 | 112000 | 0.4443 | 0.3508 |
0.2218 | 4.4403 | 113000 | 0.4410 | 0.3472 |
0.2152 | 4.4795 | 114000 | 0.4468 | 0.3535 |
0.2174 | 4.5188 | 115000 | 0.4499 | 0.3470 |
0.212 | 4.5581 | 116000 | 0.4454 | 0.3440 |
0.2039 | 4.5974 | 117000 | 0.4424 | 0.3489 |
0.2073 | 4.6367 | 118000 | 0.4437 | 0.3466 |
0.2177 | 4.6760 | 119000 | 0.4392 | 0.3422 |
0.2121 | 4.7153 | 120000 | 0.4443 | 0.3437 |
0.2072 | 4.7546 | 121000 | 0.4268 | 0.3462 |
0.2138 | 4.7939 | 122000 | 0.4272 | 0.3432 |
0.2145 | 4.8332 | 123000 | 0.4332 | 0.3454 |
0.2217 | 4.8725 | 124000 | 0.4210 | 0.3391 |
0.2069 | 4.9118 | 125000 | 0.4278 | 0.3376 |
0.2068 | 4.9511 | 126000 | 0.4216 | 0.3387 |
0.2129 | 4.9904 | 127000 | 0.4210 | 0.3362 |
0.1774 | 5.0297 | 128000 | 0.4340 | 0.3304 |
0.1705 | 5.0690 | 129000 | 0.4422 | 0.3299 |
0.1746 | 5.1083 | 130000 | 0.4306 | 0.3363 |
0.1813 | 5.1476 | 131000 | 0.4181 | 0.3334 |
0.1729 | 5.1868 | 132000 | 0.4319 | 0.3369 |
0.1777 | 5.2261 | 133000 | 0.4190 | 0.3327 |
0.18 | 5.2654 | 134000 | 0.4228 | 0.3338 |
0.1747 | 5.3047 | 135000 | 0.4268 | 0.3323 |
0.1737 | 5.3440 | 136000 | 0.4193 | 0.3325 |
0.1709 | 5.3833 | 137000 | 0.4229 | 0.3279 |
0.1726 | 5.4226 | 138000 | 0.4179 | 0.3271 |
0.1741 | 5.4619 | 139000 | 0.4205 | 0.3255 |
0.1723 | 5.5012 | 140000 | 0.4140 | 0.3294 |
0.1676 | 5.5405 | 141000 | 0.4256 | 0.3254 |
0.1769 | 5.5798 | 142000 | 0.4180 | 0.3279 |
0.1718 | 5.6191 | 143000 | 0.4158 | 0.3204 |
0.1735 | 5.6584 | 144000 | 0.4174 | 0.3209 |
0.1693 | 5.6977 | 145000 | 0.4166 | 0.3198 |
0.1745 | 5.7370 | 146000 | 0.4165 | 0.3245 |
0.1692 | 5.7763 | 147000 | 0.4148 | 0.3230 |
0.1641 | 5.8156 | 148000 | 0.4116 | 0.3216 |
0.173 | 5.8548 | 149000 | 0.4041 | 0.3237 |
0.1664 | 5.8941 | 150000 | 0.4039 | 0.3184 |
0.1648 | 5.9334 | 151000 | 0.4072 | 0.3166 |
0.1709 | 5.9727 | 152000 | 0.4022 | 0.3206 |
0.151 | 6.0120 | 153000 | 0.4034 | 0.3188 |
0.1353 | 6.0513 | 154000 | 0.4128 | 0.3257 |
0.1476 | 6.0906 | 155000 | 0.4197 | 0.3200 |
0.1465 | 6.1299 | 156000 | 0.4073 | 0.3167 |
0.139 | 6.1692 | 157000 | 0.4228 | 0.3212 |
0.1404 | 6.2085 | 158000 | 0.4117 | 0.3245 |
0.1338 | 6.2478 | 159000 | 0.4180 | 0.3153 |
0.1436 | 6.2871 | 160000 | 0.4264 | 0.3167 |
0.1317 | 6.3264 | 161000 | 0.4118 | 0.3152 |
0.1395 | 6.3657 | 162000 | 0.4269 | 0.3118 |
0.1267 | 6.4050 | 163000 | 0.4241 | 0.3135 |
0.1334 | 6.4443 | 164000 | 0.4058 | 0.3169 |
0.1369 | 6.4836 | 165000 | 0.4050 | 0.3130 |
0.1322 | 6.5228 | 166000 | 0.4097 | 0.3140 |
0.1358 | 6.5621 | 167000 | 0.4142 | 0.3129 |
0.1345 | 6.6014 | 168000 | 0.4009 | 0.3123 |
0.1321 | 6.6407 | 169000 | 0.4005 | 0.3093 |
0.1299 | 6.6800 | 170000 | 0.3996 | 0.3054 |
0.1345 | 6.7193 | 171000 | 0.4041 | 0.3071 |
0.1328 | 6.7586 | 172000 | 0.3997 | 0.3070 |
0.1245 | 6.7979 | 173000 | 0.3974 | 0.3045 |
0.1356 | 6.8372 | 174000 | 0.3999 | 0.3009 |
0.1208 | 6.8765 | 175000 | 0.3953 | 0.3019 |
0.1316 | 6.9158 | 176000 | 0.3974 | 0.3056 |
0.1232 | 6.9551 | 177000 | 0.3921 | 0.3033 |
0.1261 | 6.9944 | 178000 | 0.3985 | 0.3035 |
0.1184 | 7.0337 | 179000 | 0.4006 | 0.3061 |
0.1115 | 7.0730 | 180000 | 0.4096 | 0.3050 |
0.1109 | 7.1123 | 181000 | 0.4138 | 0.3038 |
0.1113 | 7.1516 | 182000 | 0.4119 | 0.3052 |
0.1075 | 7.1909 | 183000 | 0.4170 | 0.3007 |
0.1081 | 7.2301 | 184000 | 0.4135 | 0.3031 |
0.1108 | 7.2694 | 185000 | 0.4129 | 0.3003 |
0.1044 | 7.3087 | 186000 | 0.4130 | 0.3023 |
0.1121 | 7.3480 | 187000 | 0.4079 | 0.2993 |
0.1052 | 7.3873 | 188000 | 0.4048 | 0.3019 |
0.103 | 7.4266 | 189000 | 0.4154 | 0.3015 |
0.1105 | 7.4659 | 190000 | 0.4120 | 0.3019 |
0.1093 | 7.5052 | 191000 | 0.4105 | 0.3007 |
0.1058 | 7.5445 | 192000 | 0.4102 | 0.3011 |
0.1043 | 7.5838 | 193000 | 0.4101 | 0.2994 |
0.1098 | 7.6231 | 194000 | 0.4085 | 0.2998 |
0.1057 | 7.6624 | 195000 | 0.4072 | 0.2982 |
0.1021 | 7.7017 | 196000 | 0.4079 | 0.2974 |
0.0994 | 7.7410 | 197000 | 0.4089 | 0.2987 |
0.1065 | 7.7803 | 198000 | 0.4066 | 0.2974 |
0.1111 | 7.8196 | 199000 | 0.4071 | 0.2982 |
0.1065 | 7.8589 | 200000 | 0.4064 | 0.2984 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1