--- license: apache-2.0 tags: - generated_from_trainer model-index: - name: ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1-5gram results: [] --- # ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1-5gram This model is a fine-tuned version of [gary109/ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1-5gram](https://huggingface.co/gary109/ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1-5gram) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.4904 - Wer: 0.2087 ## 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: 5e-05 - train_batch_size: 8 - eval_batch_size: 8 - seed: 42 - gradient_accumulation_steps: 8 - total_train_batch_size: 64 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 100 - num_epochs: 100.0 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Wer | |:-------------:|:-----:|:----:|:---------------:|:------:| | 0.504 | 1.0 | 72 | 0.4810 | 0.2313 | | 0.5274 | 2.0 | 144 | 0.4828 | 0.2388 | | 0.5021 | 3.0 | 216 | 0.5008 | 0.2409 | | 0.4956 | 4.0 | 288 | 0.4897 | 0.2401 | | 0.543 | 5.0 | 360 | 0.5795 | 0.2473 | | 0.5239 | 6.0 | 432 | 0.5019 | 0.2490 | | 0.4928 | 7.0 | 504 | 0.4563 | 0.2312 | | 0.5236 | 8.0 | 576 | 0.5311 | 0.2648 | | 0.5059 | 9.0 | 648 | 0.4729 | 0.2294 | | 0.5064 | 10.0 | 720 | 0.5001 | 0.2464 | | 0.5032 | 11.0 | 792 | 0.4968 | 0.2335 | | 0.5064 | 12.0 | 864 | 0.5238 | 0.2365 | | 0.5001 | 13.0 | 936 | 0.4774 | 0.2325 | | 0.5003 | 14.0 | 1008 | 0.4961 | 0.2339 | | 0.5085 | 15.0 | 1080 | 0.4646 | 0.2271 | | 0.4931 | 16.0 | 1152 | 0.5007 | 0.2320 | | 0.4807 | 17.0 | 1224 | 0.4818 | 0.2258 | | 0.4894 | 18.0 | 1296 | 0.4954 | 0.2311 | | 0.4746 | 19.0 | 1368 | 0.4951 | 0.2293 | | 0.4795 | 20.0 | 1440 | 0.4873 | 0.2304 | | 0.4964 | 21.0 | 1512 | 0.4998 | 0.2431 | | 0.4947 | 22.0 | 1584 | 0.4986 | 0.2324 | | 0.4789 | 23.0 | 1656 | 0.5380 | 0.2381 | | 0.4603 | 24.0 | 1728 | 0.4819 | 0.2339 | | 0.4591 | 25.0 | 1800 | 0.4869 | 0.2202 | | 0.4675 | 26.0 | 1872 | 0.4873 | 0.2214 | | 0.4777 | 27.0 | 1944 | 0.4872 | 0.2291 | | 0.475 | 28.0 | 2016 | 0.4920 | 0.2285 | | 0.4921 | 29.0 | 2088 | 0.4855 | 0.2223 | | 0.4875 | 30.0 | 2160 | 0.4978 | 0.2211 | | 0.4699 | 31.0 | 2232 | 0.4962 | 0.2336 | | 0.4787 | 32.0 | 2304 | 0.5130 | 0.2276 | | 0.4631 | 33.0 | 2376 | 0.5091 | 0.2258 | | 0.4697 | 34.0 | 2448 | 0.5217 | 0.2480 | | 0.4677 | 35.0 | 2520 | 0.4852 | 0.2254 | | 0.4644 | 36.0 | 2592 | 0.5171 | 0.2333 | | 0.476 | 37.0 | 2664 | 0.4745 | 0.2209 | | 0.4491 | 38.0 | 2736 | 0.4898 | 0.2282 | | 0.4432 | 39.0 | 2808 | 0.4738 | 0.2237 | | 0.4804 | 40.0 | 2880 | 0.4744 | 0.2245 | | 0.4479 | 41.0 | 2952 | 0.5006 | 0.2202 | | 0.4379 | 42.0 | 3024 | 0.4752 | 0.2214 | | 0.4412 | 43.0 | 3096 | 0.4722 | 0.2211 | | 0.4459 | 44.0 | 3168 | 0.4859 | 0.2203 | | 0.4541 | 45.0 | 3240 | 0.4850 | 0.2189 | | 0.4509 | 46.0 | 3312 | 0.4838 | 0.2162 | | 0.4472 | 47.0 | 3384 | 0.4893 | 0.2311 | | 0.4341 | 48.0 | 3456 | 0.4900 | 0.2198 | | 0.4306 | 49.0 | 3528 | 0.5052 | 0.2166 | | 0.44 | 50.0 | 3600 | 0.5087 | 0.2230 | | 0.4425 | 51.0 | 3672 | 0.4850 | 0.2200 | | 0.4329 | 52.0 | 3744 | 0.4713 | 0.2189 | | 0.4335 | 53.0 | 3816 | 0.4829 | 0.2138 | | 0.4184 | 54.0 | 3888 | 0.5034 | 0.2128 | | 0.4303 | 55.0 | 3960 | 0.4945 | 0.2166 | | 0.4519 | 56.0 | 4032 | 0.5089 | 0.2106 | | 0.4336 | 57.0 | 4104 | 0.5325 | 0.2146 | | 0.4274 | 58.0 | 4176 | 0.5189 | 0.2220 | | 0.4217 | 59.0 | 4248 | 0.5071 | 0.2119 | | 0.4173 | 60.0 | 4320 | 0.5163 | 0.2185 | | 0.4221 | 61.0 | 4392 | 0.4850 | 0.2189 | | 0.4244 | 62.0 | 4464 | 0.4900 | 0.2153 | | 0.4147 | 63.0 | 4536 | 0.4966 | 0.2078 | | 0.4135 | 64.0 | 4608 | 0.5243 | 0.2166 | | 0.4127 | 65.0 | 4680 | 0.4889 | 0.2164 | | 0.3979 | 66.0 | 4752 | 0.4928 | 0.2173 | | 0.4151 | 67.0 | 4824 | 0.4978 | 0.2132 | | 0.4157 | 68.0 | 4896 | 0.5227 | 0.2169 | | 0.4135 | 69.0 | 4968 | 0.4866 | 0.2130 | | 0.4109 | 70.0 | 5040 | 0.4848 | 0.2094 | | 0.4084 | 71.0 | 5112 | 0.5012 | 0.2168 | | 0.4078 | 72.0 | 5184 | 0.4836 | 0.2186 | | 0.4106 | 73.0 | 5256 | 0.4835 | 0.2132 | | 0.3977 | 74.0 | 5328 | 0.5025 | 0.2151 | | 0.4056 | 75.0 | 5400 | 0.4893 | 0.2191 | | 0.4019 | 76.0 | 5472 | 0.4853 | 0.2108 | | 0.4147 | 77.0 | 5544 | 0.4849 | 0.2131 | | 0.4075 | 78.0 | 5616 | 0.5036 | 0.2117 | | 0.4064 | 79.0 | 5688 | 0.4793 | 0.2133 | | 0.4035 | 80.0 | 5760 | 0.4917 | 0.2083 | | 0.4083 | 81.0 | 5832 | 0.4725 | 0.2116 | | 0.3954 | 82.0 | 5904 | 0.4958 | 0.2119 | | 0.395 | 83.0 | 5976 | 0.4898 | 0.2135 | | 0.391 | 84.0 | 6048 | 0.4822 | 0.2151 | | 0.3983 | 85.0 | 6120 | 0.4809 | 0.2110 | | 0.3883 | 86.0 | 6192 | 0.4981 | 0.2108 | | 0.3955 | 87.0 | 6264 | 0.4877 | 0.2111 | | 0.3983 | 88.0 | 6336 | 0.4879 | 0.2135 | | 0.3877 | 89.0 | 6408 | 0.5022 | 0.2150 | | 0.3992 | 90.0 | 6480 | 0.5071 | 0.2117 | | 0.393 | 91.0 | 6552 | 0.4834 | 0.2108 | | 0.3941 | 92.0 | 6624 | 0.4962 | 0.2119 | | 0.3877 | 93.0 | 6696 | 0.4879 | 0.2076 | | 0.3709 | 94.0 | 6768 | 0.4877 | 0.2090 | | 0.3905 | 95.0 | 6840 | 0.4970 | 0.2130 | | 0.3918 | 96.0 | 6912 | 0.4895 | 0.2106 | | 0.3954 | 97.0 | 6984 | 0.4884 | 0.2079 | | 0.3979 | 98.0 | 7056 | 0.4895 | 0.2097 | | 0.373 | 99.0 | 7128 | 0.4875 | 0.2076 | | 0.3769 | 100.0 | 7200 | 0.4904 | 0.2087 | ### Framework versions - Transformers 4.21.0.dev0 - Pytorch 1.9.1+cu102 - Datasets 2.3.3.dev0 - Tokenizers 0.12.1