--- tags: - automatic-speech-recognition - gary109/AI_Light_Dance - generated_from_trainer model-index: - name: ai-light-dance_singing3_ft_pretrain2_wav2vec2-large-xlsr-53 results: [] --- # ai-light-dance_singing3_ft_pretrain2_wav2vec2-large-xlsr-53 This model is a fine-tuned version of [gary109/ai-light-dance_singing3_ft_pretrain2_wav2vec2-large-xlsr-53](https://huggingface.co/gary109/ai-light-dance_singing3_ft_pretrain2_wav2vec2-large-xlsr-53) on the GARY109/AI_LIGHT_DANCE - ONSET-SINGING3 dataset. It achieves the following results on the evaluation set: - Loss: 2.4279 - Wer: 1.0087 ## 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 | |:-------------:|:-----:|:----:|:---------------:|:------:| | 1.209 | 1.0 | 72 | 2.5599 | 0.9889 | | 1.3395 | 2.0 | 144 | 2.7188 | 0.9877 | | 1.2695 | 3.0 | 216 | 2.9989 | 0.9709 | | 1.2818 | 4.0 | 288 | 3.2352 | 0.9757 | | 1.2389 | 5.0 | 360 | 3.6867 | 0.9783 | | 1.2368 | 6.0 | 432 | 3.3189 | 0.9811 | | 1.2307 | 7.0 | 504 | 3.0786 | 0.9657 | | 1.2607 | 8.0 | 576 | 2.9720 | 0.9677 | | 1.2584 | 9.0 | 648 | 2.5613 | 0.9702 | | 1.2266 | 10.0 | 720 | 2.6937 | 0.9610 | | 1.262 | 11.0 | 792 | 3.9060 | 0.9745 | | 1.2361 | 12.0 | 864 | 3.6138 | 0.9718 | | 1.2348 | 13.0 | 936 | 3.4838 | 0.9745 | | 1.2715 | 14.0 | 1008 | 3.3128 | 0.9751 | | 1.2505 | 15.0 | 1080 | 3.2015 | 0.9710 | | 1.211 | 16.0 | 1152 | 3.4709 | 0.9709 | | 1.2067 | 17.0 | 1224 | 3.0566 | 0.9673 | | 1.2536 | 18.0 | 1296 | 2.5479 | 0.9789 | | 1.2297 | 19.0 | 1368 | 2.8307 | 0.9710 | | 1.1949 | 20.0 | 1440 | 3.4112 | 0.9777 | | 1.2181 | 21.0 | 1512 | 2.6784 | 0.9682 | | 1.195 | 22.0 | 1584 | 3.0395 | 0.9639 | | 1.2047 | 23.0 | 1656 | 3.1935 | 0.9726 | | 1.2306 | 24.0 | 1728 | 3.2649 | 0.9723 | | 1.199 | 25.0 | 1800 | 3.1378 | 0.9645 | | 1.1945 | 26.0 | 1872 | 2.8143 | 0.9596 | | 1.19 | 27.0 | 1944 | 3.5174 | 0.9787 | | 1.1976 | 28.0 | 2016 | 2.9666 | 0.9594 | | 1.2229 | 29.0 | 2088 | 2.8672 | 0.9589 | | 1.1548 | 30.0 | 2160 | 2.6568 | 0.9627 | | 1.169 | 31.0 | 2232 | 2.8799 | 0.9654 | | 1.1857 | 32.0 | 2304 | 2.8691 | 0.9625 | | 1.1862 | 33.0 | 2376 | 2.8251 | 0.9555 | | 1.1721 | 34.0 | 2448 | 3.5968 | 0.9726 | | 1.1293 | 35.0 | 2520 | 3.4130 | 0.9651 | | 1.1513 | 36.0 | 2592 | 2.8804 | 0.9630 | | 1.1537 | 37.0 | 2664 | 2.5824 | 0.9575 | | 1.1818 | 38.0 | 2736 | 2.8443 | 0.9613 | | 1.1835 | 39.0 | 2808 | 2.6431 | 0.9619 | | 1.1457 | 40.0 | 2880 | 2.9254 | 0.9639 | | 1.1591 | 41.0 | 2952 | 2.8194 | 0.9561 | | 1.1284 | 42.0 | 3024 | 2.6432 | 0.9806 | | 1.1602 | 43.0 | 3096 | 2.4279 | 1.0087 | | 1.1556 | 44.0 | 3168 | 2.5040 | 1.0030 | | 1.1256 | 45.0 | 3240 | 3.1641 | 0.9608 | | 1.1256 | 46.0 | 3312 | 2.9522 | 0.9677 | | 1.1211 | 47.0 | 3384 | 2.6318 | 0.9580 | | 1.1142 | 48.0 | 3456 | 2.7298 | 0.9533 | | 1.1237 | 49.0 | 3528 | 2.5442 | 0.9673 | | 1.0976 | 50.0 | 3600 | 2.7767 | 0.9610 | | 1.1154 | 51.0 | 3672 | 2.6849 | 0.9646 | | 1.1012 | 52.0 | 3744 | 2.5384 | 0.9621 | | 1.1077 | 53.0 | 3816 | 2.4505 | 1.0067 | | 1.0936 | 54.0 | 3888 | 2.5847 | 0.9687 | | 1.0772 | 55.0 | 3960 | 2.4575 | 0.9761 | | 1.092 | 56.0 | 4032 | 2.4889 | 0.9802 | | 1.0868 | 57.0 | 4104 | 2.5885 | 0.9664 | | 1.0979 | 58.0 | 4176 | 2.6370 | 0.9607 | | 1.094 | 59.0 | 4248 | 2.6195 | 0.9605 | | 1.0745 | 60.0 | 4320 | 2.5346 | 0.9834 | | 1.1057 | 61.0 | 4392 | 2.6879 | 0.9603 | | 1.0722 | 62.0 | 4464 | 2.5426 | 0.9735 | | 1.0731 | 63.0 | 4536 | 2.8259 | 0.9535 | | 1.0862 | 64.0 | 4608 | 2.7632 | 0.9559 | | 1.0396 | 65.0 | 4680 | 2.5401 | 0.9807 | | 1.0581 | 66.0 | 4752 | 2.6977 | 0.9687 | | 1.0647 | 67.0 | 4824 | 2.6968 | 0.9694 | | 1.0549 | 68.0 | 4896 | 2.6439 | 0.9807 | | 1.0607 | 69.0 | 4968 | 2.6822 | 0.9771 | | 1.05 | 70.0 | 5040 | 2.7011 | 0.9607 | | 1.042 | 71.0 | 5112 | 2.5766 | 0.9713 | | 1.042 | 72.0 | 5184 | 2.5720 | 0.9747 | | 1.0594 | 73.0 | 5256 | 2.7176 | 0.9704 | | 1.0425 | 74.0 | 5328 | 2.7458 | 0.9614 | | 1.0199 | 75.0 | 5400 | 2.5906 | 0.9987 | | 1.0198 | 76.0 | 5472 | 2.5534 | 1.0087 | | 1.0193 | 77.0 | 5544 | 2.5421 | 0.9933 | | 1.0379 | 78.0 | 5616 | 2.5139 | 0.9994 | | 1.025 | 79.0 | 5688 | 2.4850 | 1.0313 | | 1.0054 | 80.0 | 5760 | 2.5803 | 0.9814 | | 1.0218 | 81.0 | 5832 | 2.5696 | 0.9867 | | 1.0177 | 82.0 | 5904 | 2.6011 | 1.0065 | | 1.0094 | 83.0 | 5976 | 2.6166 | 0.9855 | | 1.0202 | 84.0 | 6048 | 2.5557 | 1.0204 | | 1.0148 | 85.0 | 6120 | 2.6118 | 1.0033 | | 1.0117 | 86.0 | 6192 | 2.5671 | 1.0120 | | 1.0195 | 87.0 | 6264 | 2.5443 | 1.0041 | | 1.0114 | 88.0 | 6336 | 2.5627 | 1.0049 | | 1.0074 | 89.0 | 6408 | 2.5670 | 1.0255 | | 0.9883 | 90.0 | 6480 | 2.5338 | 1.0306 | | 1.0112 | 91.0 | 6552 | 2.5615 | 1.0142 | | 0.9986 | 92.0 | 6624 | 2.5566 | 1.0415 | | 0.9939 | 93.0 | 6696 | 2.5728 | 1.0287 | | 0.9954 | 94.0 | 6768 | 2.5617 | 1.0138 | | 0.9643 | 95.0 | 6840 | 2.5890 | 1.0145 | | 0.9892 | 96.0 | 6912 | 2.5918 | 1.0119 | | 0.983 | 97.0 | 6984 | 2.5862 | 1.0175 | | 0.988 | 98.0 | 7056 | 2.5873 | 1.0147 | | 0.9908 | 99.0 | 7128 | 2.5973 | 1.0073 | | 0.9696 | 100.0 | 7200 | 2.5938 | 1.0156 | ### Framework versions - Transformers 4.21.0.dev0 - Pytorch 1.9.1+cu102 - Datasets 2.3.3.dev0 - Tokenizers 0.12.1