Hugging Face's logo

gary109
/
ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1-5gram

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
gary109/AI_Light_Dance
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
ai-light-dance_singing3_ft_wav2vec2-large-xlsr-53-v1-5gram / README.md
gary109's picture
gary109
update model card README.md
9357af2
preview code
|
raw
history blame
7.64 kB
metadata
license: apache-2.0
tags:
  - automatic-speech-recognition
  - gary109/AI_Light_Dance
  - 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 on the GARY109/AI_LIGHT_DANCE - ONSET-SINGING3 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4695
  • Wer: 0.2106

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
0.3993 1.0 72 0.5761 0.2678
0.4252 2.0 144 0.5669 0.2987
0.4309 3.0 216 0.5532 0.2914
0.4151 4.0 288 0.5907 0.2773
0.4229 5.0 360 0.5353 0.2502
0.4229 6.0 432 0.5442 0.2607
0.4106 7.0 504 0.5387 0.2513
0.4161 8.0 576 0.5407 0.2592
0.4072 9.0 648 0.5012 0.2493
0.4301 10.0 720 0.5449 0.2542
0.419 11.0 792 0.5025 0.2558
0.4336 12.0 864 0.5267 0.2451
0.4188 13.0 936 0.5068 0.2582
0.4005 14.0 1008 0.5392 0.2529
0.4213 15.0 1080 0.5083 0.2492
0.4123 16.0 1152 0.4909 0.2481
0.3854 17.0 1224 0.5270 0.2593
0.3887 18.0 1296 0.5059 0.2503
0.3934 19.0 1368 0.4929 0.2353
0.4005 20.0 1440 0.5170 0.2444
0.4018 21.0 1512 0.4931 0.2406
0.3981 22.0 1584 0.4988 0.2345
0.3645 23.0 1656 0.5221 0.2495
0.3727 24.0 1728 0.5192 0.2361
0.3751 25.0 1800 0.4996 0.2403
0.3839 26.0 1872 0.4840 0.2227
0.3739 27.0 1944 0.4790 0.2265
0.3781 28.0 2016 0.5081 0.2247
0.3655 29.0 2088 0.4903 0.2263
0.3832 30.0 2160 0.5196 0.2383
0.3662 31.0 2232 0.5286 0.2363
0.3692 32.0 2304 0.5283 0.2460
0.3484 33.0 2376 0.4906 0.2267
0.3493 34.0 2448 0.4983 0.2214
0.3783 35.0 2520 0.5359 0.2319
0.3604 36.0 2592 0.5129 0.2371
0.3796 37.0 2664 0.4994 0.2320
0.3629 38.0 2736 0.4726 0.2226
0.3428 39.0 2808 0.5130 0.2225
0.375 40.0 2880 0.5155 0.2285
0.3581 41.0 2952 0.4994 0.2267
0.3394 42.0 3024 0.4989 0.2177
0.3421 43.0 3096 0.4820 0.2274
0.342 44.0 3168 0.4839 0.2189
0.3482 45.0 3240 0.4979 0.2120
0.3494 46.0 3312 0.5069 0.2186
0.345 47.0 3384 0.4998 0.2321
0.327 48.0 3456 0.4978 0.2189
0.3266 49.0 3528 0.5079 0.2187
0.3353 50.0 3600 0.5302 0.2235
0.3384 51.0 3672 0.4838 0.2117
0.3259 52.0 3744 0.5022 0.2169
0.336 53.0 3816 0.4834 0.2214
0.3123 54.0 3888 0.5113 0.2164
0.3306 55.0 3960 0.4770 0.2112
0.3344 56.0 4032 0.4752 0.2130
0.3279 57.0 4104 0.4953 0.2123
0.3067 58.0 4176 0.4823 0.2175
0.3131 59.0 4248 0.4857 0.2137
0.3178 60.0 4320 0.4969 0.2154
0.3306 61.0 4392 0.4816 0.2166
0.3152 62.0 4464 0.4695 0.2106
0.3119 63.0 4536 0.5111 0.2106
0.3124 64.0 4608 0.5017 0.2119
0.317 65.0 4680 0.5071 0.2165
0.2934 66.0 4752 0.4819 0.2168
0.3103 67.0 4824 0.4944 0.2143
0.3078 68.0 4896 0.5004 0.2110
0.3084 69.0 4968 0.4998 0.2148
0.3147 70.0 5040 0.4885 0.2099
0.3081 71.0 5112 0.5062 0.2096
0.3046 72.0 5184 0.4790 0.2124
0.2993 73.0 5256 0.5045 0.2096
0.2844 74.0 5328 0.4865 0.2060
0.3061 75.0 5400 0.4997 0.2103
0.305 76.0 5472 0.5118 0.2120
0.3015 77.0 5544 0.4900 0.2096
0.2969 78.0 5616 0.4984 0.2106
0.2988 79.0 5688 0.4948 0.2065
0.3006 80.0 5760 0.4850 0.2032
0.2982 81.0 5832 0.4933 0.2047
0.2938 82.0 5904 0.4997 0.2052
0.2903 83.0 5976 0.4919 0.2034
0.297 84.0 6048 0.4807 0.2026
0.2955 85.0 6120 0.4953 0.2048
0.2847 86.0 6192 0.4951 0.2022
0.2891 87.0 6264 0.4942 0.2036
0.2895 88.0 6336 0.4888 0.2020
0.2823 89.0 6408 0.4968 0.2033
0.2885 90.0 6480 0.5042 0.2040
0.2876 91.0 6552 0.4943 0.2025
0.2765 92.0 6624 0.4961 0.2004
0.282 93.0 6696 0.4847 0.2017
0.277 94.0 6768 0.4849 0.1994
0.2883 95.0 6840 0.4849 0.2003
0.2906 96.0 6912 0.4912 0.2002
0.2781 97.0 6984 0.4874 0.2016
0.2836 98.0 7056 0.4825 0.1996
0.2676 99.0 7128 0.4806 0.1995
0.28 100.0 7200 0.4814 0.1994

Framework versions

  • Transformers 4.21.0.dev0
  • Pytorch 1.9.1+cu102
  • Datasets 2.3.3.dev0
  • Tokenizers 0.12.1