Hugging Face's logo

gary109
/
ai-light-dance_singing6_ft_wav2vec2-large-xlsr-53-5gram-v4-2

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
gary109/AI_Light_Dance
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
1
ai-light-dance_singing6_ft_wav2vec2-large-xlsr-53-5gram-v4-2 / README.md
gary109's picture
gary109
update model card README.md
8b4f6a3
preview code
|
raw
history blame
4.64 kB
metadata
tags:
  - automatic-speech-recognition
  - gary109/AI_Light_Dance
  - generated_from_trainer
model-index:
  - name: ai-light-dance_singing6_ft_wav2vec2-large-xlsr-53-5gram-v4-2
    results: []

ai-light-dance_singing6_ft_wav2vec2-large-xlsr-53-5gram-v4-2

This model is a fine-tuned version of gary109/ai-light-dance_singing4_ft_wav2vec2-large-xlsr-53-5gram-v4-2 on the GARY109/AI_LIGHT_DANCE - ONSET-SINGING6 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1803
  • Wer: 0.0981

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: 4e-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: 50.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.6742 1.0 171 0.2202 0.1309
0.6303 2.0 342 0.2230 0.1305
0.6036 3.0 513 0.1992 0.1087
0.6224 4.0 684 0.2169 0.1221
0.5761 5.0 855 0.2380 0.1452
0.5694 6.0 1026 0.2175 0.1233
0.5753 7.0 1197 0.2028 0.1085
0.5503 8.0 1368 0.2122 0.1193
0.5322 9.0 1539 0.2042 0.1170
0.6088 10.0 1710 0.2084 0.1157
0.561 11.0 1881 0.2091 0.1137
0.5319 12.0 2052 0.1959 0.1091
0.5272 13.0 2223 0.2109 0.1176
0.5102 14.0 2394 0.2037 0.1131
0.5401 15.0 2565 0.2039 0.1151
0.5146 16.0 2736 0.2046 0.1132
0.5316 17.0 2907 0.2135 0.1197
0.5099 18.0 3078 0.2059 0.1109
0.5391 19.0 3249 0.2007 0.1118
0.5323 20.0 3420 0.1973 0.1049
0.5253 21.0 3591 0.1938 0.1058
0.5016 22.0 3762 0.2034 0.1126
0.4687 23.0 3933 0.2016 0.1127
0.4843 24.0 4104 0.2009 0.1119
0.4889 25.0 4275 0.1903 0.1025
0.5106 26.0 4446 0.1940 0.1072
0.4861 27.0 4617 0.1854 0.1012
0.4883 28.0 4788 0.1880 0.1018
0.4805 29.0 4959 0.1956 0.1052
0.4905 30.0 5130 0.1964 0.1055
0.4776 31.0 5301 0.1865 0.1000
0.496 32.0 5472 0.1900 0.1029
0.4634 33.0 5643 0.1886 0.1005
0.4596 34.0 5814 0.1858 0.1014
0.4709 35.0 5985 0.1877 0.1033
0.4576 36.0 6156 0.1910 0.1045
0.4928 37.0 6327 0.1835 0.1018
0.4585 38.0 6498 0.1857 0.0999
0.4613 39.0 6669 0.1874 0.1036
0.4759 40.0 6840 0.1858 0.1019
0.4681 41.0 7011 0.1830 0.0991
0.4603 42.0 7182 0.1869 0.1003
0.456 43.0 7353 0.1809 0.0980
0.4628 44.0 7524 0.1840 0.0996
0.4391 45.0 7695 0.1814 0.0960
0.4372 46.0 7866 0.1817 0.0976
0.4542 47.0 8037 0.1816 0.0994
0.4495 48.0 8208 0.1819 0.0988
0.4182 49.0 8379 0.1803 0.0981
0.454 50.0 8550 0.1803 0.0993

Framework versions

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