CoolFace
Modelpublic

EzraWilliam/wav2vec2-xlsr-53-CV-demo-google-colab-Ezra_William_Prod19

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face
0likes29downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

wav2vec2-xlsr-53-CV-demo-google-colab-EzraWilliamProd19

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the commonvoice13_0 dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.3817
  • —Wer: 0.3751

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: 0.0001
  • —trainbatchsize: 16
  • —evalbatchsize: 8
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —num_epochs: 9
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossWer
2.94611.02782.92131.0
2.87072.05562.78421.0
1.28353.08340.66520.6095
0.7314.011120.49220.4826
0.62995.013900.44290.4414
0.51396.016680.41230.4020
0.47857.019460.39320.3874
0.46368.022240.38950.3798
0.43819.025020.38170.3751

Framework versions

  • —Transformers 4.40.1
  • —Pytorch 2.3.0+cu121
  • —Datasets 2.19.0
  • —Tokenizers 0.19.1