CoolFace
Modelpublic

Gummybear05/wav2vec2-1b-Elderly

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face
0likes6downloads
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. -->

wav2vec2-1b-Elderly

This model is a fine-tuned version of facebook/wav2vec2-xls-r-1b on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4157
  • Cer: 11.1137

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: 1
  • evalbatchsize: 8
  • seed: 42
  • gradientaccumulationsteps: 16
  • totaltrainbatch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lrschedulertype: linear
  • lrschedulerwarmup_steps: 50
  • num_epochs: 3
  • mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossCer
8.6450.25802002.262543.7735
1.49370.51614001.231127.9488
1.09810.77416000.980123.6020
0.8981.03228000.805020.0658
0.72841.290210000.732018.3447
0.66441.548312000.609215.7484
0.5921.806314000.599615.8893
0.49262.064316000.565414.6147
0.40542.322418000.477412.5822
0.39022.580420000.444611.9772
0.3612.838522000.415711.1137

Framework versions

  • Transformers 4.45.2
  • Pytorch 2.3.1.post100
  • Datasets 2.19.1
  • Tokenizers 0.20.1