CoolFace
Modelpublic

speech31/wav2vec2-large-TIMIT-IPA2

sourceHugging Faceapache-2.0updated 4y agoView on Hugging Face
1likes52downloads
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-large-TIMIT-IPA2

This model is a fine-tuned version of facebook/wav2vec2-large on the None dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.1531
  • —Per: 0.0638

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: 64
  • —evalbatchsize: 16
  • —seed: 42
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 1000
  • —num_epochs: 100
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossPer
2.08466.855000.18100.0991
0.185713.710000.14110.0691
0.094820.5515000.13450.0666
0.064627.420000.14440.0673
0.050234.2525000.14360.0628
0.038141.130000.13830.0637
0.030947.9535000.15330.0638
0.026154.7940000.15310.0638

Framework versions

  • —Transformers 4.20.0
  • —Pytorch 1.12.1+cu113
  • —Datasets 2.6.2.dev0
  • —Tokenizers 0.12.1