CoolFace
Modelpublic

hrishikeshpai30/wavlm-libri-clean-100h-large

sourceHugging Faceupdated 3y agoView on Hugging Face
1likes8downloads
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. -->

wavlm-libri-clean-100h-large

This model is a fine-tuned version of microsoft/wavlm-large on the AHAZEEMI/LIBRISPEECH10H - CLEAN dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.0893
  • —Wer: 0.0655

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: 0.0003
  • —trainbatchsize: 2
  • —evalbatchsize: 2
  • —seed: 42
  • —gradientaccumulationsteps: 2
  • —totaltrainbatch_size: 4
  • —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 500
  • —num_epochs: 3.0

Training results

Training LossEpochStepValidation LossWer
0.01440.423000.09470.0749
0.14080.846000.13470.1363
0.03961.269000.10900.0935
0.03531.6812000.10320.0832
0.0512.115000.09690.0774
0.02542.5218000.09300.0715
0.05792.9421000.08940.0660

Framework versions

  • —Transformers 4.26.1
  • —Pytorch 2.0.0+cpu
  • —Datasets 2.9.0
  • —Tokenizers 0.13.2