CoolFace
Modelpublic

Nereboss/WhisperTiny15hCommonVoice5hAugment

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face
0likes3downloads
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. -->

WhisperTiny15hCommonVoice5hAugment

This model is a fine-tuned version of OpenAI/whisper-tiny on the audiofolder dataset. It achieves the following results on the evaluation set:

  • —Loss: nan
  • —Wer: 1.0

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: 1e-05
  • —trainbatchsize: 16
  • —evalbatchsize: 8
  • —seed: 42
  • —optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 500
  • —training_steps: 5000

Training results

Training LossEpochStepValidation LossWer
0.83890.66935001.02650.4802
0.64591.338710000.95290.4519
0.56062.008015000.91140.4389
0.46972.677420000.90710.4558
0.35793.346725000.88380.4251
0.3494.016130000.88780.4238
0.28614.685435000.88600.4266
0.24445.354840000.88920.4266
0.23976.024145000.88750.4237
0.06.69345000nan1.0

Framework versions

  • —Transformers 4.48.3
  • —Pytorch 2.6.0
  • —Datasets 3.3.0
  • —Tokenizers 0.21.0