CoolFace
Modelpublic

Nereboss/WhisperTiny15hCommonVoice45hAugmentV2

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. -->

WhisperTiny15hCommonVoice45hAugmentV2

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: 0.8412
  • —Wer: 0.4180

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
0.95280.22335000.99530.4696
0.77940.446610000.91610.4818
0.68880.669915000.88450.4304
0.59050.893320000.86640.4301
0.45041.116625000.85530.4267
0.45211.339930000.85200.4231
0.40611.563235000.84580.4194
0.391.786540000.84750.4179
0.38822.009845000.84400.4168
0.32372.233150000.84120.4180

Framework versions

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