CoolFace
Modelpublic

JacobLinCool/whisper-large-v3-turbo-common_voice_16_1-zh-TW-pissa-merged

sourceHugging Facemitupdated 2y agoView on Hugging Face
0likes8downloads
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. -->

whisper-large-v3-turbo-commonvoice16_1-zh-TW-pissa

This model is a fine-tuned version of openai/whisper-large-v3-turbo on the JacobLinCool/mozilla-foundation-commonvoice16_1-zh-TW-preprocessed dataset. It achieves the following results on the evaluation set:

  • —Loss: 0.5133
  • —Wer: 63.6656
  • —Cer: 23.5752

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: 0.0005
  • —trainbatchsize: 4
  • —evalbatchsize: 4
  • —seed: 42
  • —gradientaccumulationsteps: 8
  • —totaltrainbatch_size: 32
  • —optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
  • —lrschedulertype: linear
  • —num_epochs: 10

Training results

Training LossEpochStepValidation LossWerCer
No log002.752077.612520.7783
7.69820.99873770.874487.942141.2804
5.16772.07550.749982.596536.6407
3.36472.998711320.643376.808731.6068
3.47114.015100.639776.246030.2862
1.56944.998718870.577971.543427.5471
0.79516.022650.566471.322327.0600
0.47096.998726420.549268.870626.0131
0.1168.030200.542766.760524.8104
0.05128.998733970.529866.137524.8632
0.02739.986837700.513363.665623.5752

Framework versions

  • —PEFT 0.13.2
  • —Transformers 4.46.0
  • —Pytorch 2.4.0
  • —Datasets 3.0.2
  • —Tokenizers 0.20.1