CoolFace
Modelpublic

thunderboltc/whisper-small-santali-sanlish-revised2_211sen_withouthyphen

sourceHugging Faceapache-2.0updated 2mo 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-small-santali-sanlish-revised2211senwithouthyphen

This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:

  • —Loss: 1.1378
  • —Wer: 51.7241
  • —Cer: 15.4955

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: 4
  • —evalbatchsize: 4
  • —seed: 42
  • —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • —lrschedulertype: linear
  • —lrschedulerwarmup_steps: 25
  • —num_epochs: 20
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossWerCer
2.27621.0421.151982.758622.5225
0.62592.0840.925452.586217.2973
0.21533.01260.953453.448317.2973
0.08434.01680.969949.137915.4955
0.03865.02101.067666.379318.0180
0.03106.02521.016954.310315.6757
0.01227.02940.993149.137915.8559
0.00418.03361.064253.448316.2162
0.00889.03781.082152.586215.8559
0.001010.04201.147254.310316.3964
0.000811.04621.128153.448316.0360
0.000512.05041.137851.724115.4955

Framework versions

  • —Transformers 5.13.1
  • —Pytorch 2.11.0+cu128
  • —Datasets 4.0.0
  • —Tokenizers 0.22.2