Edit model card

Whisper Small ta - EswarDivi

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2420
  • Wer: 58.9959

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
  • train_batch_size: 16
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 4000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.1954 0.75 1000 0.2666 64.2226
0.1276 1.5 2000 0.2475 56.4822
0.0755 2.26 3000 0.2451 58.8562
0.0684 3.01 4000 0.2420 58.9959

Framework versions

  • Transformers 4.28.0.dev0
  • Pytorch 2.0.0+cu117
  • Datasets 2.10.1
  • Tokenizers 0.13.2
Downloads last month
3

Dataset used to train eswardivi/whisper-small-ta

Space using eswardivi/whisper-small-ta 1

Evaluation results