Whisper Medium Vi v1 - Shiv Kumar Ganesh

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: 1.0641
  • Wer: 34.0974

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: 32
  • eval_batch_size: 16
  • 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: 7000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.0005 31.0 500 0.7179 33.7464
0.0002 62.0 1000 0.7837 32.4742
0.0001 93.0 1500 0.8267 34.2729
0.0001 124.0 2000 0.8677 35.1722
0.0 156.0 2500 0.9045 35.3257
0.0 187.0 3000 0.9316 33.9877
0.0 218.0 3500 0.9585 34.0097
0.0 249.0 4000 0.9846 33.3626
0.0 281.0 4500 1.0082 33.4832
0.0 312.0 5000 1.0247 33.7026
0.0 343.0 5500 1.0391 32.8691
0.0 374.0 6000 1.0516 32.9020
0.0 406.0 6500 1.0606 33.6477
0.0 437.0 7000 1.0641 34.0974

Framework versions

  • Transformers 4.26.0.dev0
  • Pytorch 1.13.0+cu116
  • Datasets 2.8.1.dev0
  • Tokenizers 0.13.2
Downloads last month
7
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train shivkumarganesh/whisper-small-vi-v1

Evaluation results