metadata

library_name: transformers
language:
  - ta
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - sp03/tamil
metrics:
  - wer
model-index:
  - name: Whisper Small ta - Sp03
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 11.0
          type: sp03/tamil
          config: default
          split: None
          args: 'config: ta, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 100

Whisper Small ta - Sp03

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: 3.6056
Wer: 100.0

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: 0.001
train_batch_size: 4
eval_batch_size: 1
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.4066	1.0	4	9.3832	100.0
21.3162	2.0	8	21.2056	100.0
11.2822	3.0	12	7.6851	100.0
6.0345	4.0	16	4.8468	100.0
3.8909	5.0	20	3.6056	100.0

Framework versions

Transformers 4.47.1
Pytorch 2.5.1+cu124
Datasets 3.2.0
Tokenizers 0.21.0