metadata

library_name: transformers
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
datasets:
  - abdouaziiz/dyula_train
metrics:
  - wer
model-index:
  - name: whisper-small_dyu
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: abdouaziiz/dyula_train
          type: abdouaziiz/dyula_train
        metrics:
          - name: Wer
            type: wer
            value: 0.8582430119794637

whisper-small_dyu

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

Loss: 1.6047
Wer: 0.8582

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: 0.0002
train_batch_size: 12
eval_batch_size: 12
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 40
training_steps: 16000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.3886	1.4881	1000	1.6487	0.8525
0.9518	2.9762	2000	1.6047	0.8582
0.547	4.4643	3000	1.8343	0.8620
0.3519	5.9524	4000	1.9764	0.8331
0.2128	7.4405	5000	2.1710	0.8491

Framework versions

Transformers 4.47.0.dev0
Pytorch 2.5.0+cu124
Datasets 3.0.2
Tokenizers 0.20.1