metadata

library_name: transformers
language:
  - ur
license: apache-2.0
base_model: GogetaBlueMUI/whisper-medium-ur-jalandhary
tags:
  - generated_from_trainer
datasets:
  - mirfan899/jalandhary_asr
metrics:
  - wer
model-index:
  - name: Whisper Medium Ur - Jalandhary ASR Fine-Tuned
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Jalandhary ASR
          type: mirfan899/jalandhary_asr
          args: 'split: test'
        metrics:
          - name: Wer
            type: wer
            value: 19.807797769827385

Whisper Medium Ur - Jalandhary ASR Fine-Tuned

This model is a fine-tuned version of GogetaBlueMUI/whisper-medium-ur-jalandhary on the Jalandhary ASR dataset. It achieves the following results on the evaluation set:

Loss: 0.1012
Wer: 19.8078

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-06
train_batch_size: 8
eval_batch_size: 8
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: 300
training_steps: 2400
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1097	0.5831	600	0.1066	18.6509
0.0664	1.1662	1200	0.1020	19.1575
0.0821	1.7493	1800	0.1016	19.2725
0.0567	2.3324	2400	0.1012	19.8078

Framework versions

Transformers 4.49.0
Pytorch 2.5.1+cu121
Datasets 3.3.2
Tokenizers 0.21.0