idaracecilia/whisper-finetuned-small

Whisper Fine tuned Small

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

• Loss: 0.1965
• Wer: 11.2724

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2
• training_steps: 102
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1398	0.4	20	0.2211	13.1623
0.0941	0.8	40	0.2144	11.8124
0.048	1.2	60	0.1997	11.2386
0.0481	1.6	80	0.1979	11.3736
0.0337	2.0	100	0.1965	11.2724

Framework versions

• Transformers 4.44.0
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1