Zarnabh/whisper-small-2-ps

Whisper small Ps (lowest) - ZFA

This model is a fine-tuned version of openai/whisper-small on the Common Voice 20.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.8034
• Wer: 50.4008

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: 2e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• 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
• lr_scheduler_warmup_steps: 200
• training_steps: 1500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	0.8909	300	0.7274	54.5796
0.064	1.7810	600	0.7439	57.6156
0.064	2.6711	900	0.7716	52.4817
0.0437	3.5612	1200	0.7934	51.0831
0.0075	4.4514	1500	0.8034	50.4008

Framework versions

• Transformers 4.57.3
• Pytorch 2.7.0+cu126
• Datasets 4.4.1
• Tokenizers 0.22.1