efik-whisper

This model is a fine-tuned version of offiongbassey/efik-asr on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1576

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 500
• num_epochs: 6
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.2799	0.4193	100	0.1487
0.2567	0.8386	200	0.1377
0.1838	1.2558	300	0.1434
0.1970	1.6751	400	0.1477
0.1627	2.0922	500	0.1550
0.1272	2.5115	600	0.1520
1.0981	2.9308	700	0.6144
0.1212	3.3480	800	0.1595
0.1055	3.7673	900	0.1635
0.0623	4.1845	1000	0.1671
0.0628	4.6038	1100	0.1620
0.0478	5.0210	1200	0.1571
0.0327	5.4403	1300	0.1598
0.0241	5.8595	1400	0.1576

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2