misterkissi/whisper-small-lomwe

wisper-small-lomwe

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

• Loss: 0.9452
• Wer: 0.5130

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: 64
• eval_batch_size: 16
• 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: 500
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.7718	41.6667	500	0.7511	0.4298
0.002	83.3333	1000	0.8245	0.4611
0.0004	125.0	1500	0.8562	0.4593
0.0002	166.6667	2000	0.8773	0.4924
0.0001	208.3333	2500	0.8991	0.4861
0.0001	250.0	3000	0.9142	0.4835
0.0001	291.6667	3500	0.9251	0.4826
0.0001	333.3333	4000	0.9336	0.4808
0.0001	375.0	4500	0.9421	0.5147
0.0001	416.6667	5000	0.9452	0.5130

Framework versions

• Transformers 4.53.2
• Pytorch 2.6.0+cu124
• Datasets 2.18.0
• Tokenizers 0.21.2