wme_30s_Static_atWall_1.3

This model is a fine-tuned version of openai/whisper-medium.en on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.9007
• Wer: 27.5260

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: 4e-05
• train_batch_size: 48
• eval_batch_size: 32
• seed: 42
• 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: 26
• training_steps: 264
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	0	0	1.5148	34.2009
0.8009	0.1667	44	1.0184	32.0882
0.5762	0.3333	88	0.9739	29.8836
0.553	0.5	132	0.9501	32.0269
0.4237	1.0038	176	0.9252	29.4550
0.2473	1.1705	220	0.9069	30.1286
0.1507	1.3371	264	0.9007	27.5260

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1