output_model_whisper_base_shunya_ideal_data_augmentation_enabled

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

• Loss: 0.5311
• Wer: 24.5163

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: 0.0001
• train_batch_size: 2
• eval_batch_size: 4
• 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: 500
• training_steps: 20000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.6698	1.5625	2000	0.9466	47.5705
0.2225	3.125	4000	0.7431	43.1251
0.1520	4.6875	6000	0.6466	36.1985
0.0715	6.25	8000	0.6520	36.2871
0.0582	7.8125	10000	0.6448	34.0127
0.0200	9.375	12000	0.6326	31.3691
0.0082	10.9375	14000	0.6005	29.0799
0.0082	12.5	16000	0.5570	26.8055
0.0172	14.0625	18000	0.5496	25.0480
0.0006	15.625	20000	0.5311	24.5163

Framework versions

• Transformers 5.3.0.dev0
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2