ASR_Whisper_Disease_General

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

• Loss: 0.1755
• Cer: 24.2276
• Wer: 29.8574

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 8
• 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: 339
• training_steps: 6780
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer	Wer
0.3018	0.4423	1000	0.3222	67.9472	69.5226
0.2378	0.8846	2000	0.2460	31.9183	42.8595
0.1207	1.3268	3000	0.2177	34.8328	37.2898
0.1288	1.7691	4000	0.1923	26.9850	33.9050
0.0438	2.2114	5000	0.1850	28.4133	34.9564
0.0375	2.6537	6000	0.1755	24.2276	29.8574

Framework versions

• Transformers 4.53.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1