whisper-medium-swagen-male-5hrs-52

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

• Loss: 0.5615
• Wer: 0.3048

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: 2
• eval_batch_size: 2
• seed: 52
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• 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: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.6491	0.4840	200	0.8047	0.4599
0.5128	0.9679	400	0.6416	0.3715
0.3122	1.4501	600	0.6107	0.3646
0.2424	1.9341	800	0.5730	0.3276
0.1145	2.4162	1000	0.5764	0.3246
0.1273	2.9002	1200	0.5586	0.3141
0.0423	3.3823	1400	0.5821	0.3174
0.0526	3.8663	1600	0.5872	0.3364
0.0202	4.3485	1800	0.6037	0.3033
0.0192	4.8324	2000	0.5869	0.3035

Framework versions

• Transformers 4.53.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.0