miosipof/asr_EN_medium_v1

Whisper Medium

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

• Loss: 0.3170
• Wer: 20.5788

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: 0.001
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 32
• training_steps: 128
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.8843	1.0847	32	0.8819	135.0482
0.3624	2.1695	64	0.3312	47.1061
0.1637	3.2542	96	0.3231	22.1865
0.0903	4.3390	128	0.3170	20.5788

Framework versions

• PEFT 0.12.0
• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1