whisper_medium_newds

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

• Loss: 0.6611
• Cer: 19.0911
• Wer: 30.7521

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: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer	Wer
1.4986	1.0	2598	0.7039	22.7752	36.6440
0.7329	2.0	5196	0.6673	19.6293	31.8468
0.6202	3.0	7794	0.6611	19.0911	30.7521

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2+cu118
• Datasets 2.19.0
• Tokenizers 0.19.1