w_small

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

• Loss: 0.7832
• Wer: 82.6298

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.9114	0.4548	1000	0.8773	80.5271
0.8239	0.9095	2000	0.8073	72.1577
0.6064	1.3643	3000	0.7840	74.4663
0.6283	1.8190	4000	0.7717	78.3562
0.5439	2.2738	5000	0.7827	78.6556
0.5574	2.7285	6000	0.7720	71.1815
0.454	3.1833	7000	0.7840	89.8216
0.4246	3.6380	8000	0.7832	82.6298

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu118
• Datasets 3.0.0
• Tokenizers 0.19.1