whisper-tiny-no-is-fo-100h-5k-steps_v2

This model is a fine-tuned version of davidilag/whisper-tiny-no-is-5k-steps on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4825
• Wer: 37.0143

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
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.8277	0.2320	1000	0.7912	52.0838
0.6065	0.4640	2000	0.6015	43.4316
0.5406	0.6961	3000	0.5281	39.9990
0.5429	0.9281	4000	0.4930	37.6736
0.4447	1.1601	5000	0.4825	37.0143

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1