msrpaza

This model is a fine-tuned version of microsoft/paza-whisper-large-v3-turbo on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2267
• Wer: 17.43

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: 48
• eval_batch_size: 64
• seed: 42
• 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: 100
• num_epochs: 7

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1862	0.9843	250	0.2480	19.19
0.1467	1.9685	500	0.2287	17.81
0.1192	2.9528	750	0.2250	17.53
0.1032	3.9370	1000	0.2255	17.51
0.0905	4.9213	1250	0.2260	17.46
0.0892	5.9055	1500	0.2272	17.32
0.0848	6.8898	1750	0.2267	17.32
0.0905	7.0	1778	0.2267	17.43

Framework versions

• Transformers 5.16.1
• Pytorch 2.11.0+cu128
• Datasets 4.8.5
• Tokenizers 0.23.1