EdwardFang09/whisper-base-TA-2025

CIT Smart Personal Assistant v1 2025

This model is a fine-tuned version of openai/whisper-base on the IEE4912_Dataset dataset. It achieves the following results on the evaluation set:

• Loss: nan
• Wer: 100.0

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: 32
• eval_batch_size: 16
• 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: 125
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0	62.5	250	0.0048	0.0
0.0	125.0	500	nan	100.0
0.0	187.5	750	nan	100.0
0.0	250.0	1000	nan	100.0

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu126
• Datasets 3.4.1
• Tokenizers 0.21.0