whisper-small-vi

This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1870
• Wer: 21.6303

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: 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: 500
• training_steps: 2000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2573	0.32	200	0.2834	29.3170
0.2007	0.64	400	0.2387	22.8230
0.1962	0.96	600	0.2174	20.1278
0.137	1.28	800	0.2048	21.7818
0.126	1.6	1000	0.1947	22.1146
0.1261	1.92	1200	0.1870	21.6303

Framework versions

• Transformers 4.48.0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4