hotien2107/whisper-small-vi

Whisper Small Vi - VT

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.7620
• Wer: 28.6325

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: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1304	2.8736	500	0.5556	28.9505
0.0258	5.7471	1000	0.6184	29.3563
0.0051	8.6207	1500	0.6678	29.0931
0.0015	11.4943	2000	0.6893	28.2926
0.0004	14.3678	2500	0.7158	28.5558
0.0002	17.2414	3000	0.7309	28.4900
0.0002	20.1149	3500	0.7452	28.4571
0.0002	22.9885	4000	0.7527	28.4790
0.0002	25.8621	4500	0.7596	28.5887
0.0001	28.7356	5000	0.7620	28.6325

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu124
• Datasets 3.4.0
• Tokenizers 0.21.0