arun100/whisper-small-vi-2

Whisper Small Vienamese

This model is a fine-tuned version of openai/whisper-small on the mozilla-foundation/common_voice_16_0 vi dataset. It achieves the following results on the evaluation set:

• Loss: 0.6705
• Wer: 24.5680

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-06
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• training_steps: 2500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0174	33.0	500	0.6207	24.6696
0.0045	66.0	1000	0.6705	24.5680
0.0027	99.01	1500	0.6945	25.2795
0.002	133.0	2000	0.7079	26.4790
0.0018	166.0	2500	0.7127	26.3976

Framework versions

• Transformers 4.37.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 2.16.2.dev0
• Tokenizers 0.15.0