joey234/whisper-medium-vi

Whisper Medium Vietnamese

This model is a fine-tuned version of openai/whisper-medium on the mozilla-foundation/common_voice_11_0 vi dataset. It achieves the following results on the evaluation set:

• Loss: 0.7136
• Wer: 15.4925

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: 64
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• 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
0.0001	124.0	1000	0.7136	15.4925
0.0001	249.0	2000	0.8532	17.0045
0.0	374.0	3000	0.9251	19.0972
0.0	499.0	4000	0.9787	21.5953
0.0	624.0	5000	0.9921	21.4638

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0
• Datasets 2.7.1
• Tokenizers 0.12.1