whisper-medium-chinese-tw-minnan

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

• Loss: 1.0459
• Wer: 75.3550
• Cer: 42.1433

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• 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	Cer
0.0997	3.6381	1000	1.0399	82.0487	48.6755
0.0131	7.2735	2000	1.0279	75.9635	43.4377
0.0009	10.9116	3000	1.0553	75.4564	43.2872
0.0001	14.5469	4000	1.0423	75.5578	42.6851
0.0	18.1823	5000	1.0459	75.3550	42.1433

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1