whisper-small-zh

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

Whisper smail zh - Song train

This model is a fine-tuned version of openai/whisper-small on the Chinese songs * 58 dataset. It achieves the following results on the evaluation set:

• Loss: 2.3571
• Wer: 23.1198

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: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• 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: 200
• training_steps: 10000
• mixed_precision_training: Native AMP
• label_smoothing_factor: 0.1

Training results

Framework versions

• Transformers 4.56.2
• Pytorch 2.7.1+cu118
• Datasets 4.1.1
• Tokenizers 0.22.1