AkshitSaxena1/whisper-tiny_chinese_accent_v4

Whisper tiny Chinese

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

• Loss: 0.4727
• Wer: 17.6648

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: 3.75e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• training_steps: 1500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0440	0.5128	100	0.3567	14.8848
0.0645	1.0256	200	0.4228	16.1716
0.0293	1.5385	300	0.4174	15.4885
0.0272	2.0513	400	0.4532	17.9984
0.0151	2.5641	500	0.4478	17.0770
0.0139	3.0769	600	0.4727	17.6648

Framework versions

• Transformers 5.5.0.dev0
• Pytorch 2.10.0+cu128
• Datasets 4.8.4
• Tokenizers 0.22.2