whisper-small-clean_6-v4

This model is a fine-tuned version of openai/whisper-small on the lyhourt/clean_6 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2886
• Wer: 24.0149

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: 50
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1207	0.5	200	0.3080	25.4138
0.1827	1.0	400	0.2953	24.7144
0.0907	1.1342	600	0.2921	24.3413
0.0904	1.5123	800	0.2900	24.3064
0.0823	1.8904	1000	0.2886	24.0149

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1