base-asr-with-tpbs-cv4

This model is a fine-tuned version of Botnoi/base-asr-with-lm on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5538
• Wer: 0.2576
• Cer: 0.0695
• Clean Cer: 0.0599

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer	Clean Cer
0.1306	26.67	400	0.4897	0.2766	0.0733	0.0642
0.0437	53.33	800	0.4923	0.2566	0.0685	0.0596
0.0293	80.0	1200	0.5538	0.2576	0.0695	0.0599

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.1+cu117
• Datasets 2.11.0
• Tokenizers 0.13.2