wav2vec2-large-xlsr-53-second-train

This model is a fine-tuned version of fathyy/wav2vec2-large-xlsr-53-first-try on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.3022
• Wer: 0.9529

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: 0.001
• train_batch_size: 8
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1200
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.0702	2.6667	500	3.0498	1.0
3.003	5.3333	1000	3.0485	1.0
2.9097	8.0	1500	2.9847	1.0
2.7703	10.6667	2000	2.8414	0.9912
2.601	13.3333	2500	2.8046	0.9706
2.4095	16.0	3000	2.4821	0.9529
2.2222	18.6667	3500	2.3022	0.9529

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0