test_pelby_v3

This model is a fine-tuned version of Osolon/wav2vec2-large-xls-r-300m-pl on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: inf
• Wer: 0.0850

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: 8
• eval_batch_size: 4
• 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: 100
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2464	0.1241	400	inf	0.2219
0.2056	0.2481	800	inf	0.1526
0.1711	0.3722	1200	inf	0.1356
0.1399	0.4963	1600	inf	0.1103
0.1453	0.6203	2000	inf	0.1087
0.1221	0.7444	2400	inf	0.0938
0.1122	0.8685	2800	inf	0.0877
0.109	0.9926	3200	inf	0.0850

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1