wav2vec2-large-mms-1b-tun_test

This model is a fine-tuned version of facebook/mms-1b-all on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: inf
• Wer: 0.8202

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.7143	1.0	293	inf	0.9356
2.3549	2.0	586	inf	0.8881
2.2428	3.0	879	inf	0.8788
2.2523	4.0	1172	inf	0.8702
2.1636	5.0	1465	inf	0.8788
2.0924	6.0	1758	inf	0.8405
1.9987	7.0	2051	inf	0.8265
1.9648	8.0	2344	inf	0.8257
1.9403	9.0	2637	inf	0.8292
1.9393	10.0	2930	inf	0.8202

Framework versions

• Transformers 4.41.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1