hubert1

This model is a fine-tuned version of facebook/hubert-base-ls960 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.5840
• Wer: 1.0

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: 4
• 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: 1000
• num_epochs: 7
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
8.3756	0.6821	500	5.6168	1.0
4.8753	1.3643	1000	3.5609	1.0
4.1974	2.0464	1500	3.5191	1.0
3.9847	2.7285	2000	3.5168	1.0
3.9535	3.4106	2500	3.6024	1.0
3.8797	4.0928	3000	3.4174	1.0
3.935	4.7749	3500	3.5542	1.0
3.7183	5.4570	4000	3.5183	1.0
3.7562	6.1392	4500	3.5689	1.0
3.7423	6.8213	5000	3.5840	1.0

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1