speecht5_finetuned_voxpopuli_hy

This model is a fine-tuned version of Edmon02/speecht5_finetuned_voxpopuli_hy on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4910

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: 2e-05
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5849	33.8983	500	0.5350
0.5475	67.7966	1000	0.5175
0.5241	101.6949	1500	0.5054
0.5105	135.5932	2000	0.4974
0.5013	169.4915	2500	0.4956
0.4971	203.3898	3000	0.4913
0.4874	237.2881	3500	0.4884
0.4872	271.1864	4000	0.4910

Framework versions

• Transformers 4.43.3
• Pytorch 2.4.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1