speecht5_Mehdi_Final_Model_2nd

This model is a fine-tuned version of MehdiAslam/speecht5_Mehdi_Final_Model on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4152

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 200
• training_steps: 3000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.3962	3.2924	250	0.4002
0.3925	6.5847	500	0.4058
0.3924	9.8771	750	0.4070
0.3851	13.1595	1000	0.4048
0.3803	16.4518	1250	0.4059
0.3829	19.7442	1500	0.4059
0.3774	23.0266	1750	0.4034
0.3723	26.3189	2000	0.4096
0.3706	29.6113	2250	0.4122
0.3789	32.9037	2500	0.4171
0.3688	36.1860	2750	0.4156
0.3721	39.4784	3000	0.4152

Framework versions

• Transformers 4.52.2
• Pytorch 2.6.0+cu124
• Datasets 2.14.4
• Tokenizers 0.21.1