drownnila/Ergonomic_TTS_ID_v3

Ergonomic_TTS_ID_v2_v3

This model is a fine-tuned version of drownnila/Ergonomic_TTS_ID_v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4585

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: 1e-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_FUSED 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: 10000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5004	1.2469	1000	0.4644
0.4953	2.4938	2000	0.4644
0.4938	3.7408	3000	0.4598
0.4912	4.9877	4000	0.4607
0.4889	6.2345	5000	0.4612
0.4873	7.4814	6000	0.4619
0.4858	8.7283	7000	0.4577
0.4844	9.9752	8000	0.4587
0.4827	11.2220	9000	0.4581
0.4829	12.4689	10000	0.4585

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu126
• Datasets 2.15.0
• Tokenizers 0.22.0