speecht5_clartts_finetuned_TTS-PLS

This model is a fine-tuned version of MBZUAI/speecht5_tts_clartts_ar on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5869

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: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss
0.642	2.8169	250	0.5850
0.588	5.6338	500	0.5583
0.5758	8.4507	750	0.5506
0.56	11.2676	1000	0.5511
0.5522	14.0845	1250	0.5469
0.5398	16.9014	1500	0.5490
0.5341	19.7183	1750	0.5551
0.5294	22.5352	2000	0.5555
0.5235	25.3521	2250	0.5591
0.5197	28.1690	2500	0.5637
0.5154	30.9859	2750	0.5738
0.5109	33.8028	3000	0.5694
0.5094	36.6197	3250	0.5742
0.507	39.4366	3500	0.5740
0.5063	42.2535	3750	0.5791
0.5018	45.0704	4000	0.5811
0.4988	47.8873	4250	0.5844
0.4989	50.7042	4500	0.5835
0.4989	53.5211	4750	0.5850
0.5001	56.3380	5000	0.5869

Framework versions

• Transformers 4.43.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1