speecht5_finetuned_haitian_creole_tts

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3680

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
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• 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: 100
• training_steps: 17504

Training results

Training Loss	Epoch	Step	Validation Loss
2.627	15.1550	500	0.3270
2.5152	30.3101	1000	0.3280
2.443	45.4651	1500	0.3289
2.4247	60.6202	2000	0.3360
2.4638	75.7752	2500	0.3318
2.4189	90.9302	3000	0.3377
2.311	106.0620	3500	0.3432
2.2877	121.2171	4000	0.3439
2.2857	136.3721	4500	0.3434
2.2751	151.5271	5000	0.3427
2.2665	166.6822	5500	0.3456
2.2909	181.8372	6000	0.3531
2.2922	196.9922	6500	0.3479
2.2163	212.1240	7000	0.3468
2.1916	227.2791	7500	0.3465
2.1993	242.4341	8000	0.3517
2.1799	257.5891	8500	0.3534
2.1627	272.7442	9000	0.3481
2.2402	287.8992	9500	0.3542
2.1602	303.0310	10000	0.3541
2.1541	318.1860	10500	0.3506
2.1236	333.3411	11000	0.3619
2.1321	348.4961	11500	0.3519
2.1113	363.6512	12000	0.3588
2.1757	378.8062	12500	0.3512
2.1742	393.9612	13000	0.3578
2.0891	409.0930	13500	0.3593
2.0869	424.2481	14000	0.3601
2.0978	439.4031	14500	0.3589
2.0819	454.5581	15000	0.3637
2.0664	469.7132	15500	0.3589
2.1653	484.8682	16000	0.3586
2.0797	500.0	16500	0.3606
2.0506	515.1550	17000	0.3661
2.0713	530.3101	17500	0.3680

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0