speecht5_finetuned_voxpopuli_fi_lim_speakrs

This model is a fine-tuned version of microsoft/speecht5_tts on the voxpopuli/fi dataset. It achieves the following results on the evaluation set:

• Loss: 0.4429

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: 10
• eval_batch_size: 2
• seed: 42
• optimizer: Use 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: 250
• training_steps: 8000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.5587	1.25	500	0.4937
0.5176	2.5	1000	0.4703
0.4983	3.75	1500	0.4594
0.5016	5.0	2000	0.4584
0.4797	6.25	2500	0.4539
0.4845	7.5	3000	0.4512
0.4882	8.75	3500	0.4489
0.4721	10.0	4000	0.4469
0.4829	11.25	4500	0.4456
0.4974	12.5	5000	0.4457
0.4776	13.75	5500	0.4442
0.4866	15.0	6000	0.4441
0.4752	16.25	6500	0.4430
0.4765	17.5	7000	0.4430
0.4668	18.75	7500	0.4431
0.4823	20.0	8000	0.4429

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3