speecht5_tts_voxpopuli_nl

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

• Loss: 0.4761

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• 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
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.6506	1.0	239	0.5907
0.5796	2.0	478	0.5158
0.5326	3.0	717	0.5087
0.5343	4.0	956	0.5023
0.5319	5.0	1195	0.4958
0.5193	6.0	1434	0.4878
0.5253	7.0	1673	0.4854
0.4992	8.0	1912	0.4851
0.4983	9.0	2151	0.4807
0.5039	10.0	2390	0.4781
0.5153	11.0	2629	0.4777
0.4882	12.0	2868	0.4773
0.4917	13.0	3107	0.4819
0.4942	14.0	3346	0.4758
0.4817	15.0	3585	0.4771
0.4832	16.0	3824	0.4763
0.4721	17.0	4063	0.4776
0.4975	18.0	4302	0.4789
0.493	19.0	4541	0.4770
0.4955	20.0	4780	0.4761

Framework versions

• Transformers 4.49.0.dev0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0