speecht5_finetuned_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: 0.3776

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 125
• training_steps: 3000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.4834	2.3529	250	0.4367
0.4405	4.7059	500	0.4013
0.4323	7.0588	750	0.3927
0.4244	9.4118	1000	0.3877
0.4186	11.7647	1250	0.3844
0.4176	14.1176	1500	0.3831
0.4137	16.4706	1750	0.3807
0.4098	18.8235	2000	0.3804
0.413	21.1765	2250	0.3768
0.4094	23.5294	2500	0.3781
0.409	25.8824	2750	0.3789
0.4028	28.2353	3000	0.3776

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1