maguette9315/speecht5_tts_voxpopuli_nl

SpeechT5 TTS MT V1

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

• Loss: 0.7553

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: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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: 10
• training_steps: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
No log	6.6667	10	1.0697
No log	13.3333	20	0.9269
1.1237	20.0	30	0.8719
1.1237	26.6667	40	0.8235
0.8225	33.3333	50	0.8047
0.8225	40.0	60	0.7953
0.8225	46.6667	70	0.7818
0.7747	53.3333	80	0.7663
0.7747	60.0	90	0.7559
0.7322	66.6667	100	0.7553

Framework versions

• Transformers 4.46.3
• Pytorch 2.6.0+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3