speecht5_finetuned_voxpopuli_en - Kaiky

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

• Loss: 0.4526

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_FUSED 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: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.0789	3.2051	500	0.4801
1.0035	6.4103	1000	0.4636
0.9923	9.6154	1500	0.4590
0.9552	12.8205	2000	0.4558
0.9722	16.0256	2500	0.4553
0.9492	19.2308	3000	0.4537
0.9686	22.4359	3500	0.4531
0.9686	25.6410	4000	0.4526

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.8.5
• Tokenizers 0.22.2