speecht5-chabacano-tts

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: 2.1987

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: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• 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
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
3.0013	3.5714	500	2.7360
2.5914	7.1429	1000	2.4185
2.4372	10.7143	1500	2.2980
2.371	14.2857	2000	2.2287
2.3281	17.8571	2500	2.1987

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1