Edit model card

SpeechT5 TTS Tunisie

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

  • Loss: 0.9733

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: 1
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 20
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss
No log 0.4167 10 0.9837
No log 0.8333 20 0.9733

Framework versions

  • Transformers 4.42.0.dev0
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
6
Safetensors
Model size
144M params
Tensor type
F32
·

Finetuned from

Dataset used to train Arbi-Houssem/speecht5_ar_tn_1.0