nonoJDWAOIDAWKDA/test2_finetune_StyleTTS2

StyleTTS2 Fine-tuned Model

This model is a fine-tuned version of StyleTTS2.

Model Details

• Base Model: StyleTTS2-LibriTTS
• Architecture: StyleTTS2
• Task: Text-to-Speech
• Last Checkpoint: epoch_2nd_00004.pth

Training Details

• Total Epochs: 5
• Completed Epochs: 4
• Total Iterations: 411
• Batch Size: 2
• Max Length: 120
• Learning Rate: 0.0001
• Final Validation Loss: 0.430844

Loss Parameters

• Diff Epoch: 10
• Joint Epoch: 110
• Lambda Parameters:
  • Mel: 5.0
  • F0: 1.0
  • Duration: 1.0
  • Style: 1.0

Model Components

• bert
• bert_encoder
• predictor
• decoder
• text_encoder
• predictor_encoder
• style_encoder
• diffusion
• text_aligner
• pitch_extractor
• mpd
• msd
• wd

Training Metrics

Training metrics visualization is available in training_metrics.png