Parler TTS Roleplay Model V2

Optimized fine-tuned version of Parler-TTS Mini v1 for voice generation. Based on the original model from https://huggingface.co/parler-tts/parler-tts-mini-v1

Training improvements:

  • Increased batch size to 64
  • Cosine learning rate with restarts
  • Weight decay 0.01
  • Group by length enabled
Downloads last month
53
Safetensors
Model size
878M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and HF Inference API was unable to determine this model's library.