walston's picture
Add model configuration and model card
7c7712c verified
|
Raw
History Blame Contribute Delete
1.01 kB
metadata
library_name: pytorch
license: mit
pipeline_tag: text-to-speech
tags:
  - accent-tts
  - mandarin
  - joycent
  - grad-tts
  - whisaid

Joycent trained with WhisAID accent embeddings

This Joycent Mandarin accent TTS acoustic model was trained using accent embeddings extracted by walston/whisaid-medium-grl. The released checkpoint is epoch 100.

Download

from huggingface_hub import hf_hub_download

checkpoint_path = hf_hub_download(
    repo_id="walston/joycent-medium-grl-add",
    filename="grad_100.pt",
)

Pass the path to joycent/inference_joycent.py with --acoustic-checkpoint. Full synthesis also requires the Joycent vocoder.

Checkpoint

  • Epoch: 100
  • Acoustic model: Joycent / Grad-TTS
  • Accent embedding dimension: 256