---
library_name: pytorch
pipeline_tag: text-to-speech
license: apache-2.0
tags:
  - text-to-speech
  - audio-visual
  - lip-sync
  - face-conditioned
datasets:
  - lrs3
---

# SyncVoice

Official model weights trained on LRS3 dataset.

## Download

Download the model from
[kaidi555/SyncVoice](https://huggingface.co/kaidi555/SyncVoice):

```bash
hf download kaidi555/SyncVoice \
  --include "lrs_exp/*" \
  --local-dir checkpoints/syncvoice
```

The downloaded files are stored under:

```text
checkpoints/syncvoice/lrs_exp/
  checkpoint-55000.pt
  model.json
  tokens.txt
```