---
language: 
  - zh
  - en
tags:
  - speech-synthesis
  - speech-to-speech
  - voice-conversion
  - pytorch
  - audio
  - chinese-tts
  - multi-speaker
  - convolution
  - encoder-decoder
license: apache-2.0
datasets:
- vctk
library_name: pytorch
---

# Convbased

Github: [https://github.com/Convbased/Convbased-Studio](https://github.com/Convbased/Convbased-Studio)

This project focuses on training high-quality pre-trained models.


| Feature Extraction | Vocoder | Sample Rate 40k | Sample Rate 48k |
|-----------|--------|-----|-----|
| contentvec | hifigannsf | ❌ | ✅ |
| contentvec | sifigan | ❌ | ✅ |
| contentvec | bigvgan | ✅ | ❌ |
| spin | hifigannsf | ❌ | ✅ |
| spin | sifigan | ❌ | ✅ |
| spin-v2 | bigvgan | 🧱 | ❌ |
| chinese-hubert-base | hifigannsf | ❌ | ✅ |


*Training code from [Applio](https://github.com/IAHispano/Applio).*

*Dedicated to advancing Chinese speech synthesis technology. These base models have been used for fine-tuning most models at [Convbased Studio](https://weights.chat/).*