innnky/MagVITS

VITS with phoneme-level prosody modeling based on MaskGIT

stars

commits

Python

primary language

Aug 31, 2024

updated

README

MagVITS

VITS with phoneme-level prosody modeling based on MaskGIT （WIP）

feature: inference speed ~= bert-vits2 & prosody > bert-vits2 (maybe)

目前代码正在重构，可能还跑不通，目前不建议跑中文预训练模型不久后会上传（数据：原神中文+aishell 共200h多一些）

structure

Acknowledgements

Thanks to the support of the GPUs by leng-yue fishaudio
VITS
MaskGIT
AuxiliaryASR and styletts2
MegaTTS
descript-audio-codec
visinger

Contributors

innnky

11 commits

Similar repos

innnky/ar-vits

text to speech using autoregressive transformer and VITS

247

same org

innnky/emotional-vits

无需情感标注的情感可控语音合成模型，基于VITS

1,391

same org

xincanfeng/vitsgpt-vits

the code for vits in the vitsGPT project

liuhuang/Megatts2_HierSpeechpp

Model

amphion/Vevo1.5

Model

OpenMOSS-Team/MOSS-VoiceGenerator

Model

LSimon95/megatts2

Unoffical implementation of Megatts2

285

Blizaine/maestro-seedvc

seed-vc voice conversion adapted for Maestro (GPL-3.0). Cloned automatically at install time by…

Languages

Python

100.0%

README

MagVITS

VITS with phoneme-level prosody modeling based on MaskGIT （WIP）

feature: inference speed ~= bert-vits2 & prosody > bert-vits2 (maybe)

目前代码正在重构，可能还跑不通，目前不建议跑中文预训练模型不久后会上传（数据：原神中文+aishell 共200h多一些）

structure

Acknowledgements

Thanks to the support of the GPUs by leng-yue fishaudio
VITS
MaskGIT
AuxiliaryASR and styletts2
MegaTTS
descript-audio-codec
visinger