Code Monkey home page Code Monkey logo

dragonianvoice's Introduction

这里是 纳鲁赛繆 · 尤梅米 · 希娜卡纳

  • ❤️ 这里是 MoeSS/MoeVoiceStudio/LibDlVoiceCodec(LibSvc & LibTTS & LibSvs) 的开发者。其中LibDlVoiceCodec为多个 "Svc/TTS/Svs" 仓库的 C/Cpp/C# Onnx 推理程序仓库以及自研张量库和基于自研张量库的"Svc/TTS/Svs"推理仓库,目前后者正处于开发阶段,而前者已经可以稳定使用;而MoeSS/MoeVoiceStudio则为基于LibDlVoiceCodec开发的UI应用软件,可以比较方便的编辑LibDlVoiceCodec中的参数,以及实现专业歌声合成软件的部分功能。

  • ✉️ 我的邮箱:[email protected]

  • ⚡ 爱发电地址:纳鲁赛繆 · 尤梅米 · 希娜卡纳


My Skill Set

使用的语言

C C++ C# Java Kotlin Python

AI相关

pytorch TensorFlow

杂项

Premiere Pro Photoshop Illustrator PowerShell

GitHub状态

dragonianvoice's People

Contributors

narusemioshirakana avatar ricecakey06 avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

dragonianvoice's Issues

使用SoVits未找到模型

配置文件:

{
    "Folder" : "yue",
    "Name" : "yue",
    "Type" : "SoVits",
    "Symbol" : "_,.!?-~…AEINOQUabdefghijkmnoprstuvwyzʃʧʦ↓↑" ,
    "Cleaner" : "LowerCharacters",
    "Rate" : 22050,
    "Hop" : 0,
    "Hifigan": "hifigan",
    "SoVits3": true,
    "Hubert": "hubert",
    "Pndm" : 100,
    "MelBins" : 128,
    "Characters" : []
}

onnx文件:
image
界面提示:
image
请问是哪里我没有配置对吗?

RVC 的json模板有误

image
如图,这个字段为true时,根据代码,软件会将其识别成SoVits模型(在加载模型的时候提示找不到 xxx_SoVits.onnx)
image

[求助]关于集成libsvc并推理RVC-ONNX的一些问题

你好,我集成了libsvc至我的工程中,但在使用NativeAPI.h提供的接口进行推理时,遇到了一个棘手的问题,请问能否帮忙解答一下呢?

  1. 我之前使用pytorch和pth的时候,pcm的输入个数与输出相同。
    但是在使用LibSvcInferSlice接口时,发现pcm的输入和输出无法应对上。
    跟进函数内部后,发现推理前会先进行hubert推理,此时得到的hubert输出大小与我预期的onnx-rvc模型的pcm输入个数也有出入。
    即便我调整hubert输入大小,使得hubert输出大小刚好匹配上 ,使之能够得到finaOut时,却发现finaOut固定输出大小为20000,即便最后通过采样率计算resize之后,也与我实际的输入pcm个数不匹配。最终导致我输出的音频播放要么断断续续,要么声音不正确。
    请问,我该如何确保LibSvcInferSlice接口的输入大小与输出大小匹配呢?

在加载模型时闪退

正在使用CPU版本;Models使用的模型来源为他人分享;目录结构如下:
image

点击该“加载模型”按钮时,GUI将无征兆地闪退。且内存、显存并无明显波动。
image
CUDA版本为12.1,不符合GPU版本使用条件;但是考虑到使用的是CPU,理应没有影响。

log内容如下:

[Info] Removing Env & Release Memory
[Info] Complete!
[Info] Creating Env
[Info] Env Created

烦请大佬解疑释惑,提前感谢。

新版仍然无法使用

前置包已经下完按说明解压。

CPU版加载模型后出现错误提示。
666666

GPU版加载模型后没有错误提示,几秒后自己闪退。

Model not recognized, So Vits 4.0

MoeSS not recognizing model:

01

Hubert4.0 Model file directory:

02

Mods folder directory:

03

Model folder directory:

04

Config json file:

{
    "Folder" : "hapiv2",
    "Name" : "hapiv2",
    "Type" : "SoVits",
    "Rate" : 44100,
    "Hop" : 512,
    "Cleaner" : "",
    "Hubert": "hubert4.0",
    "SoVits4": true,
    "Characters" : [""]
}
//Hop:HopLength of the model, if you don't know what it is you are advised to look up the information on the internet. This must be filled in the configuration file of the SoVits model.(The value must be the one you set during training and can be seen in the configuration file you used to train the model)
//Cleaner:The name of the plugin,can be left blank, but if it is filled in, the corresponding CleanerDll must be placed in the Cleaner folder, if the Dll does not exist or if there is an internal error in the Dll, it will report an error when loading the model
//Hubert:Hubert model name, required and must be placed in the "Hubert" folder for Hubert models downloaded from the sub-model repository
//Characters:For multi-speaker model this must be filled in as a list of your speakers' names, for single-speaker model it can be left out

Am I missing something here? It doesnt seem to be recognizing the model.

Failed to Inference

弹窗

屏幕截图 2023-07-16 182517.png

推理切片或用旧版方式推理均有该弹窗,

RVC的json文件:

{
    "Folder" : "c_RVC",
    "Name" : "c",
    "Type" : "RVC",
    "Rate" : 40000,
    "Hop" : 400,
    "Cleaner" : "",
    "Hubert": "hubert4.0",
    "Diffusion": false,
    "CharaMix": true,
    "Volume": false,
    "HiddenSize": 256,
    "Characters" : []
}

设置

设置.png

bert-vits2 中文特化版支持吗?

请问如果支持的话该如何使用呢?
我看了中文特化用到的二郎神bert,好像不太好导出onnx?我导出时产生了300多个文件而不是一个单独的onnx文件。

不考虑支持一下4.0-Vec768-Layer12吗 似乎只支持到V2

Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]

Describe the solution you'd like
A clear and concise description of what you want to happen.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.

exe运行问题

1
当我这样放置json文件时,点击exe无法出现ui界面,
2
当我把json文件放入模型文件夹中,可以运行exe,但是无法载入模型。
{6D868C1D-FA30-48b5-9C32-D70601C1C099}
我的模型是通过ddsp 5.0训练的,这是我的json文件内容,
请问问题出在哪里呢?

关于新版的一些问题

  1. 新版tts支持生成特殊语气的语音么,比如特别愤怒,特别开心,等不同的语气和情绪
  2. 新版tts支持中文生成不,是只要模型支持就可以是吧

请问`转换ONNX的程序`在哪?

模型需要转换为ONNX模型,转换ONNX的程序我已经pull到每个项目的源仓库了,PTH不能直接用!!!!!!!!!!!!!

请问ReadMe中提到的pth转换ONNX的程序在哪里?

onnx 模型

您好,我想用onnx推理,但是项目里用python推得脚本我不确定输入输出是什么。换个说法就是我该用哪个模型转的onnx来作为model.onnx呢?我用sovits-4.1训好了一个,但是转onnx后不可以直接用该项目的python脚本推理,非常感谢!

无法加载任何模型

当前的Release版本,下载后同时去下载了三个模型,并解压到了Mods文件夹
image

打开exe不弹窗报错,但是模型列表为空,仅显示 未选中
image
// ====
经查看.mod文件,发现与readme不符,
image
将 Mdid 修改为 Folder 后重试,依旧无法加载,请问怎么才能运行呢

error when use cuda

环境创建失败!
OrtException:
[ln "EnwManager.cpp" Line 68]
D: a work)1s onnxruntime core session(provider bridge ort.cc:1209
onnxruntimeProviderLibrary..Get [ONNXRuntimeError] : 1 : FAIL :
loadLibrary failed with error 126 " when trying to load
CUsers Administrator Downloads MoeVoicestudio.-1TS MoeVoicest
udio - T1s onnxruntime providers cuda.dll
微信截图_20240306234003

没有调用GPU推理

首先感谢作者,总体使用很方便,效果不错

但是我在运行时默认调用的是CPU,而不是GPU,请问这可能是什么原因导致的?
我使用的是SoVITS的模型,之前用Python推理时没有这个问题

距离构建成功还差一步

问下作者用的VS那个版本构建的。
我用VS2019编译过了,链接出错:

Linked 错误 C1007 无法识别的标志“-Zc:nrvo”(在“p2”中)

可能是lib链接的问题。

SoVits Error: cannot open audio file

When I try to use SoVits, it worked well when I converted my first .wav file. However, after I saved the first converted file, it reports the error that it cannot open the audio file when I try to convert the second .wav file.

my system is Windows 11, and I used the cuda version of Moess.

I'm sure the file address is right, when I close the software and restart it. I can convert my sound file. The problem still happens when I try to convert the second file.

By the way, is this software support f0 inference of SoVits? How to set f0? In Sovits command it is used as an option parameter "-a".

Thanks for your attention!

遇到的问题

我使用RVC的cmd推导onnx在输入完所有信息(模型到音频路径)之后,在运行时,会直接闪退,并且没有音频文件输出出来。
屏幕截图 2024-02-26 212601
然后我在使用moevs的ui界面时,怎么导入diffusion模型(需要傻瓜式教程,我用rvc模型的)
image
我不导入diffusion模型,在加载模型会闪退,在导入音频无法导入(能对音频切片页面,但是没有导入进来)
log:
[Info] Removing Env & Release Memory
[Info] Complete!
[Info] Creating Env
[Info] Env Created

Core分支编译失败

MoeVoiceStudioCore分支的main.cpp还是对QT有依赖,例如#include <QApplication>,请问必须依赖QT是设计如此吗?

需要一个傻瓜式的教程

我是新手,不知道怎么才能下载到歌手声源,应该怎么做才能让它发声,这一切我都不懂,说实话真的需要一个详细的教程

建议这个工程不要有UI部分的功能

看到大佬最近提交了很多QT相关代码。
建议能不能像Moegoe那样,命令行和UI的工程是分离调用的。这个工程只用做核心推理就行了,我可以帮忙随便套个皮写GUI调用命令行推理。

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.