From c33328091bb4c27c5f2917da403249ecdc71004b Mon Sep 17 00:00:00 2001
From: RVC-Boss <129054828+RVC-Boss@users.noreply.github.com>
Date: Sun, 19 Jul 2026 22:23:01 +0800
Subject: [PATCH] Update Ubuntu setup and multilingual documentation

---
 README.md                   | 237 ++++++++++++++++++---------------
 docs/en/README.en.md        | 236 ++++++++++++++++++---------------
 docs/fr/README.fr.md        | 230 +++++++++++++++++++-------------
 docs/jp/README.ja.md        | 252 ++++++++++++++++++------------------
 docs/kr/README.ko.han.md    | 189 ++++++++++++++++++++-------
 docs/kr/README.ko.md        | 243 +++++++++++++++++-----------------
 docs/pt/README.pt.md        | 241 ++++++++++++++++++++--------------
 docs/tr/README.tr.md        | 219 ++++++++++++++++++++-----------
 requirments_cpu_py312.txt   |  18 +--
 requirments_cu118_py312.txt |   7 +-
 requirments_cu128_py312.txt |   7 +-
 webui.py                    |  15 ++-
 12 files changed, 1111 insertions(+), 783 deletions(-)

diff --git a/README.md b/README.md
index c86746f..108f897 100644
--- a/README.md
+++ b/README.md
@@ -1,20 +1,18 @@
 <div align="center">
 
 <h1>Retrieval-based-Voice-Conversion-WebUI</h1>
-一个基于VITS的简单易用的变声框架<br><br>
+简单易用的 语音音色转换/变声器 框架<br><br>
 
 [![madewithlove](https://img.shields.io/badge/made_with-%E2%9D%A4-red?style=for-the-badge&labelColor=orange
 )](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)
 
 <img src="https://counter.seku.su/cmoe?name=rvc&theme=r34" /><br>
 
-[![Open In Colab](https://img.shields.io/badge/Colab-F9AB00?style=for-the-badge&logo=googlecolab&color=525252)](https://colab.research.google.com/github/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/Retrieval_based_Voice_Conversion_WebUI.ipynb)
 [![Licence](https://img.shields.io/badge/LICENSE-MIT-green.svg?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
-[![Huggingface](https://img.shields.io/badge/🤗%20-Spaces-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
+[![Huggingface](https://img.shields.io/badge/🤗%20-Models-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
 
-[![Discord](https://img.shields.io/badge/RVC%20Developers-Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/HcsmBBGyVk)
 
-[**更新日志**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/docs/Changelog_CN.md) | [**常见问题解答**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98%E8%A7%A3%E7%AD%94) | [**AutoDL·5毛钱训练AI歌手**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/Autodl%E8%AE%AD%E7%BB%83RVC%C2%B7AI%E6%AD%8C%E6%89%8B%E6%95%99%E7%A8%8B) | [**对照实验记录**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/Autodl%E8%AE%AD%E7%BB%83RVC%C2%B7AI%E6%AD%8C%E6%89%8B%E6%95%99%E7%A8%8B](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%AF%B9%E7%85%A7%E5%AE%9E%E9%AA%8C%C2%B7%E5%AE%9E%E9%AA%8C%E8%AE%B0%E5%BD%95)) | [**在线演示**](https://modelscope.cn/studios/FlowerCry/RVCv2demo)
+[**更新日志**](./docs/cn/Changelog_CN.md) | [**常见问题解答**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98%E8%A7%A3%E7%AD%94) | [**AutoDL·5毛钱训练AI歌手**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/Autodl%E8%AE%AD%E7%BB%83RVC%C2%B7AI%E6%AD%8C%E6%89%8B%E6%95%99%E7%A8%8B) | [**对照实验记录**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%AF%B9%E7%85%A7%E5%AE%9E%E9%AA%8C%C2%B7%E5%AE%9E%E9%AA%8C%E8%AE%B0%E5%BD%95) | [**在线演示**](https://modelscope.cn/studios/FlowerCry/RVCv2demo)
 
 [**English**](./docs/en/README.en.md) | [**中文简体**](./README.md) | [**日本語**](./docs/jp/README.ja.md) | [**한국어**](./docs/kr/README.ko.md) ([**韓國語**](./docs/kr/README.ko.han.md)) | [**Français**](./docs/fr/README.fr.md) | [**Türkçe**](./docs/tr/README.tr.md) | [**Português**](./docs/pt/README.pt.md)
 
@@ -52,140 +50,173 @@
 + 简单易用的网页界面
 + 可调用UVR5模型来快速分离人声和伴奏
 + 使用最先进的[人声音高提取算法InterSpeech2023-RMVPE](#参考项目)根绝哑音问题，速度快、资源占用小
-+ A卡I卡加速支持
++ A卡/I卡使用 CPU 依赖方案；Windows 可使用 DirectML，Linux 使用 CPU
 
 点此查看我们的[演示视频](https://www.bilibili.com/video/BV1pm4y1z7Gm/) !
 
 ## 环境配置
-以下指令需在 Python 版本大于3.8的环境中执行。  
 
-### Windows/Linux/MacOS等平台通用方法
-下列方法任选其一。
-#### 1. 通过 pip 安装依赖
-1. 安装Pytorch及其核心依赖，若已安装则跳过。参考自: https://pytorch.org/get-started/locally/
+本分支面向 **Python 3.12 x64**，请先进入仓库根目录。Ubuntu 推荐使用 Ubuntu 24.04 x86_64。
+
+### Ubuntu 24.04
+
 ```bash
-pip install torch torchvision torchaudio
-```
-2. 如果是 win 系统 + Nvidia Ampere 架构(RTX30xx)，根据 #21 的经验，需要指定 pytorch 对应的 cuda 版本
-```bash
-pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
-```
-3. 根据自己的显卡安装对应依赖
-- N卡
-```bash
-pip install -r requirements.txt
-```
-- A卡/I卡
-```bash
-pip install -r requirements-dml.txt
-```
-- A卡ROCM(Linux)
-```bash
-pip install -r requirements-amd.txt
-```
-#### 2. 通过 poetry 来安装依赖
-安装 Poetry 依赖管理工具，若已安装则跳过。参考自: https://python-poetry.org/docs/#installation
-```bash
-curl -sSL https://install.python-poetry.org | python3 -
+sudo apt update
+sudo apt install -y python3.12 python3.12-venv python3.12-dev ffmpeg unzip libsndfile1 libportaudio2
+
+python3.12 -m venv .venv
+source .venv/bin/activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-通过 Poetry 安装依赖时，python 建议使用 3.7-3.10 版本，其余版本在安装 llvmlite==0.39.0 时会出现冲突
-```bash
-poetry init -n
-poetry env use "path to your python.exe"
-poetry run pip install -r requirments.txt
+### Windows
+
+安装 Python 3.12 x64 后创建虚拟环境：
+
+```powershell
+py -3.12 -m venv .venv
+.venv\Scripts\activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-### MacOS
-可以通过 `run.sh` 来安装依赖
+### 按硬件选择依赖
+
+| 硬件 | 安装方式 |
+| --- | --- |
+| CPU、AMD、Intel | 使用 `requirments_cpu_py312.txt`；Windows 可使用 DirectML，Linux 使用 CPU |
+| NVIDIA RTX 50 系 | 先安装 CUDA 12.8 版 Torch，再安装 `requirments_cu128_py312.txt` |
+| NVIDIA RTX 50 系以前 | 先安装 CUDA 11.8 版 Torch，再安装 `requirments_cu118_py312.txt` |
+
+#### CPU、AMD、Intel
+
 ```bash
-sh ./run.sh
+python -m pip install -r requirments_cpu_py312.txt
 ```
 
-## 其他预模型准备
-RVC需要其他一些预模型来推理和训练。
+#### NVIDIA RTX 50 系：两阶段安装
 
-你可以从我们的[Hugging Face space](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)下载到这些模型。
-
-### 1. 下载 assets
-以下是一份清单，包括了所有RVC所需的预模型和其他文件的名称。你可以在`tools`文件夹找到下载它们的脚本。
-
-- ./assets/hubert_base
-
-- ./assets/pretrained 
-
-- ./assets/uvr5_weights
-
-想使用v2版本模型的话，需要额外下载
-
-- ./assets/pretrained_v2
-
-### 2. 安装 ffmpeg
-若ffmpeg和ffprobe已安装则跳过。
-
-#### Ubuntu/Debian 用户
 ```bash
-sudo apt install ffmpeg
+python -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 \
+  --index-url https://download.pytorch.org/whl/cu128 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu128_py312.txt
 ```
-#### MacOS 用户
+
+#### NVIDIA RTX 50 系以前：两阶段安装
+
 ```bash
-brew install ffmpeg
+python -m pip install torch==2.7.1+cu118 torchaudio==2.7.1+cu118 \
+  --index-url https://download.pytorch.org/whl/cu118 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu118_py312.txt
 ```
-#### Windows 用户
-下载后放置在根目录。
-- 下载[ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffmpeg.exe)
 
-- 下载[ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffprobe.exe)
+检查 Torch 与 CUDA 状态：
 
-### 3. 下载 rmvpe 人声音高提取算法所需文件
+```bash
+python -c "import torch; print('torch:', torch.__version__); print('cuda:', torch.version.cuda); print('cuda available:', torch.cuda.is_available())"
+```
 
-如果你想使用最新的RMVPE人声音高提取算法，则你需要下载音高提取模型参数并放置于RVC根目录。
+项目代码还会检查 NVIDIA GPU 的显存和计算能力。显存约低于 4 GiB 或计算能力低于 SM 5.3 时会使用 CPU。
 
-- 下载[rmvpe.pt](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.pt)
+### 修改下载源
 
-#### 下载 rmvpe 的 dml 环境(可选, A卡/I卡用户)
+三个 `requirments_*.txt` 顶部已经包含下载源。中国大陆用户可保留默认镜像；需要使用官方源时，只替换 `--index-url` 和 `--extra-index-url`，保留包版本、CUDA 后缀和两阶段顺序。
 
-- 下载[rmvpe.onnx](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.onnx)
+| Default mirror | Official source |
+| --- | --- |
+| `https://mirrors.pku.edu.cn/pypi/simple` | `https://pypi.org/simple` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cpu` | `https://download.pytorch.org/whl/cpu` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu118` | `https://download.pytorch.org/whl/cu118` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu128` | `https://download.pytorch.org/whl/cu128` |
 
-### 4. AMD显卡Rocm(可选, 仅Linux)
+## 模型与运行目录
 
-如果你想基于AMD的Rocm技术在Linux系统上运行RVC，请先在[这里](https://rocm.docs.amd.com/en/latest/deploy/linux/os-native/install.html)安装所需的驱动。
+WebUI 会自动创建运行目录。模型请从 [Hugging Face 模型仓库](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main) 下载，并保持以下路径：
 
-若你使用的是Arch Linux，可以使用pacman来安装所需驱动：
-````
-pacman -S rocm-hip-sdk rocm-opencl-sdk
-````
-对于某些型号的显卡，你可能需要额外配置如下的环境变量（如：RX6700XT）：
-````
-export ROCM_PATH=/opt/rocm
-export HSA_OVERRIDE_GFX_VERSION=10.3.0
-````
-同时确保你的当前用户处于`render`与`video`用户组内：
-````
-sudo usermod -aG render $USERNAME
-sudo usermod -aG video $USERNAME
-````
+```text
+assets/
+├── hubert_base/
+│   ├── config.json
+│   ├── preprocessor_config.json
+│   └── pytorch_model.bin
+├── rmvpe/rmvpe.pt
+├── pretrained/
+├── pretrained_v2/
+├── uvr5_weights/
+├── weights/        # user RVC .pth models
+└── indices/        # user .index files
+logs/
+└── mute/           # training silence samples
+
+# Exact paths used by the code
+assets/hubert_base/config.json
+assets/hubert_base/preprocessor_config.json
+assets/hubert_base/pytorch_model.bin
+assets/rmvpe/rmvpe.pt
+assets/pretrained/*.pth
+assets/pretrained_v2/*.pth
+assets/uvr5_weights/*
+assets/weights/*.pth
+assets/indices/*.index
+logs/mute/*
+```
+
+### 下载模型
+
+```bash
+python -m pip install --upgrade huggingface_hub
+
+# Required for inference and feature extraction
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "hubert_base/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI rmvpe.pt --revision main \
+  --local-dir assets/rmvpe
+
+# Required for v1/v2 training
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "pretrained/*" "pretrained_v2/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI mute.zip --revision main \
+  --local-dir .model-downloads
+python -m zipfile -e .model-downloads/mute.zip logs
+
+# Required only for UVR5 vocal separation
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "uvr5_weights/*" --local-dir assets
+```
+
+仅 Windows AMD/Intel DirectML 环境还需要：
+
+```bash
+hf download lj1995/VoiceConversionWebUI rmvpe.onnx --revision main \
+  --local-dir assets/rmvpe
+```
+
+`hubert_base.pt` 是旧格式；当前代码使用 `assets/hubert_base/` 下的 Transformers 三文件模型。FCPE 模型由 `torchfcpe` 包提供。
+
+### FFmpeg
+
+Ubuntu 已在前面的系统依赖命令中安装 FFmpeg。Windows 用户可把下面两个文件放到项目根目录：
+
+- [ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffmpeg.exe?download=true)
+- [ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffprobe.exe?download=true)
 
 ## 开始使用
-### 直接启动
-使用以下指令来启动 WebUI
+
+启动 WebUI：
+
 ```bash
 python webui.py
 ```
 
-若先前使用 Poetry 安装依赖，则可以通过以下方式启动WebUI
+无桌面的 Ubuntu 服务器：
+
 ```bash
-poetry run python webui.py
+python webui.py --noautoopen
 ```
 
-### 使用整合包
-下载并解压`RVC-beta.7z`
-#### Windows 用户
-双击`go-webui.bat`
-#### MacOS 用户
-```bash
-sh ./run.sh
-```
+默认服务监听端口为 `7865`。用户自己的 `.pth` 模型放入 `assets/weights/`，`.index` 文件放入 `assets/indices/`。
+
 ## 参考项目
 + [ContentVec](https://github.com/auspicious3000/contentvec/)
 + [VITS](https://github.com/jaywalnut310/vits)
diff --git a/docs/en/README.en.md b/docs/en/README.en.md
index 630ab1e..788bbeb 100644
--- a/docs/en/README.en.md
+++ b/docs/en/README.en.md
@@ -1,20 +1,18 @@
 <div align="center">
 
 <h1>Retrieval-based-Voice-Conversion-WebUI</h1>
-An easy-to-use Voice Conversion framework based on VITS.<br><br>
+A simple, easy-to-use voice timbre conversion / voice changer framework.<br><br>
 
 [![madewithlove](https://img.shields.io/badge/made_with-%E2%9D%A4-red?style=for-the-badge&labelColor=orange
 )](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)
 
 <img src="https://counter.seku.su/cmoe?name=rvc&theme=r34" /><br>
-  
-[![Open In Colab](https://img.shields.io/badge/Colab-F9AB00?style=for-the-badge&logo=googlecolab&color=525252)](https://colab.research.google.com/github/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/Retrieval_based_Voice_Conversion_WebUI.ipynb)
+
 [![Licence](https://img.shields.io/github/license/RVC-Project/Retrieval-based-Voice-Conversion-WebUI?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
-[![Huggingface](https://img.shields.io/badge/🤗%20-Spaces-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
+[![Huggingface](https://img.shields.io/badge/🤗%20-Models-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
 
-[![Discord](https://img.shields.io/badge/RVC%20Developers-Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/HcsmBBGyVk)
 
-[**Changelog**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/docs/Changelog_EN.md) | [**FAQ (Frequently Asked Questions)**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/FAQ-(Frequently-Asked-Questions)) 
+[**Changelog**](./Changelog_EN.md) | [**FAQ (Frequently Asked Questions)**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/FAQ-(Frequently-Asked-Questions))
 
 [**English**](../en/README.en.md) | [**中文简体**](../../README.md) | [**日本語**](../jp/README.ja.md) | [**한국어**](../kr/README.ko.md) ([**韓國語**](../kr/README.ko.han.md)) | [**Français**](../fr/README.fr.md) | [**Türkçe**](../tr/README.tr.md) | [**Português**](../pt/README.pt.md)
 
@@ -55,135 +53,169 @@ An easy-to-use Voice Conversion framework based on VITS.<br><br>
 + Easy-to-use WebUI;
 + UVR5 model to quickly separate vocals and instruments;
 + High-pitch Voice Extraction Algorithm [InterSpeech2023-RMVPE](#Credits) to prevent a muted sound problem. Provides the best results (significantly) and is faster with lower resource consumption than Crepe_full;
-+ AMD/Intel graphics cards acceleration supported;
++ AMD/Intel systems use the CPU dependency set; Windows may use DirectML and Linux uses CPU;
 
-## Preparing the environment
-The following commands need to be executed with Python 3.8 or higher.
+## Environment setup
+
+This branch targets **Python 3.12 x64**. Run every command from the repository root. Ubuntu 24.04 x86_64 is recommended.
+
+### Ubuntu 24.04
 
-(Windows/Linux)
-First install the main dependencies through pip:
 ```bash
-# Install PyTorch-related core dependencies, skip if installed
-# Reference: https://pytorch.org/get-started/locally/
-pip install torch torchvision torchaudio
+sudo apt update
+sudo apt install -y python3.12 python3.12-venv python3.12-dev ffmpeg unzip libsndfile1 libportaudio2
 
-#For Windows + Nvidia Ampere Architecture(RTX30xx), you need to specify the cuda version corresponding to pytorch according to the experience of https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/21
-#pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
-
-#For Linux + AMD Cards, you need to use the following pytorch versions:
-#pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm5.4.2
+python3.12 -m venv .venv
+source .venv/bin/activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-Then can use poetry to install the other dependencies:
-```bash
-# Install the Poetry dependency management tool, skip if installed
-# Reference: https://python-poetry.org/docs/#installation
-curl -sSL https://install.python-poetry.org | python3 -
+### Windows
 
-# Install the project dependencies
-poetry install
+Install Python 3.12 x64, then create a virtual environment:
+
+```powershell
+py -3.12 -m venv .venv
+.venv\Scripts\activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-You can also use pip to install them:
+### Choose dependencies by hardware
+
+| Hardware | Installation |
+| --- | --- |
+| CPU, AMD, Intel | Use `requirments_cpu_py312.txt`; Windows may use DirectML, while Linux uses CPU |
+| NVIDIA RTX 50 series | Install the CUDA 12.8 Torch pair first, then `requirments_cu128_py312.txt` |
+| NVIDIA GPUs before the RTX 50 series | Install the CUDA 11.8 Torch pair first, then `requirments_cu118_py312.txt` |
+
+#### CPU, AMD, Intel
+
 ```bash
-
-for Nvidia graphics cards
-  pip install -r requirements.txt
-
-for AMD/Intel graphics cards on Windows (DirectML)：
-  pip install -r requirements-dml.txt
-
-for AMD graphics cards on Linux (ROCm):
-  pip install -r requirements-amd.txt
+python -m pip install -r requirments_cpu_py312.txt
 ```
 
-------
-Mac users can install dependencies via `run.sh`:
+#### NVIDIA RTX 50 series: two stages
+
 ```bash
-sh ./run.sh
+python -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 \
+  --index-url https://download.pytorch.org/whl/cu128 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu128_py312.txt
 ```
 
-## Preparation of other Pre-models
-RVC requires other pre-models to infer and train.
+#### NVIDIA GPUs before the RTX 50 series: two stages
 
-Download them from our [Huggingface space](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/).
-
-Here's a list of Pre-models and other files that RVC needs:
 ```bash
-./assets/hubert_base
-
-./assets/pretrained 
-
-./assets/uvr5_weights
-
-Additional downloads are required if you want to test the v2 version of the model.
-
-./assets/pretrained_v2
-
-If you want to test the v2 version model (the v2 version model has changed the input from the 256 dimensional feature of 9-layer Hubert+final_proj to the 768 dimensional feature of 12-layer Hubert, and has added 3 period discriminators), you will need to download additional features
-
-./assets/pretrained_v2
-
-If you want to use the latest SOTA RMVPE vocal pitch extraction algorithm, you need to download the RMVPE weights and place them in the RVC root directory
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.pt
-
-    For AMD/Intel graphics cards users you need download:
-
-    https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.onnx
-
+python -m pip install torch==2.7.1+cu118 torchaudio==2.7.1+cu118 \
+  --index-url https://download.pytorch.org/whl/cu118 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu118_py312.txt
 ```
 
-### 2. Install FFmpeg
-If you have FFmpeg and FFprobe installed on your computer, you can skip this step.
+Verify Torch and CUDA:
 
-#### For Ubuntu/Debian users
 ```bash
-sudo apt install ffmpeg
+python -c "import torch; print('torch:', torch.__version__); print('cuda:', torch.version.cuda); print('cuda available:', torch.cuda.is_available())"
 ```
-#### For MacOS users
+
+The application also checks NVIDIA GPU memory and compute capability. GPUs with roughly less than 4 GiB or below SM 5.3 use the CPU path.
+
+### Package indexes
+
+The three `requirments_*.txt` files define their package indexes at the top. Keep the default mirrors in mainland China. To use official indexes, replace only `--index-url` and `--extra-index-url`; keep package versions, CUDA suffixes, and the two-stage order unchanged.
+
+| Default mirror | Official source |
+| --- | --- |
+| `https://mirrors.pku.edu.cn/pypi/simple` | `https://pypi.org/simple` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cpu` | `https://download.pytorch.org/whl/cpu` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu118` | `https://download.pytorch.org/whl/cu118` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu128` | `https://download.pytorch.org/whl/cu128` |
+
+## Models and runtime directories
+
+The WebUI creates runtime directories automatically. Download models from the [Hugging Face model repository](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main) and keep this layout:
+
+```text
+assets/
+├── hubert_base/
+│   ├── config.json
+│   ├── preprocessor_config.json
+│   └── pytorch_model.bin
+├── rmvpe/rmvpe.pt
+├── pretrained/
+├── pretrained_v2/
+├── uvr5_weights/
+├── weights/        # user RVC .pth models
+└── indices/        # user .index files
+logs/
+└── mute/           # training silence samples
+
+# Exact paths used by the code
+assets/hubert_base/config.json
+assets/hubert_base/preprocessor_config.json
+assets/hubert_base/pytorch_model.bin
+assets/rmvpe/rmvpe.pt
+assets/pretrained/*.pth
+assets/pretrained_v2/*.pth
+assets/uvr5_weights/*
+assets/weights/*.pth
+assets/indices/*.index
+logs/mute/*
+```
+
+### Download models
+
 ```bash
-brew install ffmpeg
+python -m pip install --upgrade huggingface_hub
+
+# Required for inference and feature extraction
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "hubert_base/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI rmvpe.pt --revision main \
+  --local-dir assets/rmvpe
+
+# Required for v1/v2 training
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "pretrained/*" "pretrained_v2/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI mute.zip --revision main \
+  --local-dir .model-downloads
+python -m zipfile -e .model-downloads/mute.zip logs
+
+# Required only for UVR5 vocal separation
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "uvr5_weights/*" --local-dir assets
 ```
-#### For Windwos users
-Download these files and place them in the root folder:
-- [ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffmpeg.exe)
 
-- [ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffprobe.exe)
+Windows AMD/Intel DirectML environments additionally need:
 
-## ROCm Support for AMD graphic cards (Linux only)
-To use ROCm on Linux install all required drivers as described [here](https://rocm.docs.amd.com/en/latest/deploy/linux/os-native/install.html).
+```bash
+hf download lj1995/VoiceConversionWebUI rmvpe.onnx --revision main \
+  --local-dir assets/rmvpe
+```
 
-On Arch use pacman to install the driver:
-````
-pacman -S rocm-hip-sdk rocm-opencl-sdk
-````
+The legacy `hubert_base.pt` file is not used by this branch. The current code uses the three-file Transformers model under `assets/hubert_base/`. FCPE is bundled with `torchfcpe`.
 
-You might also need to set these environment variables (e.g. on a RX6700XT):
-````
-export ROCM_PATH=/opt/rocm
-export HSA_OVERRIDE_GFX_VERSION=10.3.0
-````
-Make sure your user is part of the `render` and `video` group:
-````
-sudo usermod -aG render $USERNAME
-sudo usermod -aG video $USERNAME
-````
+### FFmpeg
+
+The Ubuntu setup command above installs FFmpeg. On Windows, place these files in the repository root:
+
+- [ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffmpeg.exe?download=true)
+- [ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffprobe.exe?download=true)
+
+## Start the WebUI
 
-## Get started
-### start up directly
-Use the following command to start WebUI:
 ```bash
 python webui.py
 ```
-### Use the integration package
-Download and extract file `RVC-beta.7z`, then follow the steps below according to your system:
-#### For Windows users
-双击`go-webui.bat`
-#### For MacOS users
+
+For a headless Ubuntu server:
+
 ```bash
-sh ./run.sh
+python webui.py --noautoopen
 ```
+
+The default port is `7865`. Put personal `.pth` models in `assets/weights/` and `.index` files in `assets/indices/`.
+
 ## Credits
 + [ContentVec](https://github.com/auspicious3000/contentvec/)
 + [VITS](https://github.com/jaywalnut310/vits)
@@ -194,7 +226,7 @@ sh ./run.sh
 + [audio-slicer](https://github.com/openvpi/audio-slicer)
 + [Vocal pitch extraction:RMVPE](https://github.com/Dream-High/RMVPE)
   + The pretrained model is trained and tested by [yxlllc](https://github.com/yxlllc/RMVPE) and [RVC-Boss](https://github.com/RVC-Boss).
-  
+
 ## Thanks to all contributors for their efforts
 <a href="https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/graphs/contributors" target="_blank">
   <img src="https://contrib.rocks/image?repo=RVC-Project/Retrieval-based-Voice-Conversion-WebUI" />
diff --git a/docs/fr/README.fr.md b/docs/fr/README.fr.md
index 039a592..7bdd2e9 100644
--- a/docs/fr/README.fr.md
+++ b/docs/fr/README.fr.md
@@ -1,20 +1,18 @@
 <div align="center">
 
 <h1>Retrieval-based-Voice-Conversion-WebUI</h1>
-Un framework simple et facile à utiliser pour la conversion vocale (modificateur de voix) basé sur VITS<br><br>
+Un framework simple et facile à utiliser pour la conversion du timbre vocal / le changement de voix.<br><br>
 
 [![madewithlove](https://img.shields.io/badge/made_with-%E2%9D%A4-red?style=for-the-badge&labelColor=orange
 )](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)
 
 <img src="https://counter.seku.su/cmoe?name=rvc&theme=r34" /><br>
 
-[![Open In Colab](https://img.shields.io/badge/Colab-F9AB00?style=for-the-badge&logo=googlecolab&color=525252)](https://colab.research.google.com/github/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/Retrieval_based_Voice_Conversion_WebUI.ipynb)
 [![Licence](https://img.shields.io/badge/LICENSE-MIT-green.svg?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
-[![Huggingface](https://img.shields.io/badge/🤗%20-Spaces-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
+[![Huggingface](https://img.shields.io/badge/🤗%20-Models-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
 
-[![Discord](https://img.shields.io/badge/RVC%20Developers-Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/HcsmBBGyVk)
 
-[**Journal de mise à jour**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/docs/Changelog_CN.md) | [**FAQ**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98%E8%A7%A3%E7%AD%94) | [**AutoDL·Formation d'un chanteur AI pour 5 centimes**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/Autodl%E8%AE%AD%E7%BB%83RVC%C2%B7AI%E6%AD%8C%E6%89%8B%E6%95%99%E7%A8%8B) | [**Enregistrement des expériences comparatives**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%AF%B9%E7%85%A7%E5%AE%9E%E9%AA%8C%C2%B7%E5%AE%9E%E9%AA%8C%E8%AE%B0%E5%BD%95) | [**Démonstration en ligne**](https://huggingface.co/spaces/Ricecake123/RVC-demo)
+[**Journal de mise à jour**](./Changelog_FR.md) | [**FAQ**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98%E8%A7%A3%E7%AD%94) | [**AutoDL·Formation d'un chanteur AI pour 5 centimes**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/Autodl%E8%AE%AD%E7%BB%83RVC%C2%B7AI%E6%AD%8C%E6%89%8B%E6%95%99%E7%A8%8B) | [**Enregistrement des expériences comparatives**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%AF%B9%E7%85%A7%E5%AE%9E%E9%AA%8C%C2%B7%E5%AE%9E%E9%AA%8C%E8%AE%B0%E5%BD%95) | [**Démonstration en ligne**](https://huggingface.co/spaces/Ricecake123/RVC-demo)
 
 </div>
 
@@ -39,119 +37,169 @@ Ce dépôt a les caractéristiques suivantes :
 + Interface web simple et facile à utiliser.
 + Peut appeler le modèle UVR5 pour séparer rapidement la voix et l'accompagnement.
 + Utilise l'algorithme de pitch vocal le plus avancé [InterSpeech2023-RMVPE](#projets-référencés) pour éliminer les problèmes de voix muette. Meilleurs résultats, plus rapide que crepe_full, et moins gourmand en ressources.
-+ Support d'accélération pour les cartes AMD et Intel.
++ Les systèmes AMD/Intel utilisent les dépendances CPU ; Windows peut utiliser DirectML et Linux utilise le CPU.
 
 ## Configuration de l'environnement
-Exécutez les commandes suivantes dans un environnement Python de version 3.8 ou supérieure.
 
-(Windows/Linux)  
-Installez d'abord les dépendances principales via pip :
+Cette branche cible **Python 3.12 x64**. Exécutez toutes les commandes depuis la racine du dépôt. Ubuntu 24.04 x86_64 est recommandé.
+
+### Ubuntu 24.04
+
 ```bash
-# Installez Pytorch et ses dépendances essentielles, sautez si déjà installé.
-# Voir : https://pytorch.org/get-started/locally/
-pip install torch torchvision torchaudio
+sudo apt update
+sudo apt install -y python3.12 python3.12-venv python3.12-dev ffmpeg unzip libsndfile1 libportaudio2
 
-# Pour les utilisateurs de Windows avec une architecture Nvidia Ampere (RTX30xx), en se basant sur l'expérience #21, spécifiez la version CUDA correspondante pour Pytorch.
-pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
-
-# Pour Linux + carte AMD, utilisez cette version de Pytorch:
-pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm5.4.2
+python3.12 -m venv .venv
+source .venv/bin/activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-Vous pouvez utiliser poetry pour installer les dépendances :
-```bash
-# Installez l'outil de gestion des dépendances Poetry, sautez si déjà installé.
-# Voir : https://python-poetry.org/docs/#installation
-curl -sSL https://install.python-poetry.org | python3 -
+### Windows
 
-# Installez les dépendances avec poetry.
-poetry install
+Installez Python 3.12 x64, puis créez un environnement virtuel :
+
+```powershell
+py -3.12 -m venv .venv
+.venv\Scripts\activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-Ou vous pouvez utiliser pip pour installer les dépendances :
+### Choisir les dépendances selon le matériel
+
+| Matériel | Installation |
+| --- | --- |
+| CPU, AMD, Intel | Utiliser `requirments_cpu_py312.txt` ; Windows peut utiliser DirectML, Linux utilise le CPU |
+| NVIDIA RTX série 50 | Installer d'abord Torch CUDA 12.8, puis `requirments_cu128_py312.txt` |
+| NVIDIA antérieure à la série RTX 50 | Installer d'abord Torch CUDA 11.8, puis `requirments_cu118_py312.txt` |
+
+#### CPU, AMD, Intel
+
 ```bash
-# Cartes Nvidia :
-pip install -r requirements.txt
-
-# Cartes AMD/Intel :
-pip install -r requirements-dml.txt
-
-# Cartes AMD sur Linux (ROCm)
-pip install -r requirements-amd.txt
+python -m pip install -r requirments_cpu_py312.txt
 ```
 
-------
-Les utilisateurs de Mac peuvent exécuter `run.sh` pour installer les dépendances :
+#### NVIDIA RTX série 50 : installation en deux étapes
+
 ```bash
-sh ./run.sh
+python -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 \
+  --index-url https://download.pytorch.org/whl/cu128 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu128_py312.txt
 ```
 
-## Préparation d'autres modèles pré-entraînés
-RVC nécessite d'autres modèles pré-entraînés pour l'inférence et la formation.
+#### NVIDIA antérieure à la série RTX 50 : installation en deux étapes
 
-Vous pouvez télécharger ces modèles depuis notre [espace Hugging Face](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/).
-
-Voici une liste des modèles et autres fichiers requis par RVC :
 ```bash
-./assets/hubert_base
-
-./assets/pretrained 
-
-./assets/uvr5_weights
-
-# Pour tester la version v2 du modèle, téléchargez également :
-
-./assets/pretrained_v2
-
-# Si vous utilisez Windows, vous pourriez avoir besoin de ces fichiers pour ffmpeg et ffprobe, sautez cette étape si vous avez déjà installé ffmpeg et ffprobe. Les utilisateurs d'ubuntu/debian peuvent installer ces deux bibliothèques avec apt install ffmpeg. Les utilisateurs de Mac peuvent les installer avec brew install ffmpeg (prérequis : avoir installé brew).
-
-# ./ffmpeg
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffmpeg.exe
-
-# ./ffprobe
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffprobe.exe
-
-# Si vous souhaitez utiliser le dernier algorithme RMVPE de pitch vocal, téléchargez les paramètres du modèle de pitch et placez-les dans le répertoire racine de RVC.
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.pt
-
-    # Les utilisateurs de cartes AMD/Intel nécessitant l'environnement DML doivent télécharger :
-
-    https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.onnx
-
+python -m pip install torch==2.7.1+cu118 torchaudio==2.7.1+cu118 \
+  --index-url https://download.pytorch.org/whl/cu118 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu118_py312.txt
 ```
-Ensuite, exécutez la commande suivante pour démarrer WebUI :
+
+Vérifiez Torch et CUDA :
+
+```bash
+python -c "import torch; print('torch:', torch.__version__); print('cuda:', torch.version.cuda); print('cuda available:', torch.cuda.is_available())"
+```
+
+Le programme vérifie aussi la mémoire et la capacité de calcul NVIDIA. Une carte d'environ moins de 4 Gio ou inférieure à SM 5.3 utilise le CPU.
+
+### Sources de paquets
+
+Les trois fichiers `requirments_*.txt` définissent leurs sources en haut. Pour utiliser les sources officielles, remplacez uniquement `--index-url` et `--extra-index-url`, en conservant les versions, les suffixes CUDA et l'ordre des deux étapes.
+
+| Default mirror | Official source |
+| --- | --- |
+| `https://mirrors.pku.edu.cn/pypi/simple` | `https://pypi.org/simple` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cpu` | `https://download.pytorch.org/whl/cpu` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu118` | `https://download.pytorch.org/whl/cu118` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu128` | `https://download.pytorch.org/whl/cu128` |
+
+## Modèles et répertoires d'exécution
+
+Le WebUI crée automatiquement les répertoires d'exécution. Téléchargez les modèles depuis le [dépôt de modèles Hugging Face](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main) et conservez cette structure :
+
+```text
+assets/
+├── hubert_base/
+│   ├── config.json
+│   ├── preprocessor_config.json
+│   └── pytorch_model.bin
+├── rmvpe/rmvpe.pt
+├── pretrained/
+├── pretrained_v2/
+├── uvr5_weights/
+├── weights/        # user RVC .pth models
+└── indices/        # user .index files
+logs/
+└── mute/           # training silence samples
+
+# Exact paths used by the code
+assets/hubert_base/config.json
+assets/hubert_base/preprocessor_config.json
+assets/hubert_base/pytorch_model.bin
+assets/rmvpe/rmvpe.pt
+assets/pretrained/*.pth
+assets/pretrained_v2/*.pth
+assets/uvr5_weights/*
+assets/weights/*.pth
+assets/indices/*.index
+logs/mute/*
+```
+
+### Télécharger les modèles
+
+```bash
+python -m pip install --upgrade huggingface_hub
+
+# Required for inference and feature extraction
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "hubert_base/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI rmvpe.pt --revision main \
+  --local-dir assets/rmvpe
+
+# Required for v1/v2 training
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "pretrained/*" "pretrained_v2/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI mute.zip --revision main \
+  --local-dir .model-downloads
+python -m zipfile -e .model-downloads/mute.zip logs
+
+# Required only for UVR5 vocal separation
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "uvr5_weights/*" --local-dir assets
+```
+
+Les environnements Windows AMD/Intel DirectML nécessitent aussi :
+
+```bash
+hf download lj1995/VoiceConversionWebUI rmvpe.onnx --revision main \
+  --local-dir assets/rmvpe
+```
+
+L'ancien fichier `hubert_base.pt` n'est pas utilisé ici. Le code actuel utilise le modèle Transformers dans `assets/hubert_base/`. FCPE est inclus dans `torchfcpe`.
+
+### FFmpeg
+
+La commande Ubuntu précédente installe FFmpeg. Sous Windows, placez ces fichiers à la racine du dépôt :
+
+- [ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffmpeg.exe?download=true)
+- [ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffprobe.exe?download=true)
+
+## Démarrer le WebUI
+
 ```bash
 python webui.py
 ```
 
-Si vous utilisez Windows ou macOS, vous pouvez télécharger et extraire `RVC-beta.7z`. Les utilisateurs de Windows peuvent exécuter `go-webui.bat` pour démarrer WebUI, tandis que les utilisateurs de macOS peuvent exécuter `sh ./run.sh`.
+Serveur Ubuntu sans interface graphique :
 
-## Compatibilité ROCm pour les cartes AMD (seulement Linux)
-Installez tous les pilotes décrits [ici](https://rocm.docs.amd.com/en/latest/deploy/linux/os-native/install.html).
-
-Sur Arch utilisez pacman pour installer le pilote:
-````
-pacman -S rocm-hip-sdk rocm-opencl-sdk
-````
-
-Vous devrez peut-être créer ces variables d'environnement (par exemple avec RX6700XT):
-````
-export ROCM_PATH=/opt/rocm
-export HSA_OVERRIDE_GFX_VERSION=10.3.0
-````
-Assurez-vous que votre utilisateur est dans les groupes `render` et `video`:
-````
-sudo usermod -aG render $USERNAME
-sudo usermod -aG video $USERNAME
-````
-Enfin vous pouvez exécuter WebUI:
 ```bash
-python webui.py
+python webui.py --noautoopen
 ```
 
+Le port par défaut est `7865`. Placez les modèles `.pth` dans `assets/weights/` et les fichiers `.index` dans `assets/indices/`.
+
 ## Crédits
 + [ContentVec](https://github.com/auspicious3000/contentvec/)
 + [VITS](https://github.com/jaywalnut310/vits)
diff --git a/docs/jp/README.ja.md b/docs/jp/README.ja.md
index 715e503..c9445c9 100644
--- a/docs/jp/README.ja.md
+++ b/docs/jp/README.ja.md
@@ -1,17 +1,15 @@
 <div align="center">
 
 <h1>Retrieval-based-Voice-Conversion-WebUI</h1>
-VITSに基づく使いやすい音声変換（voice changer）framework<br><br>
+シンプルで使いやすい声質変換／ボイスチェンジャーフレームワーク。<br><br>
 
 [![madewithlove](https://img.shields.io/badge/made_with-%E2%9D%A4-red?style=for-the-badge&labelColor=orange)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)
 
 <img src="https://counter.seku.su/cmoe?name=rvc&theme=r34" /><br>
 
-[![Open In Colab](https://img.shields.io/badge/Colab-F9AB00?style=for-the-badge&logo=googlecolab&color=525252)](https://colab.research.google.com/github/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/Retrieval_based_Voice_Conversion_WebUI.ipynb)
 [![Licence](https://img.shields.io/badge/LICENSE-MIT-green.svg?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
-[![Huggingface](https://img.shields.io/badge/🤗%20-Spaces-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
+[![Huggingface](https://img.shields.io/badge/🤗%20-Models-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
 
-[![Discord](https://img.shields.io/badge/RVC%20Developers-Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/HcsmBBGyVk)
 
 [**更新日誌**](./Changelog_JA.md) | [**よくある質問**](./faq_ja.md) | [**AutoDL·5 円で AI 歌手をトレーニング**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/Autodl%E8%AE%AD%E7%BB%83RVC%C2%B7AI%E6%AD%8C%E6%89%8B%E6%95%99%E7%A8%8B) | [**対照実験記録**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%AF%B9%E7%85%A7%E5%AE%9E%E9%AA%8C%C2%B7%E5%AE%9E%E9%AA%8C%E8%AE%B0%E5%BD%95) | [**オンラインデモ**](https://modelscope.cn/studios/FlowerCry/RVCv2demo)
 
@@ -63,168 +61,164 @@ VITSに基づく使いやすい音声変換（voice changer）framework<br><br>
 
 ## 環境構築
 
-下記のコマンドは、Python3.8 以上の環境で実行する必要があります:
+このブランチは **Python 3.12 x64** を対象としています。すべてのコマンドはリポジトリのルートで実行してください。Ubuntu 24.04 x86_64 を推奨します。
 
-### Windows/Linux/MacOS などのプラットフォーム共通方法
-
-以下の方法のいずれかを選択してください。
-
-#### 1. pip を通じた依存関係のインストール
-
-1. Pytorch 及びその主要な依存関係のインストール、すでにインストールされている場合はスキップ。参照：https://pytorch.org/get-started/locally/
+### Ubuntu 24.04
 
 ```bash
-pip install torch torchvision torchaudio
+sudo apt update
+sudo apt install -y python3.12 python3.12-venv python3.12-dev ffmpeg unzip libsndfile1 libportaudio2
+
+python3.12 -m venv .venv
+source .venv/bin/activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-2. win システム + Nvidia Ampere アーキテクチャ（RTX30xx）の場合、#21 の経験に基づいて pytorch に対応する cuda バージョンを指定
+### Windows
+
+Python 3.12 x64 をインストールし、仮想環境を作成します。
+
+```powershell
+py -3.12 -m venv .venv
+.venv\Scripts\activate
+python -m pip install --upgrade pip setuptools wheel
+```
+
+### ハードウェア別の依存関係
+
+| ハードウェア | インストール方法 |
+| --- | --- |
+| CPU、AMD、Intel | `requirments_cpu_py312.txt` を使用。Windows は DirectML、Linux は CPU を使用 |
+| NVIDIA RTX 50 シリーズ | CUDA 12.8 版 Torch を先にインストールし、その後 `requirments_cu128_py312.txt` |
+| RTX 50 シリーズより前の NVIDIA | CUDA 11.8 版 Torch を先にインストールし、その後 `requirments_cu118_py312.txt` |
+
+#### CPU、AMD、Intel
 
 ```bash
-pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
+python -m pip install -r requirments_cpu_py312.txt
 ```
 
-3. 自分のグラフィックカードに合わせた依存関係のインストール
-
-- N カード
+#### NVIDIA RTX 50 シリーズ：2 段階
 
 ```bash
-pip install -r requirements.txt
+python -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 \
+  --index-url https://download.pytorch.org/whl/cu128 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu128_py312.txt
 ```
 
-- A カード/I カード
+#### RTX 50 シリーズより前の NVIDIA：2 段階
 
 ```bash
-pip install -r requirements-dml.txt
+python -m pip install torch==2.7.1+cu118 torchaudio==2.7.1+cu118 \
+  --index-url https://download.pytorch.org/whl/cu118 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu118_py312.txt
 ```
 
-- A カード ROCM(Linux)
+Torch と CUDA を確認します。
 
 ```bash
-pip install -r requirements-amd.txt
+python -c "import torch; print('torch:', torch.__version__); print('cuda:', torch.version.cuda); print('cuda available:', torch.cuda.is_available())"
 ```
 
-#### 2. poetry を通じた依存関係のインストール
+アプリは NVIDIA GPU のメモリと計算能力も確認します。約 4 GiB 未満、または SM 5.3 未満の場合は CPU を使用します。
 
-Poetry 依存関係管理ツールのインストール、すでにインストールされている場合はスキップ。参照：https://python-poetry.org/docs/#installation
+### パッケージのダウンロード元
+
+3 つの `requirments_*.txt` の先頭にダウンロード元があります。公式の配布元を使用する場合は `--index-url` と `--extra-index-url` だけを置き換え、バージョン、CUDA 接尾辞、2 段階の順序は維持してください。
+
+| Default mirror | Official source |
+| --- | --- |
+| `https://mirrors.pku.edu.cn/pypi/simple` | `https://pypi.org/simple` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cpu` | `https://download.pytorch.org/whl/cpu` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu118` | `https://download.pytorch.org/whl/cu118` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu128` | `https://download.pytorch.org/whl/cu128` |
+
+## モデルと実行ディレクトリ
+
+WebUI は実行ディレクトリを自動作成します。[Hugging Face モデルリポジトリ](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main)からモデルをダウンロードし、次の構成を維持してください。
+
+```text
+assets/
+├── hubert_base/
+│   ├── config.json
+│   ├── preprocessor_config.json
+│   └── pytorch_model.bin
+├── rmvpe/rmvpe.pt
+├── pretrained/
+├── pretrained_v2/
+├── uvr5_weights/
+├── weights/        # user RVC .pth models
+└── indices/        # user .index files
+logs/
+└── mute/           # training silence samples
+
+# Exact paths used by the code
+assets/hubert_base/config.json
+assets/hubert_base/preprocessor_config.json
+assets/hubert_base/pytorch_model.bin
+assets/rmvpe/rmvpe.pt
+assets/pretrained/*.pth
+assets/pretrained_v2/*.pth
+assets/uvr5_weights/*
+assets/weights/*.pth
+assets/indices/*.index
+logs/mute/*
+```
+
+### モデルのダウンロード
 
 ```bash
-curl -sSL https://install.python-poetry.org | python3 -
+python -m pip install --upgrade huggingface_hub
+
+# Required for inference and feature extraction
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "hubert_base/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI rmvpe.pt --revision main \
+  --local-dir assets/rmvpe
+
+# Required for v1/v2 training
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "pretrained/*" "pretrained_v2/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI mute.zip --revision main \
+  --local-dir .model-downloads
+python -m zipfile -e .model-downloads/mute.zip logs
+
+# Required only for UVR5 vocal separation
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "uvr5_weights/*" --local-dir assets
 ```
 
-poetry を使って依存関係をインストール
+Windows の AMD/Intel DirectML 環境では、さらに次のファイルが必要です。
 
 ```bash
-poetry install
+hf download lj1995/VoiceConversionWebUI rmvpe.onnx --revision main \
+  --local-dir assets/rmvpe
 ```
 
-### MacOS
+旧形式の `hubert_base.pt` はこのブランチでは使用しません。現在は `assets/hubert_base/` の Transformers モデルを使用します。FCPE は `torchfcpe` に含まれます。
 
-`run.sh`を使って依存関係をインストールできます
+### FFmpeg
+
+上記の Ubuntu コマンドで FFmpeg がインストールされます。Windows では次のファイルをリポジトリのルートに配置します。
+
+- [ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffmpeg.exe?download=true)
+- [ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffprobe.exe?download=true)
+
+## WebUI の起動
 
 ```bash
-sh ./run.sh
-```
-
-## その他の事前訓練されたモデルの準備
-
-RVC は推論とトレーニングのために他のいくつかの事前訓練されたモデルが必要です。
-
-これらのモデルは私たちの[Hugging Face space](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)でダウンロードできます。
-
-### 1. assets のダウンロード
-
-以下は、RVC に必要なすべての事前学習モデルとその他のファイルのリストです。`tools`フォルダーにこれらをダウンロードするスクリプトがあります。
-
-- ./assets/hubert_base
-
-- ./assets/pretrained
-
-- ./assets/uvr5_weights
-
-v2 バージョンのモデルを使用する場合、追加で以下をダウンロードする必要があります。
-
-- ./assets/pretrained_v2
-
-### 2. ffmpeg のインストール
-
-ffmpeg と ffprobe が既にインストールされている場合はスキップします。
-
-#### Ubuntu/Debian ユーザー
-
-```bash
-sudo apt install ffmpeg
-```
-
-#### MacOS ユーザー
-
-```bash
-brew install ffmpeg
-```
-
-#### Windows ユーザー
-
-ダウンロード後、ルートディレクトリに配置してください。
-
-- [ffmpeg.exe をダウンロード](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffmpeg.exe)
-
-- [ffprobe.exe をダウンロード](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffprobe.exe)
-
-### 3. RMVPE 人間の声のピッチ抽出アルゴリズムに必要なファイルのダウンロード
-
-最新の RMVPE 人間の声のピッチ抽出アルゴリズムを使用する場合、ピッチ抽出モデルのパラメータをダウンロードして RVC のルートディレクトリに配置する必要があります。
-
-- [rmvpe.pt をダウンロード](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.pt)
-
-#### dml 環境の RMVPE をダウンロード(オプション、A カード/I カードユーザー)
-
-- [rmvpe.onnx をダウンロード](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.onnx)
-
-### 4. AMD グラフィックカード Rocm(オプション、Linux のみ)
-
-Linux システムで AMD の Rocm 技術をベースに RVC を実行したい場合、[こちら](https://rocm.docs.amd.com/en/latest/deploy/linux/os-native/install.html)で必要なドライバーを先にインストールしてください。
-
-Arch Linux を使用している場合、pacman を使用して必要なドライバーをインストールできます。
-
-```
-pacman -S rocm-hip-sdk rocm-opencl-sdk
-```
-
-一部のモデルのグラフィックカード（例：RX6700XT）の場合、以下のような環境変数を追加で設定する必要があるかもしれません。
-
-```
-export ROCM_PATH=/opt/rocm
-export HSA_OVERRIDE_GFX_VERSION=10.3.0
-```
-
-同時に、現在のユーザーが`render`および`video`ユーザーグループに属していることを確認してください。
-
-```
-sudo usermod -aG render $USERNAME
-sudo usermod -aG video $USERNAME
-```
-
-## 使用開始
-
-### 直接起動
-
-以下のコマンドで WebUI を起動します
-'''bash
 python webui.py
-'''
+```
 
-### 統合パッケージの使用
+画面のない Ubuntu サーバー：
 
-`RVC-beta.7z`をダウンロードして解凍
+```bash
+python webui.py --noautoopen
+```
 
-#### Windows ユーザー
-
-`go-webui.bat`をダブルクリック
-
-#### MacOS ユーザー
-
-'''bash
-sh ./run.sh
-'''
+既定のポートは `7865` です。`.pth` モデルは `assets/weights/`、`.index` ファイルは `assets/indices/` に配置します。
 
 ## 参考プロジェクト
 
diff --git a/docs/kr/README.ko.han.md b/docs/kr/README.ko.han.md
index 507dbe4..5f6e57f 100644
--- a/docs/kr/README.ko.han.md
+++ b/docs/kr/README.ko.han.md
@@ -1,23 +1,21 @@
 <div align="center">
 
 <h1>Retrieval-based-Voice-Conversion-WebUI</h1>
-VITS基盤의 簡單하고使用하기 쉬운音聲變換틀<br><br>
+簡單하고 使用하기 쉬운 音色變換/變聲器 프레임워크.<br><br>
 
 [![madewithlove](https://img.shields.io/badge/made_with-%E2%9D%A4-red?style=for-the-badge&labelColor=orange
 )](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)
 
 <img src="https://counter.seku.su/cmoe?name=rvc&theme=r34" /><br>
-  
-[![Open In Colab](https://img.shields.io/badge/Colab-F9AB00?style=for-the-badge&logo=googlecolab&color=525252)](https://colab.research.google.com/github/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/Retrieval_based_Voice_Conversion_WebUI.ipynb)
-[![Licence](https://img.shields.io/github/license/RVC-Project/Retrieval-based-Voice-Conversion-WebUI?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
-[![Huggingface](https://img.shields.io/badge/🤗%20-Spaces-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
 
-[![Discord](https://img.shields.io/badge/RVC%20Developers-Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/HcsmBBGyVk)
+[![Licence](https://img.shields.io/github/license/RVC-Project/Retrieval-based-Voice-Conversion-WebUI?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
+[![Huggingface](https://img.shields.io/badge/🤗%20-Models-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
+
 
 </div>
 
 ------
-[**更新日誌**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/docs/Changelog_KO.md)
+[**更新日誌**](./Changelog_KO.md)
 
 [**English**](../en/README.en.md) | [**中文简体**](../../README.md) | [**日本語**](../jp/README.ja.md) | [**한국어**](../kr/README.ko.md) ([**韓國語**](../kr/README.ko.han.md)) | [**Français**](../fr/README.fr.md) | [**Türkçe**](../tr/README.tr.md) | [**Português**](../pt/README.pt.md)
 
@@ -38,56 +36,159 @@ VITS基盤의 簡單하고使用하기 쉬운音聲變換틀<br><br>
 + 使用하기 쉬운 WebUI (웹 使用者인터페이스);
 + UVR5 모델을 利用하여 목소리와 背景音樂의 빠른 分離;
 
-## 環境의準備
-poetry를通해 依存를設置하는 것을 勸獎합니다.
+## 環境의 準備
 
-다음命令은 Python 버전3.8以上의環境에서 實行되어야 합니다:
-```bash
-# PyTorch 關聯主要依存設置, 이미設置되어 있는 境遇 건너뛰기 可能
-# 參照: https://pytorch.org/get-started/locally/
-pip install torch torchvision torchaudio
+이 브랜치는 **Python 3.12 x64**를 對象으로 합니다. 모든 命令은 저장소 根目錄에서 실행하세요. Ubuntu 24.04 x86_64를 권장합니다.
 
-# Windows + Nvidia Ampere Architecture(RTX30xx)를 使用하고 있다面, #21 에서 명시된 것과 같이 PyTorch에 맞는 CUDA 버전을 指定해야 합니다.
-#pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
-
-# Poetry 設置, 이미設置되어 있는 境遇 건너뛰기 可能
-# Reference: https://python-poetry.org/docs/#installation
-curl -sSL https://install.python-poetry.org | python3 -
-
-# 依存設置
-poetry install
-```
-pip를 活用하여依存를 設置하여도 無妨합니다.
+### Ubuntu 24.04
 
 ```bash
-pip install -r requirements.txt
+sudo apt update
+sudo apt install -y python3.12 python3.12-venv python3.12-dev ffmpeg unzip libsndfile1 libportaudio2
+
+python3.12 -m venv .venv
+source .venv/bin/activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-## 其他預備모델準備
-RVC 모델은 推論과訓練을 依하여 다른 預備모델이 必要합니다.
+### Windows
 
-[Huggingface space](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)를 通해서 다운로드 할 수 있습니다.
+Python 3.12 x64를 설치한 뒤 假想環境을 만드세요.
+
+```powershell
+py -3.12 -m venv .venv
+.venv\Scripts\activate
+python -m pip install --upgrade pip setuptools wheel
+```
+
+### 하드웨어別 依存性
+
+| 하드웨어 | 設置 方法 |
+| --- | --- |
+| CPU, AMD, Intel | `requirments_cpu_py312.txt` 使用. Windows는 DirectML, Linux는 CPU 使用 |
+| NVIDIA RTX 50 系列 | CUDA 12.8 Torch를 먼저 설치한 뒤 `requirments_cu128_py312.txt` 設置 |
+| RTX 50 系列 以前 NVIDIA | CUDA 11.8 Torch를 먼저 설치한 뒤 `requirments_cu118_py312.txt` 設置 |
+
+#### CPU, AMD, Intel
 
-다음은 RVC에 必要한 預備모델 및 其他 파일 目錄입니다:
 ```bash
-./assets/hubert_base
-
-./assets/pretrained 
-
-./assets/uvr5_weights
-
-V2 버전 모델을 테스트하려면 추가 다운로드가 필요합니다.
-
-./assets/pretrained_v2
-
-# Windows를 使用하는境遇 이 사전도 必要할 수 있습니다. FFmpeg가 設置되어 있으면 건너뛰어도 됩니다.
-ffmpeg.exe
+python -m pip install -r requirments_cpu_py312.txt
 ```
-그後 以下의 命令을 使用하여 WebUI를 始作할 수 있습니다:
+
+#### NVIDIA RTX 50 系列：2段階 設置
+
+```bash
+python -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 \
+  --index-url https://download.pytorch.org/whl/cu128 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu128_py312.txt
+```
+
+#### RTX 50 系列 以前 NVIDIA：2段階 設置
+
+Torch와 CUDA 狀態를 確認하세요.
+
+```bash
+python -c "import torch; print('torch:', torch.__version__); print('cuda:', torch.version.cuda); print('cuda available:', torch.cuda.is_available())"
+```
+
+프로그램은 NVIDIA GPU 記憶體와 計算能力도 확인합니다. 약 4 GiB 미만이거나 SM 5.3 미만이면 CPU 經路를 사용합니다.
+
+### 패키지 來源
+
+세 개의 `requirments_*.txt` 上端에 下載 來源이 있습니다. 공식 소스를 사용할 때는 `--index-url`과 `--extra-index-url`만 교체하고 버전, CUDA 접미사, 2段階 순서는 유지하세요.
+
+| Default mirror | Official source |
+| --- | --- |
+| `https://mirrors.pku.edu.cn/pypi/simple` | `https://pypi.org/simple` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cpu` | `https://download.pytorch.org/whl/cpu` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu118` | `https://download.pytorch.org/whl/cu118` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu128` | `https://download.pytorch.org/whl/cu128` |
+
+## 모델과 運行 目錄
+
+WebUI는 運行 目錄를 자동 생성합니다. [Hugging Face 모델 저장소](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main)에서 모델을 받고 다음 구조를 유지하세요.
+
+```text
+assets/
+├── hubert_base/
+│   ├── config.json
+│   ├── preprocessor_config.json
+│   └── pytorch_model.bin
+├── rmvpe/rmvpe.pt
+├── pretrained/
+├── pretrained_v2/
+├── uvr5_weights/
+├── weights/        # user RVC .pth models
+└── indices/        # user .index files
+logs/
+└── mute/           # training silence samples
+
+# Exact paths used by the code
+assets/hubert_base/config.json
+assets/hubert_base/preprocessor_config.json
+assets/hubert_base/pytorch_model.bin
+assets/rmvpe/rmvpe.pt
+assets/pretrained/*.pth
+assets/pretrained_v2/*.pth
+assets/uvr5_weights/*
+assets/weights/*.pth
+assets/indices/*.index
+logs/mute/*
+```
+
+### 모델 下載
+
+```bash
+python -m pip install --upgrade huggingface_hub
+
+# Required for inference and feature extraction
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "hubert_base/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI rmvpe.pt --revision main \
+  --local-dir assets/rmvpe
+
+# Required for v1/v2 training
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "pretrained/*" "pretrained_v2/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI mute.zip --revision main \
+  --local-dir .model-downloads
+python -m zipfile -e .model-downloads/mute.zip logs
+
+# Required only for UVR5 vocal separation
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "uvr5_weights/*" --local-dir assets
+```
+
+Windows AMD/Intel DirectML 환경에는 다음 파일도 필요합니다.
+
+```bash
+hf download lj1995/VoiceConversionWebUI rmvpe.onnx --revision main \
+  --local-dir assets/rmvpe
+```
+
+舊形式 `hubert_base.pt`는 이 브랜치에서 사용하지 않습니다. 현재 코드는 `assets/hubert_base/`의 Transformers 모델을 사용합니다. FCPE는 `torchfcpe`에 포함됩니다.
+
+### FFmpeg
+
+위 Ubuntu 명령은 FFmpeg를 설치합니다. Windows에서는 다음 파일을 저장소 根目錄에 배치하세요.
+
+- [ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffmpeg.exe?download=true)
+- [ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffprobe.exe?download=true)
+
+## WebUI 始作
+
 ```bash
 python webui.py
 ```
-Windows를 使用하는境遇 `RVC-beta.7z`를 다운로드 및 壓縮解除하여 RVC를 直接使用하거나 `go-webui.bat`을 使用하여 WebUi를 直接할 수 있습니다.
+
+화면이 없는 Ubuntu 서버:
+
+```bash
+python webui.py --noautoopen
+```
+
+基本 포트는 `7865`입니다. `.pth` 모델은 `assets/weights/`, `.index` 파일은 `assets/indices/`에 배치하세요.
 
 ## 參考
 + [ContentVec](https://github.com/auspicious3000/contentvec/)
diff --git a/docs/kr/README.ko.md b/docs/kr/README.ko.md
index b65f13d..4e7eaa8 100644
--- a/docs/kr/README.ko.md
+++ b/docs/kr/README.ko.md
@@ -1,17 +1,15 @@
 <div align="center">
 
 <h1>Retrieval-based-Voice-Conversion-WebUI</h1>
-VITS 기반의 간단하고 사용하기 쉬운 음성 변환 프레임워크.<br><br>
+간단하고 사용하기 쉬운 음색 변환/보이스 체인저 프레임워크.<br><br>
 
 [![madewithlove](https://img.shields.io/badge/made_with-%E2%9D%A4-red?style=for-the-badge&labelColor=orange)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)
 
 <img src="https://counter.seku.su/cmoe?name=rvc&theme=r34" /><br>
 
-[![Open In Colab](https://img.shields.io/badge/Colab-F9AB00?style=for-the-badge&logo=googlecolab&color=525252)](https://colab.research.google.com/github/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/Retrieval_based_Voice_Conversion_WebUI.ipynb)
 [![Licence](https://img.shields.io/badge/LICENSE-MIT-green.svg?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
-[![Huggingface](https://img.shields.io/badge/🤗%20-Spaces-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
+[![Huggingface](https://img.shields.io/badge/🤗%20-Models-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
 
-[![Discord](https://img.shields.io/badge/RVC%20Developers-Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/HcsmBBGyVk)
 
 [**업데이트 로그**](./Changelog_KO.md) | [**자주 묻는 질문**](./faq_ko.md) | [**AutoDL·5원으로 AI 가수 훈련**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/Autodl%E8%AE%AD%E7%BB%83RVC%C2%B7AI%E6%AD%8C%E6%89%8B%E6%95%99%E7%A8%8B) | [**대조 실험 기록**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/%E5%AF%B9%E7%85%A7%E5%AE%9E%E9%AA%8C%C2%B7%E5%AE%9E%E9%AA%8C%E8%AE%B0%E5%BD%95) | [**온라인 데모**](https://modelscope.cn/studios/FlowerCry/RVCv2demo)
 
@@ -63,170 +61,165 @@ VITS 기반의 간단하고 사용하기 쉬운 음성 변환 프레임워크.<b
 
 ## 환경 설정
 
-다음 명령은 Python 버전이 3.8 이상인 환경에서 실행해야 합니다.
+이 브랜치는 **Python 3.12 x64**를 대상으로 합니다. 모든 명령은 저장소 루트에서 실행하세요. Ubuntu 24.04 x86_64를 권장합니다.
 
-### Windows/Linux/MacOS 등 플랫폼 공통 방법
-
-아래 방법 중 하나를 선택하세요.
-
-#### 1. pip를 통한 의존성 설치
-
-1. Pytorch 및 의존성 모듈 설치, 이미 설치되어 있으면 생략. 참조: https://pytorch.org/get-started/locally/
+### Ubuntu 24.04
 
 ```bash
-pip install torch torchvision torchaudio
+sudo apt update
+sudo apt install -y python3.12 python3.12-venv python3.12-dev ffmpeg unzip libsndfile1 libportaudio2
+
+python3.12 -m venv .venv
+source .venv/bin/activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-2. win 시스템 + Nvidia Ampere 아키텍처(RTX30xx) 사용 시, #21의 사례에 따라 pytorch에 해당하는 cuda 버전을 지정
+### Windows
+
+Python 3.12 x64를 설치한 다음 가상 환경을 만드세요.
+
+```powershell
+py -3.12 -m venv .venv
+.venv\Scripts\activate
+python -m pip install --upgrade pip setuptools wheel
+```
+
+### 하드웨어별 의존성 선택
+
+| 하드웨어 | 설치 방법 |
+| --- | --- |
+| CPU, AMD, Intel | `requirments_cpu_py312.txt` 사용. Windows는 DirectML, Linux는 CPU 사용 |
+| NVIDIA RTX 50 시리즈 | CUDA 12.8 Torch를 먼저 설치한 뒤 `requirments_cu128_py312.txt` 설치 |
+| RTX 50 시리즈 이전 NVIDIA | CUDA 11.8 Torch를 먼저 설치한 뒤 `requirments_cu118_py312.txt` 설치 |
+
+#### CPU, AMD, Intel
 
 ```bash
-pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
+python -m pip install -r requirments_cpu_py312.txt
 ```
 
-3. 자신의 그래픽 카드에 맞는 의존성 설치
-
-- N카드
+#### NVIDIA RTX 50 시리즈: 2단계 설치
 
 ```bash
-pip install -r requirements.txt
+python -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 \
+  --index-url https://download.pytorch.org/whl/cu128 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu128_py312.txt
 ```
 
-- A카드/I카드
+#### RTX 50 시리즈 이전 NVIDIA: 2단계 설치
 
 ```bash
-pip install -r requirements-dml.txt
+python -m pip install torch==2.7.1+cu118 torchaudio==2.7.1+cu118 \
+  --index-url https://download.pytorch.org/whl/cu118 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu118_py312.txt
 ```
 
-- A카드ROCM(Linux)
+Torch와 CUDA 상태를 확인하세요.
 
 ```bash
-pip install -r requirements-amd.txt
+python -c "import torch; print('torch:', torch.__version__); print('cuda:', torch.version.cuda); print('cuda available:', torch.cuda.is_available())"
 ```
 
-#### 2. poetry를 통한 의존성 설치
+프로그램은 NVIDIA GPU 메모리와 연산 능력도 확인합니다. 약 4 GiB 미만이거나 SM 5.3 미만이면 CPU 경로를 사용합니다.
 
-Poetry 의존성 관리 도구 설치, 이미 설치된 경우 생략. 참조: https://python-poetry.org/docs/#installation
+### 패키지 다운로드 소스
+
+세 개의 `requirments_*.txt` 파일 맨 위에 다운로드 소스가 있습니다. 공식 소스를 사용할 때는 `--index-url`과 `--extra-index-url`만 교체하고 버전, CUDA 접미사, 2단계 순서는 유지하세요.
+
+| Default mirror | Official source |
+| --- | --- |
+| `https://mirrors.pku.edu.cn/pypi/simple` | `https://pypi.org/simple` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cpu` | `https://download.pytorch.org/whl/cpu` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu118` | `https://download.pytorch.org/whl/cu118` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu128` | `https://download.pytorch.org/whl/cu128` |
+
+## 모델 및 실행 디렉터리
+
+WebUI는 실행 디렉터리를 자동 생성합니다. [Hugging Face 모델 저장소](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main)에서 모델을 받고 다음 구조를 유지하세요.
+
+```text
+assets/
+├── hubert_base/
+│   ├── config.json
+│   ├── preprocessor_config.json
+│   └── pytorch_model.bin
+├── rmvpe/rmvpe.pt
+├── pretrained/
+├── pretrained_v2/
+├── uvr5_weights/
+├── weights/        # user RVC .pth models
+└── indices/        # user .index files
+logs/
+└── mute/           # training silence samples
+
+# Exact paths used by the code
+assets/hubert_base/config.json
+assets/hubert_base/preprocessor_config.json
+assets/hubert_base/pytorch_model.bin
+assets/rmvpe/rmvpe.pt
+assets/pretrained/*.pth
+assets/pretrained_v2/*.pth
+assets/uvr5_weights/*
+assets/weights/*.pth
+assets/indices/*.index
+logs/mute/*
+```
+
+### 모델 다운로드
 
 ```bash
-curl -sSL https://install.python-poetry.org | python3 -
+python -m pip install --upgrade huggingface_hub
+
+# Required for inference and feature extraction
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "hubert_base/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI rmvpe.pt --revision main \
+  --local-dir assets/rmvpe
+
+# Required for v1/v2 training
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "pretrained/*" "pretrained_v2/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI mute.zip --revision main \
+  --local-dir .model-downloads
+python -m zipfile -e .model-downloads/mute.zip logs
+
+# Required only for UVR5 vocal separation
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "uvr5_weights/*" --local-dir assets
 ```
 
-poetry를 통한 의존성 설치
+Windows AMD/Intel DirectML 환경에는 다음 파일도 필요합니다.
 
 ```bash
-poetry install
+hf download lj1995/VoiceConversionWebUI rmvpe.onnx --revision main \
+  --local-dir assets/rmvpe
 ```
 
-### MacOS
+이전 형식인 `hubert_base.pt`는 이 브랜치에서 사용하지 않습니다. 현재 코드는 `assets/hubert_base/`의 Transformers 모델을 사용합니다. FCPE는 `torchfcpe`에 포함됩니다.
 
-`run.sh`를 통해 의존성 설치 가능
+### FFmpeg
 
-```bash
-sh ./run.sh
-```
+위 Ubuntu 명령은 FFmpeg를 설치합니다. Windows에서는 다음 파일을 저장소 루트에 배치하세요.
 
-## 기타 사전 훈련된 모델 준비
+- [ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffmpeg.exe?download=true)
+- [ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffprobe.exe?download=true)
 
-RVC는 추론과 훈련을 위해 다른 일부 사전 훈련된 모델이 필요합니다.
-
-이러한 모델은 저희의 [Hugging Face space](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)에서 다운로드할 수 있습니다.
-
-### 1. assets 다운로드
-
-다음은 RVC에 필요한 모든 사전 훈련된 모델과 기타 파일의 목록입니다. `tools` 폴더에서 이들을 다운로드하는 스크립트를 찾을 수 있습니다.
-
-- ./assets/hubert_base
-
-- ./assets/pretrained
-
-- ./assets/uvr5_weights
-
-v2 버전 모델을 사용하려면 추가로 다음을 다운로드해야 합니다.
-
-- ./assets/pretrained_v2
-
-### 2. ffmpeg 설치
-
-ffmpeg와 ffprobe가 이미 설치되어 있다면 건너뜁니다.
-
-#### Ubuntu/Debian 사용자
-
-```bash
-sudo apt install ffmpeg
-```
-
-#### MacOS 사용자
-
-```bash
-brew install ffmpeg
-```
-
-#### Windows 사용자
-
-다운로드 후 루트 디렉토리에 배치.
-
-- [ffmpeg.exe 다운로드](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffmpeg.exe)
-
-- [ffprobe.exe 다운로드](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffprobe.exe)
-
-### 3. RMVPE 인간 음성 피치 추출 알고리즘에 필요한 파일 다운로드
-
-최신 RMVPE 인간 음성 피치 추출 알고리즘을 사용하려면 음피치 추출 모델 매개변수를 다운로드하고 RVC 루트 디렉토리에 배치해야 합니다.
-
-- [rmvpe.pt 다운로드](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.pt)
-
-#### dml 환경의 RMVPE 다운로드(선택사항, A카드/I카드 사용자)
-
-- [rmvpe.onnx 다운로드](https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.onnx)
-
-### 4. AMD 그래픽 카드 Rocm(선택사항, Linux만 해당)
-
-Linux 시스템에서 AMD의 Rocm 기술을 기반으로 RVC를 실행하려면 [여기](https://rocm.docs.amd.com/en/latest/deploy/linux/os-native/install.html)에서 필요한 드라이버를 먼저 설치하세요.
-
-Arch Linux를 사용하는 경우 pacman을 사용하여 필요한 드라이버를 설치할 수 있습니다.
-
-```
-pacman -S rocm-hip-sdk rocm-opencl-sdk
-```
-
-일부 모델의 그래픽 카드(예: RX6700XT)의 경우, 다음과 같은 환경 변수를 추가로 설정해야 할 수 있습니다.
-
-```
-export ROCM_PATH=/opt/rocm
-export HSA_OVERRIDE_GFX_VERSION=10.3.0
-```
-
-동시에 현재 사용자가 `render` 및 `video` 사용자 그룹에 속해 있는지 확인하세요.
-
-```
-sudo usermod -aG render $USERNAME
-sudo usermod -aG video $USERNAME
-```
-
-## 시작하기
-
-### 직접 시작
-
-다음 명령어로 WebUI를 시작하세요
+## WebUI 시작
 
 ```bash
 python webui.py
 ```
 
-### 통합 패키지 사용
-
-`RVC-beta.7z`를 다운로드하고 압축 해제
-
-#### Windows 사용자
-
-`go-webui.bat` 더블 클릭
-
-#### MacOS 사용자
+화면이 없는 Ubuntu 서버:
 
 ```bash
-sh ./run.sh
+python webui.py --noautoopen
 ```
 
+기본 포트는 `7865`입니다. `.pth` 모델은 `assets/weights/`, `.index` 파일은 `assets/indices/`에 배치하세요.
+
 ## 참조 프로젝트
 
 - [ContentVec](https://github.com/auspicious3000/contentvec/)
diff --git a/docs/pt/README.pt.md b/docs/pt/README.pt.md
index 42d8fd0..5691879 100644
--- a/docs/pt/README.pt.md
+++ b/docs/pt/README.pt.md
@@ -1,25 +1,23 @@
 <div align="center">
 
 <h1>Retrieval-based-Voice-Conversion-WebUI</h1>
-Uma estrutura de conversão de voz fácil de usar baseada em VITS.<br><br>
+Um framework simples e fácil de usar para conversão de timbre vocal / alteração de voz.<br><br>
 
 [![madewithlove](https://img.shields.io/badge/made_with-%E2%9D%A4-red?style=for-the-badge&labelColor=orange
 )](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)
 
 <img src="https://counter.seku.su/cmoe?name=rvc&theme=r34" /><br>
-  
-[![Open In Colab](https://img.shields.io/badge/Colab-F9AB00?style=for-the-badge&logo=googlecolab&color=525252)](https://colab.research.google.com/github/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/Retrieval_based_Voice_Conversion_WebUI.ipynb)
-[![Licence](https://img.shields.io/github/license/RVC-Project/Retrieval-based-Voice-Conversion-WebUI?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
-[![Huggingface](https://img.shields.io/badge/🤗%20-Spaces-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
 
-[![Discord](https://img.shields.io/badge/RVC%20Developers-Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/HcsmBBGyVk)
+[![Licence](https://img.shields.io/github/license/RVC-Project/Retrieval-based-Voice-Conversion-WebUI?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
+[![Huggingface](https://img.shields.io/badge/🤗%20-Models-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
+
 
 </div>
 
 ------
-[**Changelog**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/docs/Changelog_EN.md) | [**FAQ (Frequently Asked Questions)**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/FAQ-(Frequently-Asked-Questions)) 
+[**Changelog**](./Changelog_pt.md) | [**FAQ (Frequently Asked Questions)**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/FAQ-(Frequently-Asked-Questions))
 
-[**English**](../en/README.en.md) | [**中文简体**](../../README.md) | [**日本語**](../jp/README.ja.md) | [**한국어**](../kr/README.ko.md) ([**韓國語**](../kr/README.ko.han.md)) | [**Türkçe**](../tr/README.tr.md) | [**Português**](../pt/README.pt.md)
+[**English**](../en/README.en.md) | [**中文简体**](../../README.md) | [**日本語**](../jp/README.ja.md) | [**한국어**](../kr/README.ko.md) ([**韓國語**](../kr/README.ko.han.md)) | [**Français**](../fr/README.fr.md) | [**Türkçe**](../tr/README.tr.md) | [**Português**](../pt/README.pt.md)
 
 
 Confira nosso [Vídeo de demonstração](https://www.bilibili.com/video/BV1pm4y1z7Gm/) aqui!
@@ -46,124 +44,169 @@ Este repositório possui os seguintes recursos:
 + Interface Webui fácil de usar;
 + Use o modelo UVR5 para separar rapidamente vocais e instrumentos.
 + Use o mais poderoso algoritmo de extração de voz de alta frequência [InterSpeech2023-RMVPE](#Credits) para evitar o problema de som mudo. Fornece os melhores resultados (significativamente) e é mais rápido, com consumo de recursos ainda menor que o Crepe_full.
-+ Suporta aceleração de placas gráficas AMD/Intel.
++ Sistemas AMD/Intel usam as dependências de CPU; Windows pode usar DirectML e Linux usa CPU.
 
 ## Preparando o ambiente
-Os comandos a seguir precisam ser executados no ambiente Python versão 3.8 ou superior.
 
-(Windows/Linux)
-Primeiro instale as dependências principais através do pip:
+Esta branch é destinada a **Python 3.12 x64**. Execute todos os comandos na raiz do repositório. Ubuntu 24.04 x86_64 é recomendado.
+
+### Ubuntu 24.04
+
 ```bash
-# Instale as dependências principais relacionadas ao PyTorch, pule se instaladas
-# Referência: https://pytorch.org/get-started/locally/
-pip install torch torchvision torchaudio
+sudo apt update
+sudo apt install -y python3.12 python3.12-venv python3.12-dev ffmpeg unzip libsndfile1 libportaudio2
 
-#Para arquitetura Windows + Nvidia Ampere (RTX30xx), você precisa especificar a versão cuda correspondente ao pytorch de acordo com a experiência de https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/ 21
-#pip instalar tocha torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
-
-#Para placas Linux + AMD, você precisa usar as seguintes versões do pytorch:
-#pip instalar tocha torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm5.4.2
+python3.12 -m venv .venv
+source .venv/bin/activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-Então pode usar poesia para instalar as outras dependências:
-```bash
-# Instale a ferramenta de gerenciamento de dependências Poetry, pule se instalada
-# Referência: https://python-poetry.org/docs/#installation
-curl -sSL https://install.python-poetry.org | python3 -
+### Windows
 
-#Instale as dependências do projeto
-poetry install
+Instale o Python 3.12 x64 e crie um ambiente virtual:
+
+```powershell
+py -3.12 -m venv .venv
+.venv\Scripts\activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-Você também pode usar pip para instalá-los:
+### Escolha de dependências por hardware
+
+| Hardware | Instalação |
+| --- | --- |
+| CPU, AMD, Intel | Use `requirments_cpu_py312.txt`; Windows pode usar DirectML, Linux usa CPU |
+| NVIDIA RTX série 50 | Instale primeiro o Torch CUDA 12.8 e depois `requirments_cu128_py312.txt` |
+| NVIDIA anterior à série RTX 50 | Instale primeiro o Torch CUDA 11.8 e depois `requirments_cu118_py312.txt` |
+
+#### CPU, AMD, Intel
+
 ```bash
-
-for Nvidia graphics cards
-  pip install -r requirements.txt
-
-for AMD/Intel graphics cards on Windows (DirectML)：
-  pip install -r requirements-dml.txt
-
-for AMD graphics cards on Linux (ROCm):
-  pip install -r requirements-amd.txt
+python -m pip install -r requirments_cpu_py312.txt
 ```
 
-------
-Usuários de Mac podem instalar dependências via `run.sh`:
+#### NVIDIA RTX série 50: duas etapas
+
 ```bash
-sh ./run.sh
+python -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 \
+  --index-url https://download.pytorch.org/whl/cu128 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu128_py312.txt
 ```
 
-## Preparação de outros Pré-modelos
-RVC requer outros pré-modelos para inferir e treinar.
+#### NVIDIA anterior à série RTX 50: duas etapas
 
-Baixe-os em nosso [Huggingface space](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/).
-
-Aqui está uma lista de pré-modelos e outros arquivos que o RVC precisa:
 ```bash
-./assets/hubert_base
-
-./assets/pretrained 
-
-./assets/uvr5_weights
-
-Downloads adicionais são necessários se você quiser testar a versão v2 do modelo.
-
-./assets/pretrained_v2
-
-Se você deseja testar o modelo da versão v2 (o modelo da versão v2 alterou a entrada do recurso dimensional 256 do Hubert + final_proj de 9 camadas para o recurso dimensional 768 do Hubert de 12 camadas e adicionou 3 discriminadores de período), você precisará baixar recursos adicionais
-
-./assets/pretrained_v2
-
-#Se você estiver usando Windows, também pode precisar desses dois arquivos, pule se FFmpeg e FFprobe estiverem instalados
-ffmpeg.exe
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffmpeg.exe
-
-ffprobe.exe
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffprobe.exe
-
-Se quiser usar o algoritmo de extração de tom vocal SOTA RMVPE mais recente, você precisa baixar os pesos RMVPE e colocá-los no diretório raiz RVC
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.pt
-
-    Para usuários de placas gráficas AMD/Intel, você precisa baixar:
-
-    https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.onnx
-
+python -m pip install torch==2.7.1+cu118 torchaudio==2.7.1+cu118 \
+  --index-url https://download.pytorch.org/whl/cu118 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu118_py312.txt
 ```
 
-Em seguida, use este comando para iniciar o Webui:
+Verifique Torch e CUDA:
+
+```bash
+python -c "import torch; print('torch:', torch.__version__); print('cuda:', torch.version.cuda); print('cuda available:', torch.cuda.is_available())"
+```
+
+O programa também verifica a memória e a capacidade de computação da GPU NVIDIA. GPUs com aproximadamente menos de 4 GiB ou abaixo de SM 5.3 usam CPU.
+
+### Fontes de pacotes
+
+Os três arquivos `requirments_*.txt` definem suas fontes no topo. Para usar fontes oficiais, substitua somente `--index-url` e `--extra-index-url`, mantendo versões, sufixos CUDA e a ordem das duas etapas.
+
+| Default mirror | Official source |
+| --- | --- |
+| `https://mirrors.pku.edu.cn/pypi/simple` | `https://pypi.org/simple` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cpu` | `https://download.pytorch.org/whl/cpu` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu118` | `https://download.pytorch.org/whl/cu118` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu128` | `https://download.pytorch.org/whl/cu128` |
+
+## Modelos e diretórios de execução
+
+O WebUI cria os diretórios de execução automaticamente. Baixe os modelos do [repositório de modelos Hugging Face](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main) e mantenha esta estrutura:
+
+```text
+assets/
+├── hubert_base/
+│   ├── config.json
+│   ├── preprocessor_config.json
+│   └── pytorch_model.bin
+├── rmvpe/rmvpe.pt
+├── pretrained/
+├── pretrained_v2/
+├── uvr5_weights/
+├── weights/        # user RVC .pth models
+└── indices/        # user .index files
+logs/
+└── mute/           # training silence samples
+
+# Exact paths used by the code
+assets/hubert_base/config.json
+assets/hubert_base/preprocessor_config.json
+assets/hubert_base/pytorch_model.bin
+assets/rmvpe/rmvpe.pt
+assets/pretrained/*.pth
+assets/pretrained_v2/*.pth
+assets/uvr5_weights/*
+assets/weights/*.pth
+assets/indices/*.index
+logs/mute/*
+```
+
+### Baixar modelos
+
+```bash
+python -m pip install --upgrade huggingface_hub
+
+# Required for inference and feature extraction
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "hubert_base/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI rmvpe.pt --revision main \
+  --local-dir assets/rmvpe
+
+# Required for v1/v2 training
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "pretrained/*" "pretrained_v2/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI mute.zip --revision main \
+  --local-dir .model-downloads
+python -m zipfile -e .model-downloads/mute.zip logs
+
+# Required only for UVR5 vocal separation
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "uvr5_weights/*" --local-dir assets
+```
+
+Ambientes Windows AMD/Intel DirectML também precisam de:
+
+```bash
+hf download lj1995/VoiceConversionWebUI rmvpe.onnx --revision main \
+  --local-dir assets/rmvpe
+```
+
+O arquivo antigo `hubert_base.pt` não é usado nesta branch. O código atual usa o modelo Transformers em `assets/hubert_base/`. O FCPE está incluído no `torchfcpe`.
+
+### FFmpeg
+
+O comando Ubuntu acima instala o FFmpeg. No Windows, coloque estes arquivos na raiz do repositório:
+
+- [ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffmpeg.exe?download=true)
+- [ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffprobe.exe?download=true)
+
+## Iniciar o WebUI
+
 ```bash
 python webui.py
 ```
 
-Se estiver usando Windows ou macOS, você pode baixar e extrair `RVC-beta.7z` para usar RVC diretamente usando `go-webui.bat` no Windows ou `sh ./run.sh` no macOS para iniciar o Webui.
+Servidor Ubuntu sem interface gráfica:
 
-## Suporte ROCm para placas gráficas AMD (somente Linux)
-Para usar o ROCm no Linux, instale todos os drivers necessários conforme descrito [aqui](https://rocm.docs.amd.com/en/latest/deploy/linux/os-native/install.html).
-
-No Arch use pacman para instalar o driver:
-````
-pacman -S rocm-hip-sdk rocm-opencl-sdk
-````
-
-Talvez você também precise definir estas variáveis de ambiente (por exemplo, em um RX6700XT):
-````
-export ROCM_PATH=/opt/rocm
-export HSA_OVERRIDE_GFX_VERSION=10.3.0
-````
-Verifique também se seu usuário faz parte do grupo `render` e `video`:
-````
-sudo usermod -aG render $USERNAME
-sudo usermod -aG video $USERNAME
-````
-Depois disso, você pode executar o WebUI:
 ```bash
-python webui.py
+python webui.py --noautoopen
 ```
 
+A porta padrão é `7865`. Coloque modelos `.pth` em `assets/weights/` e arquivos `.index` em `assets/indices/`.
+
 ## Credits
 + [ContentVec](https://github.com/auspicious3000/contentvec/)
 + [VITS](https://github.com/jaywalnut310/vits)
@@ -174,7 +217,7 @@ python webui.py
 + [audio-slicer](https://github.com/openvpi/audio-slicer)
 + [Vocal pitch extraction:RMVPE](https://github.com/Dream-High/RMVPE)
   + The pretrained model is trained and tested by [yxlllc](https://github.com/yxlllc/RMVPE) and [RVC-Boss](https://github.com/RVC-Boss).
-  
+
 ## Thanks to all contributors for their efforts
 <a href="https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/graphs/contributors" target="_blank">
   <img src="https://contrib.rocks/image?repo=RVC-Project/Retrieval-based-Voice-Conversion-WebUI" />
diff --git a/docs/tr/README.tr.md b/docs/tr/README.tr.md
index b8829ef..84fd949 100644
--- a/docs/tr/README.tr.md
+++ b/docs/tr/README.tr.md
@@ -2,23 +2,21 @@
 <div align="center">
 
 <h1>Çekme Temelli Ses Dönüşümü Web Arayüzü</h1>
-VITS'e dayalı kullanımı kolay bir Ses Dönüşümü çerçevesi.<br><br>
+Basit ve kullanımı kolay bir ses tınısı dönüştürme / ses değiştirici çerçevesi.<br><br>
 
 [![madewithlove](https://img.shields.io/badge/made_with-%E2%9D%A4-red?style=for-the-badge&labelColor=orange
 )](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI)
 
 <img src="https://counter.seku.su/cmoe?name=rvc&theme=r34" /><br>
-  
-[![Open In Colab](https://img.shields.io/badge/Colab-F9AB00?style=for-the-badge&logo=googlecolab&color=525252)](https://colab.research.google.com/github/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/Retrieval_based_Voice_Conversion_WebUI.ipynb)
-[![Lisans](https://img.shields.io/github/license/RVC-Project/Retrieval-based-Voice-Conversion-WebUI?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
-[![Huggingface](https://img.shields.io/badge/🤗%20-Spaces-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
 
-[![Discord](https://img.shields.io/badge/RVC%20Geliştiricileri-Discord-7289DA?style=for-the-badge&logo=discord&logoColor=white)](https://discord.gg/HcsmBBGyVk)
+[![Lisans](https://img.shields.io/github/license/RVC-Project/Retrieval-based-Voice-Conversion-WebUI?style=for-the-badge)](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/LICENSE)
+[![Huggingface](https://img.shields.io/badge/🤗%20-Models-yellow.svg?style=for-the-badge)](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/)
+
 
 </div>
 
 ------
-[**Değişiklik Geçmişi**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/docs/Changelog_TR.md) | [**SSS (Sıkça Sorulan Sorular)**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/SSS-(Sıkça-Sorulan-Sorular)) 
+[**Değişiklik Geçmişi**](./Changelog_TR.md) | [**SSS (Sıkça Sorulan Sorular)**](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/wiki/SSS-(Sıkça-Sorulan-Sorular))
 
 [**İngilizce**](../en/README.en.md) | [**中文简体**](../../README.md) | [**日本語**](../jp/README.ja.md) | [**한국어**](../kr/README.ko.md) ([**韓國語**](../kr/README.ko.han.md)) | [**Français**](../fr/README.fr.md) | [**Türkçe**](../tr/README.tr.md) | [**Português**](../pt/README.pt.md)
 
@@ -41,94 +39,168 @@ Bu depo aşağıdaki özelliklere sahiptir:
 + Kullanımı kolay Web arayüzü;
 + UVR5 modelini kullanarak hızla vokalleri ve enstrümanları ayırma.
 + En güçlü Yüksek tiz Ses Çıkarma Algoritması [InterSpeech2023-RMVPE](#Krediler) sessiz ses sorununu önlemek için kullanılır. En iyi sonuçları (önemli ölçüde) sağlar ve Crepe_full'den daha hızlı çalışır, hatta daha düşük kaynak tüketimi sağlar.
-+ AMD/Intel grafik kartları hızlandırması desteklenir.
++ AMD/Intel sistemleri CPU bağımlılıklarını kullanır; Windows DirectML, Linux CPU kullanabilir.
 
 ## Ortamın Hazırlanması
-Aşağıdaki komutlar, Python sürümü 3.8 veya daha yüksek olan bir ortamda çalıştırılmalıdır.
 
-(Windows/Linux)
-İlk olarak ana bağımlılıkları pip aracılığıyla kurun:
+Bu dal **Python 3.12 x64** için hazırlanmıştır. Tüm komutları depo kökünde çalıştırın. Ubuntu 24.04 x86_64 önerilir.
+
+### Ubuntu 24.04
+
 ```bash
-# PyTorch ile ilgili temel bağımlılıkları kurun, zaten kuruluysa atlayın
-# Referans: https://pytorch.org/get-started/locally/
-pip install torch torchvision torchaudio
+sudo apt update
+sudo apt install -y python3.12 python3.12-venv python3.12-dev ffmpeg unzip libsndfile1 libportaudio2
 
-# Windows + Nvidia Ampere Mimarisi(RTX30xx) için, https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/21 deneyime göre pytorch'a karşılık gelen cuda sürümünü belirtmeniz gerekebilir
-#pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
+python3.12 -m venv .venv
+source .venv/bin/activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-Sonra poetry kullanarak diğer bağımlılıkları kurabilirsiniz:
-```bash
-# Poetry bağımlılık yönetim aracını kurun, zaten kuruluysa atlayın
-# Referans: https://python-poetry.org/docs/#installation
-curl -sSL https://install.python-poetry.org | python3 -
+### Windows
 
-# Projeyi bağımlılıkları kurun
-poetry install
+Python 3.12 x64 kurduktan sonra sanal ortam oluşturun:
+
+```powershell
+py -3.12 -m venv .venv
+.venv\Scripts\activate
+python -m pip install --upgrade pip setuptools wheel
 ```
 
-Ayrıca bunları pip kullanarak da kurabilirsiniz:
+### Donanıma göre bağımlılık seçimi
+
+| Donanım | Kurulum |
+| --- | --- |
+| CPU, AMD, Intel | `requirments_cpu_py312.txt` kullanın; Windows DirectML, Linux CPU kullanabilir |
+| NVIDIA RTX 50 serisi | Önce CUDA 12.8 Torch, ardından `requirments_cu128_py312.txt` |
+| RTX 50 serisinden önceki NVIDIA | Önce CUDA 11.8 Torch, ardından `requirments_cu118_py312.txt` |
+
+#### CPU, AMD, Intel
+
 ```bash
-
-Nvidia grafik kartları için
-  pip install -r requirements.txt
-
-AMD/Intel grafik kartları için：
-  pip install -r requirements-dml.txt
-
+python -m pip install -r requirments_cpu_py312.txt
 ```
 
-------
-Mac kullanıcıları `run.sh` aracılığıyla bağımlılıkları kurabilir:
+#### NVIDIA RTX 50 serisi: iki aşama
+
 ```bash
-sh ./run.sh
+python -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 \
+  --index-url https://download.pytorch.org/whl/cu128 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu128_py312.txt
 ```
 
-## Diğer Ön Modellerin Hazırlanması
-RVC'nin çıkarım ve eğitim yapması için diğer ön modellere ihtiyacı vardır.
+#### RTX 50 serisinden önceki NVIDIA: iki aşama
 
-Bu ön modelleri [Huggingface alanımızdan](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main/) indirmeniz gerekecektir.
-
-İşte RVC'nin ihtiyaç duyduğu diğer ön modellerin ve dosyaların bir listesi:
 ```bash
-./assets/hubert_base
-
-./assets/pretrained 
-
-./assets/uvr5_weights
-
-V2 sürümü modelini test etmek isterseniz, ek özellikler indirmeniz gerekecektir.
-
-./assets/pretrained_v2
-
-V2 sürüm modelini test etmek isterseniz (v2 sürüm modeli, 9 katmanlı Hubert+final_proj'ün 256 boyutlu özelliğini 12 katmanlı Hubert'ün 768 boyutlu özelliğiyle değiştirmiştir ve 3 periyot ayırıcı eklemiştir), ek özellikleri indirmeniz gerekecektir.
-
-./assets/pretrained_v2
-
-Eğer Windows kullanıyorsanız, FFmpeg ve FFprobe kurulu değilse bu iki dosyayı da indirmeniz gerekebilir.
-ffmpeg.exe
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffmpeg.exe
-
-ffprobe.exe
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/ffprobe.exe
-
-En son SOTA RMVPE vokal ton çıkarma algoritmasını kullanmak istiyorsanız, RMVPE ağırlıklarını indirip RVC kök dizinine koymalısınız.
-
-https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.pt
-
-    AMD/Intel grafik kartları kullanıcıları için indirmeniz gereken:
-
-    https://huggingface.co/lj1995/VoiceConversionWebUI/blob/main/rmvpe.onnx
-
+python -m pip install torch==2.7.1+cu118 torchaudio==2.7.1+cu118 \
+  --index-url https://download.pytorch.org/whl/cu118 \
+  --extra-index-url https://pypi.org/simple
+python -m pip install -r requirments_cu118_py312.txt
 ```
 
-Daha sonra bu komutu kullanarak Webui'yi başlatabilirsiniz:
+Torch ve CUDA durumunu doğrulayın:
+
+```bash
+python -c "import torch; print('torch:', torch.__version__); print('cuda:', torch.version.cuda); print('cuda available:', torch.cuda.is_available())"
+```
+
+Program NVIDIA GPU belleğini ve hesaplama yeteneğini de denetler. Yaklaşık 4 GiB altındaki veya SM 5.3 altındaki kartlar CPU yolunu kullanır.
+
+### Paket kaynakları
+
+Üç `requirments_*.txt` dosyasının başında paket kaynakları yer alır. Resmî kaynakları kullanmak için yalnızca `--index-url` ve `--extra-index-url` satırlarını değiştirin; sürümleri, CUDA eklerini ve iki aşamalı sırayı koruyun.
+
+| Default mirror | Official source |
+| --- | --- |
+| `https://mirrors.pku.edu.cn/pypi/simple` | `https://pypi.org/simple` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cpu` | `https://download.pytorch.org/whl/cpu` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu118` | `https://download.pytorch.org/whl/cu118` |
+| `https://mirrors.nju.edu.cn/pytorch/whl/cu128` | `https://download.pytorch.org/whl/cu128` |
+
+## Modeller ve çalışma dizinleri
+
+WebUI çalışma dizinlerini otomatik oluşturur. Modelleri [Hugging Face model deposundan](https://huggingface.co/lj1995/VoiceConversionWebUI/tree/main) indirin ve şu yapıyı koruyun:
+
+```text
+assets/
+├── hubert_base/
+│   ├── config.json
+│   ├── preprocessor_config.json
+│   └── pytorch_model.bin
+├── rmvpe/rmvpe.pt
+├── pretrained/
+├── pretrained_v2/
+├── uvr5_weights/
+├── weights/        # user RVC .pth models
+└── indices/        # user .index files
+logs/
+└── mute/           # training silence samples
+
+# Exact paths used by the code
+assets/hubert_base/config.json
+assets/hubert_base/preprocessor_config.json
+assets/hubert_base/pytorch_model.bin
+assets/rmvpe/rmvpe.pt
+assets/pretrained/*.pth
+assets/pretrained_v2/*.pth
+assets/uvr5_weights/*
+assets/weights/*.pth
+assets/indices/*.index
+logs/mute/*
+```
+
+### Modelleri indirme
+
+```bash
+python -m pip install --upgrade huggingface_hub
+
+# Required for inference and feature extraction
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "hubert_base/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI rmvpe.pt --revision main \
+  --local-dir assets/rmvpe
+
+# Required for v1/v2 training
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "pretrained/*" "pretrained_v2/*" --local-dir assets
+hf download lj1995/VoiceConversionWebUI mute.zip --revision main \
+  --local-dir .model-downloads
+python -m zipfile -e .model-downloads/mute.zip logs
+
+# Required only for UVR5 vocal separation
+hf download lj1995/VoiceConversionWebUI --revision main \
+  --include "uvr5_weights/*" --local-dir assets
+```
+
+Windows AMD/Intel DirectML ortamlarında ayrıca şu dosya gerekir:
+
+```bash
+hf download lj1995/VoiceConversionWebUI rmvpe.onnx --revision main \
+  --local-dir assets/rmvpe
+```
+
+Eski `hubert_base.pt` dosyası bu dalda kullanılmaz. Güncel kod `assets/hubert_base/` altındaki Transformers modelini kullanır. FCPE modeli `torchfcpe` paketine dahildir.
+
+### FFmpeg
+
+Yukarıdaki Ubuntu komutu FFmpeg'i kurar. Windows'ta şu dosyaları depo köküne yerleştirin:
+
+- [ffmpeg.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffmpeg.exe?download=true)
+- [ffprobe.exe](https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/ffprobe.exe?download=true)
+
+## WebUI'yi başlatma
+
 ```bash
 python webui.py
 ```
-Windows veya macOS kullanıyorsanız, `RVC-beta.7z` dosyasını indirip çıkararak `go-webui.bat`i kullanarak veya macOS'ta `sh ./run.sh` kullanarak doğrudan RVC'yi kullanabilirsiniz.
+
+Grafik arayüzü olmayan Ubuntu sunucusu:
+
+```bash
+python webui.py --noautoopen
+```
+
+Varsayılan bağlantı noktası `7865`'tir. `.pth` modellerini `assets/weights/`, `.index` dosyalarını `assets/indices/` içine yerleştirin.
 
 ## Krediler
 + [ContentVec](https://github.com/auspicious3000/contentvec/)
@@ -140,9 +212,8 @@ Windows veya macOS kullanıyorsanız, `RVC-beta.7z` dosyasını indirip çıkara
 + [audio-slicer](https://github.com/openvpi/audio-slicer)
 + [Vokal ton çıkarma:RMVPE](https://github.com/Dream-High/RMVPE)
   + Ön eğitimli model [yxlllc](https://github.com/yxlllc/RMVPE) ve [RVC-Boss](https://github.com/RVC-Boss) tarafından eğitilip test edilmiştir.
-  
+
 ## Katkıda Bulunan Herkese Teşekkürler
 <a href="https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/graphs/contributors" target="_blank">
   <img src="https://contrib.rocks/image?repo=RVC-Project/Retrieval-based-Voice-Conversion-WebUI" />
 </a>
-```
diff --git a/requirments_cpu_py312.txt b/requirments_cpu_py312.txt
index 8dc5b54..30735ce 100644
--- a/requirments_cpu_py312.txt
+++ b/requirments_cpu_py312.txt
@@ -1,11 +1,12 @@
-# Python 3.12 x64 / Windows / CPU + DirectML
+# Python 3.12 x64 / Windows + Linux / CPU
 #
 # Install the complete environment in one stage from the project root:
-#   runtime\python.exe -I -m pip install -r requirments_cpu_py312.txt
+#   python -m pip install -r requirments_cpu_py312.txt
+# Windows portable runtime users may replace `python` with `runtime\python.exe -I`.
 #
 # torch-directml 0.2.5 is built against the PyTorch 2.4.1 family.  These exact
-# CPU versions are therefore intentional; they also ensure that no CUDA Torch
-# binaries are introduced into this runtime.
+# CPU versions are therefore intentional. DirectML is installed only on Windows;
+# Linux AMD/Intel systems use the CPU execution path.
 
 --index-url https://mirrors.pku.edu.cn/pypi/simple
 --extra-index-url https://mirrors.nju.edu.cn/pytorch/whl/cpu
@@ -14,7 +15,7 @@
 torch==2.4.1+cpu
 torchaudio==2.4.1+cpu
 torchvision==0.19.1+cpu
-torch-directml==0.2.5.dev240914
+torch-directml==0.2.5.dev240914; platform_system == "Windows"
 
 # Packaging support. Gradio 3.14 still imports pkg_resources, so Setuptools
 # must remain on the last generation that provides it.
@@ -38,9 +39,10 @@ matplotlib>=3.8.2,<4
 networkx>=3.2.0,<4
 numpy>=1.26.4,<2
 
-# DirectML provider for RMVPE and ONNX UVR inference.  ORT variants share the
-# same Python module, so this environment contains only the DirectML package.
-onnxruntime-directml>=1.24.4,<2
+# ONNX Runtime providers are platform-specific because all variants expose the
+# same Python module. Windows uses DirectML; Linux uses the CPU provider.
+onnxruntime-directml>=1.24.4,<2; platform_system == "Windows"
+onnxruntime>=1.24.4,<2; platform_system != "Windows"
 coloredlogs>=15.0,<16
 
 opencv-python-headless>=4.10.0,<5
diff --git a/requirments_cu118_py312.txt b/requirments_cu118_py312.txt
index 5dd8356..785aa5c 100644
--- a/requirments_cu118_py312.txt
+++ b/requirments_cu118_py312.txt
@@ -1,13 +1,14 @@
-# Python 3.12 x64 / Windows / NVIDIA CUDA 11.8
+# Python 3.12 x64 / Windows + Linux / NVIDIA CUDA 11.8
 #
 # Install in two stages from the project root. Torch and Torchaudio are NOT
 # listed below, so the second command cannot replace the verified cu118 pair.
 #
 # Stage 1 - install the matching CUDA 11.8 pair (Torch must stay below 2.8):
-#   runtime\python.exe -I -m pip install torch==2.7.1+cu118 torchaudio==2.7.1+cu118 --index-url https://mirrors.nju.edu.cn/pytorch/whl/cu118 --extra-index-url https://mirrors.pku.edu.cn/pypi/simple
+#   python -m pip install torch==2.7.1+cu118 torchaudio==2.7.1+cu118 --index-url https://mirrors.nju.edu.cn/pytorch/whl/cu118 --extra-index-url https://mirrors.pku.edu.cn/pypi/simple
 #
 # Stage 2 - install all remaining project dependencies:
-#   runtime\python.exe -I -m pip install -r requirments_cu118_py312.txt
+#   python -m pip install -r requirments_cu118_py312.txt
+# Windows portable runtime users may replace `python` with `runtime\python.exe -I`.
 
 --index-url https://mirrors.pku.edu.cn/pypi/simple
 
diff --git a/requirments_cu128_py312.txt b/requirments_cu128_py312.txt
index b8382ac..2aa873f 100644
--- a/requirments_cu128_py312.txt
+++ b/requirments_cu128_py312.txt
@@ -1,13 +1,14 @@
-# Python 3.12 x64 / Windows / NVIDIA CUDA 12.8
+# Python 3.12 x64 / Windows + Linux / NVIDIA CUDA 12.8
 #
 # Install in two stages from the project root. Torch and Torchaudio are NOT
 # listed below, so the second command cannot replace the verified cu128 pair.
 #
 # Stage 1 - install the matching CUDA 12.8 pair (Torch must stay below 2.8):
-#   runtime\python.exe -I -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 --index-url https://mirrors.nju.edu.cn/pytorch/whl/cu128 --extra-index-url https://mirrors.pku.edu.cn/pypi/simple
+#   python -m pip install torch==2.7.1+cu128 torchaudio==2.7.1+cu128 --index-url https://mirrors.nju.edu.cn/pytorch/whl/cu128 --extra-index-url https://mirrors.pku.edu.cn/pypi/simple
 #
 # Stage 2 - install all remaining project dependencies:
-#   runtime\python.exe -I -m pip install -r requirments_cu128_py312.txt
+#   python -m pip install -r requirments_cu128_py312.txt
+# Windows portable runtime users may replace `python` with `runtime\python.exe -I`.
 
 --index-url https://mirrors.pku.edu.cn/pypi/simple
 
diff --git a/webui.py b/webui.py
index eee285a..95ec997 100644
--- a/webui.py
+++ b/webui.py
@@ -116,8 +116,19 @@ def launch_webui_with_port_fallback(app, config):
             )
             next_port = config.listen_port + 1
 
-os.makedirs(os.path.join(now_dir, "logs"), exist_ok=True)
-os.makedirs(os.path.join(now_dir, "assets/weights"), exist_ok=True)
+runtime_dirs = (
+    os.path.join(now_dir, "logs"),
+    os.environ["weight_root"],
+    os.environ["weight_uvr5_root"],
+    os.environ["index_root"],
+    os.environ["outside_index_root"],
+    os.environ["rmvpe_root"],
+    os.path.join(now_dir, "assets", "hubert_base"),
+    os.path.join(now_dir, "assets", "pretrained"),
+    os.path.join(now_dir, "assets", "pretrained_v2"),
+)
+for runtime_dir in runtime_dirs:
+    os.makedirs(runtime_dir, exist_ok=True)
 warnings.filterwarnings("ignore")
 torch.manual_seed(114514)