返回模型

说明文档

<div align="center">
<img alt="LOGO" src="https://huggingface.co/AnhP/Vietnamese-RVC-Project/resolve/main/ico.png" width="300" height="300" />

# Vietnamese RVC BY ANH
专为越南人设计的高质量、高性能语音转换工具。

[![Vietnamese RVC](https://img.shields.io/badge/GitHub-100000?style=for-the-badge&logo=github&logoColor=white)](https://github.com/PhamHuynhAnh16/Vietnamese-RVC)
[![Open In Colab](https://img.shields.io/badge/Colab-F9AB00?style=for-the-badge&logo=googlecolab&color=525252)](https://colab.research.google.com/github/PhamHuynhAnh16/Vietnamese-RVC-ipynb/blob/main/Vietnamese-RVC.ipynb)
[![Licence](https://img.shields.io/github/license/saltstack/salt?style=for-the-badge)](https://github.com/PhamHuynhAnh16/Vietnamese-RVC/blob/main/LICENSE)

</div>

<div align="center">

[![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/AnhP/RVC-GUI)
[![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97-Models-blue)](https://huggingface.co/AnhP/Vietnamese-RVC-Project)

</div>

# 项目描述
本项目是一个简单易用的语音转换工具，专为越南人设计。旨在打造高质量、性能最优的语音转换产品，让用户能够流畅、自然地改变声音。

# 项目功能

- 音频分离 (MDX-Net/Demucs)

- 语音转换 (文件转换/批量转换/Whisper转换/文本转换)

- 背景音乐编辑

- 为音频应用效果

- 生成训练数据 (从链接路径)

- 训练模型 (v1/v2, 高质量编码器)

- 模型融合

- 读取模型信息

- 将模型导出为ONNX

- 从现有模型库下载

- 从网络搜索模型

- 音高提取

- 支持使用ONNX模型进行音频转换推理

- RVC ONNX模型也支持索引用于推理

- 多种模型选项:

F0: `pm, dio, mangio-crepe-tiny, mangio-crepe-small, mangio-crepe-medium, mangio-crepe-large, mangio-crepe-full, crepe-tiny, crepe-small, crepe-medium, crepe-large, crepe-full, fcpe, fcpe-legacy, rmvpe, rmvpe-legacy, harvest, yin, pyin, swipe`

F0_ONNX: 部分模型已转换为ONNX以支持加速提取

F0_HYBRID: 可以将多个选项组合在一起，如 `hybrid[rmvpe+harvest]`，或者尝试将所有选项组合在一起

EMBEDDERS: `contentvec_base, hubert_base, japanese_hubert_base, korean_hubert_base, chinese_hubert_base, portuguese_hubert_base`

EMBEDDERS_ONNX: 上述所有嵌入模型都有已转换为ONNX的版本，用于加速嵌入提取

EMBEDDERS_TRANSFORMERS: 上述所有嵌入模型都有已转换为HuggingFace的版本，可替代fairseq使用

SPIN_EMBEDDERS: 一种新的嵌入提取模型，可能比旧版提取器带来更高质量。

# 使用指南

**如果我真正有空的话会有...**

# 安装和使用方法

- 第1步: **从官网安装python或 [python](https://www.python.org/ftp/python/3.10.7/python-3.10.7-amd64.exe) (需要 PYTHON 3.10.x 或 PYTHON 3.11.x)**
- 第2步: **从 [FFMPEG](https://github.com/BtbN/FFmpeg-Builds/releases) 安装ffmpeg，解压并添加到PATH**
- 第3步: **下载源码并解压**
- 第4步: **进入源码目录，打开命令提示符或终端**
- 第5步: **输入命令安装运行所需的依赖库**

python -m venv env env\Scripts\activate


如果有NVIDIA GPU，请根据你的cuda版本运行此步骤，可以将cu117改为cu128等...

如果使用 Torch 2.3.1

python -m pip install torch==2.3.1 torchaudio==2.3.1 torchvision==0.18.1 --index-url https://download.pytorch.org/whl/cu117

如果使用 Torch 2.6.0

python -m pip install torch==2.6.0 torchaudio==2.6.0 torchvision==0.21.0 --index-url https://download.pytorch.org/whl/cu117


接下来运行

python -m pip install -r requirements.txt


- 第5步: **运行run_app文件打开使用界面(注意:不要关闭界面的命令提示符或终端)**
- 或者在源码目录中使用命令提示符或终端窗口
- 如果要允许界面访问项目外的文件，请在命令中添加 --allow_all_disk

env\Scripts\python.exe main\app\app.py --open


**如果你使用Tensorboard来监控训练**

运行文件: tensorboard 或命令 env\Scripts\python.exe main/app/tensorboard.py


# 使用命令行

python main\app\parser.py --help


# 源码主要结构:

<pre>
Vietnamese-RVC-main
├── assets
│   ├── f0
│   ├── languages
│   │   ├── en-US.json
│   │   └── vi-VN.json
│   ├── logs
│   │   └── mute
│   │       ├── f0
│   │       │   └── mute.wav.npy
│   │       ├── f0_voiced
│   │       │   └── mute.wav.npy
│   │       ├── sliced_audios
│   │       │   ├── mute32000.wav
│   │       │   ├── mute40000.wav
│   │       │   └── mute48000.wav
│   │       ├── sliced_audios_16k
│   │       │   └── mute.wav
│   │       ├── v1_extracted
│   │       │   ├── mute.npy
│   │       │   └── mute_spin.npy
│   │       └── v2_extracted_spin
│   │           ├── mute.npy
│   │           └── mute_spin.npy
│   ├── models
│   │   ├── audioldm2
│   │   ├── embedders
│   │   ├── predictors
│   │   ├── pretrained_custom
│   │   ├── pretrained_v1
│   │   ├── pretrained_v2
│   │   ├── speaker_diarization
│   │   │   ├── assets
│   │   │   │   ├── gpt2.tiktoken
│   │   │   │   ├── mel_filters.npz
│   │   │   │   └── multilingual.tiktoken
│   │   │   └── models
│   │   └── uvr5
│   ├── presets
│   ├── weights
│   └── ico.png
├── audios
├── dataset
├── main
│   ├── app
│   │   ├── app.py
│   │   ├── tensorboard.py
│   │   └── parser.py
│   ├── configs
│   │   ├── v1
│   │   │   ├── 32000.json
│   │   │   ├── 40000.json
│   │   │   └── 48000.json
│   │   ├── v2
│   │   │   ├── 32000.json
│   │   │   ├── 40000.json
│   │   │   └── 48000.json
│   │   ├── config.json
│   │   └── config.py
│   ├── inference
│   │   ├── audio_effects.py
│   │   ├── audioldm2.py
│   │   ├── convert.py
│   │   ├── create_dataset.py
│   │   ├── create_index.py
│   │   ├── extract.py
│   │   ├── preprocess.py
│   │   ├── separator_music.py
│   │   └── train.py
│   ├── library
│   │   ├── algorithm
│   │   │   ├── commons.py
│   │   │   ├── modules.py
│   │   │   ├── mrf_hifigan.py
│   │   │   ├── onnx_export.py
│   │   │   ├── refinegan.py
│   │   │   ├── residuals.py
│   │   │   ├── separator.py
│   │   │   └── stftpitchshift.py
│   │   ├── architectures
│   │   │   ├── demucs_separator.py
│   │   │   ├── fairseq.py
│   │   │   └── mdx_separator.py
│   │   ├── audioldm2
│   │   │   ├── models.py
│   │   │   └── utils.py
│   │   ├── predictors
│   │   │   ├── CREPE.py
│   │   │   ├── FCPE.py
│   │   │   ├── Generator.py
│   │   │   ├── RMVPE.py
│   │   │   ├── SWIPE.py
│   │   │   └── WORLD_WRAPPER.py
│   │   ├── speaker_diarization
│   │   │   ├── audio.py
│   │   │   ├── ECAPA_TDNN.py
│   │   │   ├── embedding.py
│   │   │   ├── encoder.py
│   │   │   ├── features.py
│   │   │   ├── parameter_transfer.py
│   │   │   ├── segment.py
│   │   │   ├── speechbrain.py
│   │   │   └── whisper.py
│   │   ├── uvr5_separator
│   │   │   ├── common_separator.py
│   │   │   ├── spec_utils.py
│   │   │   └── demucs
│   │   │       ├── apply.py
│   │   │       ├── demucs.py
│   │   │       ├── hdemucs.py
│   │   │       ├── htdemucs.py
│   │   │       ├── states.py
│   │   │       └── utils.py
│   │   └── utils.py
│   └── tools
│       ├── gdown.py
│       ├── huggingface.py
│       ├── mediafire.py
│       ├── meganz.py
│       ├── noisereduce.py
│       └── pixeldrain.py
├── docker-compose-cpu.yaml
├── docker-compose-cuda118.yaml
├── docker-compose-cuda128.yaml
├── Dockerfile
├── Dockerfile.cuda118
├── Dockerfile.cuda128
├── LICENSE
├── README.md
├── requirements.txt
├── run_app.bat
└── tensorboard.bat
</pre>

# 注意事项

- **本项目仅支持NVIDIA GPU (如果我有AMD GPU进行测试，可能会在之后支持AMD)**
- **目前新编码器如MRF HIFIGAN的预训练集尚不完整**
- **MRF HIFIGAN和REFINEGAN编码器在不训练音高时不支持训练**
- **Vietnamese-RVC仓库中的模型是从AI Hub、HuggingFace和其他仓库随机收集的。可能有不同的版权许可条款 (例如: Audioldm2的模型权重带有"非商业"条款)**
- **本源代码包含根据"非商业"条款授权的第三方软件组件。任何商业使用行为，包括为衍生软件筹集资金或融资，都可能违反许可并需承担相应的法律责任。**

# 使用条款

- 你必须确保通过本项目中上传和转换的音频内容不侵犯第三方的知识产权。

- 不得将本项目用于任何非法活动，包括但不限于用于欺诈、骚扰或伤害他人。

- 你对因不当使用产品而产生的任何损害承担全部责任。

- 我对因使用本项目而产生的任何直接或间接损害不承担责任。

# 本项目基于以下项目构建

|                                                            作品名称                                                            |         作者         |  许可证  |
|--------------------------------------------------------------------------------------------------------------------------------|-------------------------|-------------|
| **[Applio](https://github.com/IAHispano/Applio/tree/main)**                                                                    | IAHispano               | MIT License |
| **[Python-audio-separator](https://github.com/nomadkaraoke/python-audio-separator/tree/main)**                                 | Nomad Karaoke           | MIT License |
| **[Retrieval-based-Voice-Conversion-WebUI](https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/tree/main)**  | RVC Project             | MIT License |
| **[RVC-ONNX-INFER-BY-Anh](https://github.com/PhamHuynhAnh16/RVC_Onnx_Infer)**                                                  | Phạm Huỳnh Anh          | MIT License |
| **[Torch-Onnx-Crepe-By-Anh](https://github.com/PhamHuynhAnh16/TORCH-ONNX-CREPE)**                                              | Phạm Huỳnh Anh          | MIT License |
| **[Hubert-No-Fairseq](https://github.com/PhamHuynhAnh16/hubert-no-fairseq)**                                                   | Phạm Huỳnh Anh          | MIT License |
| **[Local-attention](https://github.com/lucidrains/local-attention)**                                                           | Phil Wang               | MIT License |
| **[TorchFcpe](https://github.com/CNChTu/FCPE/tree/main)**                                                                      | CN_ChiTu                | MIT License |
| **[FcpeONNX](https://github.com/deiteris/voice-changer/blob/master-custom/server/utils/fcpe_onnx.py)**                         | Yury                    | MIT License |
| **[ContentVec](https://github.com/auspicious3000/contentvec)**                                                                 | Kaizhi Qian             | MIT License |
| **[Mediafiredl](https://github.com/Gann4Life/mediafiredl)**                                                                    | Santiago Ariel Mansilla | MIT License |
| **[Noisereduce](https://github.com/timsainb/noisereduce)**                                                                     | Tim Sainburg            | MIT License |
| **[World.py-By-Anh](https://github.com/PhamHuynhAnh16/world.py)**                                                              | Phạm Huỳnh Anh          | MIT License |
| **[Mega.py](https://github.com/3v1n0/mega.py)**                                                                                | Marco Trevisan          | No License  |
| **[Gdown](https://github.com/wkentaro/gdown)**                                                                                 | Kentaro Wada            | MIT License |
| **[Whisper](https://github.com/openai/whisper)**                                                                               | OpenAI                  | MIT License |
| **[PyannoteAudio](https://github.com/pyannote/pyannote-audio)**                                                                | pyannote                | MIT License |
| **[AudioEditingCode](https://github.com/HilaManor/AudioEditingCode)**                                                          | Hila Manor              | MIT License |
| **[StftPitchShift](https://github.com/jurihock/stftPitchShift)**                                                               | Jürgen Hock             | MIT License |
| **[Codename-RVC-Fork-3](https://github.com/codename0og/codename-rvc-fork-3)**                                                  | Codename;0              | MIT License |

# 模型搜索工具的模型库

- **[VOICE-MODELS.COM](https://voice-models.com/)**

# RVC中的音高提取方法

本文档详细介绍了所使用的音高提取方法，以及根据个人经验总结的每种方法的优缺点、强度和可靠性信息。

| 方法        |      类型      |          优点          |            限制           |      强度      |     可靠性     |
|--------------------|----------------|---------------------------|------------------------------|--------------------|--------------------|
| pm                 | Praat          | 快速                     | 精度较低                | 低               | 低               |
| dio                | PYWORLD        | 适合说唱         | 高频精度较低 | 中等         | 中等         |
| harvest            | PYWORLD        | 比DIO更精确         | 处理较慢               | 高                | 很高            |
| crepe              | Deep Learning  | 高精度             | 需要GPU                  | 很高            | 很高            |
| mangio-crepe       | crepe finetune | 针对RVC优化        | 有时不如原始crepe        | 中等到高 | 中等到高 |
| fcpe               | Deep Learning  | 精确，实时 | 需要强GPU                 | 较好                | 中等         |
| fcpe-legacy        | Old            | 精确，实时 | 较旧                       | 较好                | 中等         |
| rmvpe              | Deep Learning  | 对歌唱效果好    | 消耗资源               | 很高            | 出色           |
| rmvpe-legacy       | Old            | 支持旧系统        | 较旧                       | 高                | 较好                |
| yin                | Librosa        | 简单，高效        | 容易出现倍频错误                   | 中等         | 低               |
| pyin               | Librosa        | 比YIN更稳定           | 计算更复杂       | 较好                | 较好                |
| swipe              | WORLD          | 精度高          | 对噪声敏感           | 高                | 较好                |

# 错误报告

- **如果在使用本源代码时遇到错误，我对此糟糕的体验表示诚挚的歉意，你可以通过以下方式提交错误报告**
- **你可以通过使用界面中的webhook错误报告系统向我报告错误**
- **如果错误报告系统无法工作，你可以通过Discord `pham_huynh_anh` 或 [ISSUE](https://github.com/PhamHuynhAnh16/Vietnamese-RVC/issues) 向我报告错误**

# ☎️ 联系我
- Discord: **pham_huynh_anh**

AnhP/Vietnamese-RVC-Project

作者 AnhP

audio-to-audio

↓ 0 ♥ 8

创建时间: 2024-12-04 11:50:16+00:00

更新时间: 2026-03-01 10:08:09+00:00

在 Hugging Face 上查看

文件 (273)

.gitattributes

README.md

binary/decrypt.bin

binary/world.bin

embedders/fairseq/chinese_hubert_base.pt

embedders/fairseq/contentvec_base.pt

embedders/fairseq/hubert_base.pt

embedders/fairseq/japanese_hubert_base.pt

embedders/fairseq/korean_hubert_base.pt

embedders/fairseq/portuguese_hubert_base.pt

embedders/fairseq/vietnamese_hubert_base.pt

embedders/onnx/chinese_hubert_base.onnx ONNX

embedders/onnx/contentvec_base.onnx ONNX

embedders/onnx/hubert_base.onnx ONNX

embedders/onnx/japanese_hubert_base.onnx ONNX

embedders/onnx/korean_hubert_base.onnx ONNX

embedders/onnx/portuguese_hubert_base.onnx ONNX

embedders/onnx/vietnamese_hubert_base.onnx ONNX

embedders/spin/spin-v1/config.json

embedders/spin/spin-v1/model.safetensors

embedders/spin/spin-v1/pytorch_model.bin

embedders/spin/spin-v2/config.json

embedders/spin/spin-v2/model.safetensors

embedders/spin/spin-v2/pytorch_model.bin

embedders/transformers/chinese_hubert_base/config.json

embedders/transformers/chinese_hubert_base/model.safetensors

embedders/transformers/chinese_hubert_base/pytorch_model.bin

embedders/transformers/contentvec_base/config.json

embedders/transformers/contentvec_base/model.safetensors

embedders/transformers/contentvec_base/pytorch_model.bin

embedders/transformers/hubert_base/config.json

embedders/transformers/hubert_base/model.safetensors

embedders/transformers/hubert_base/pytorch_model.bin

embedders/transformers/japanese_hubert_base/config.json

embedders/transformers/japanese_hubert_base/model.safetensors

embedders/transformers/japanese_hubert_base/pytorch_model.bin

embedders/transformers/korean_hubert_base/config.json

embedders/transformers/korean_hubert_base/model.safetensors

embedders/transformers/korean_hubert_base/pytorch_model.bin

embedders/transformers/portuguese_hubert_base/config.json

embedders/transformers/portuguese_hubert_base/model.safetensors

embedders/transformers/portuguese_hubert_base/pytorch_model.bin

embedders/transformers/vietnamese_hubert_base/config.json

embedders/transformers/vietnamese_hubert_base/model.safetensors

embedders/transformers/vietnamese_hubert_base/pytorch_model.bin

ffmpeg/ffmpeg.exe

ffmpeg/ffprobe.exe

ico.png

json/custom_pretrained.json

json/model_data.json

json/uvr_models.json

predictors/World/world64.dll

predictors/World/world86.dll

predictors/World/world_android64.so

predictors/World/world_android86.so

predictors/World/world_arm64.so

predictors/World/world_arm_eabi.so

predictors/World/world_linux.so

predictors/World/world_mac.dylib

predictors/crepe_full.onnx ONNX

predictors/crepe_full.pth

predictors/crepe_large.onnx ONNX

predictors/crepe_large.pth

predictors/crepe_medium.onnx ONNX

predictors/crepe_medium.pth

predictors/crepe_small.onnx ONNX

predictors/crepe_small.pth

predictors/crepe_tiny.onnx ONNX

predictors/crepe_tiny.pth

predictors/ddsp_200k.onnx ONNX

predictors/ddsp_200k.pt

predictors/djcm-svs.onnx ONNX

predictors/djcm-svs.pt

predictors/djcm.onnx ONNX

predictors/djcm.pt

predictors/fcn.onnx ONNX

predictors/fcn.pt

predictors/fcpe.onnx ONNX

predictors/fcpe.pt

predictors/fcpe_legacy.onnx ONNX

predictors/fcpe_legacy.pt

predictors/hpa-rmvpe-112000.onnx ONNX

predictors/hpa-rmvpe-112000.pt

predictors/hpa-rmvpe-76000.onnx ONNX

predictors/hpa-rmvpe-76000.pt

predictors/pesto.onnx ONNX

predictors/pesto.pt

predictors/rmvpe.onnx ONNX

predictors/rmvpe.pt

predictors/swift.onnx ONNX

pretrained_custom/32k-TITAN-Medium.zip

pretrained_custom/40k-TITAN-Medium.zip

pretrained_custom/48k-TITAN-Medium.zip

pretrained_custom/BeatzForge_V2_32k.zip

pretrained_custom/DMR-V1.zip

pretrained_custom/F0_SingerPretrain.zip

pretrained_custom/GuideVocalPretrain.zip

pretrained_custom/IMA-TEST-V1.zip

pretrained_custom/Italla_32k.zip

pretrained_custom/KLM41_32k.zip

pretrained_custom/KLM41_48k.zip

pretrained_custom/KLM42_32k.zip

pretrained_custom/KLM42_T4_40k.zip

pretrained_custom/KLM43x1_32k.zip

pretrained_custom/KLM43x2_32k.zip

pretrained_custom/KLM43x3_32k.zip

pretrained_custom/KLM43x3_40k.zip

pretrained_custom/KLM43x3_48k.zip

pretrained_custom/KLM43x4_32k.zip

pretrained_custom/KLM43x4_40k.zip

pretrained_custom/KLM43x4_48k.zip

pretrained_custom/KLM4_32k.zip

pretrained_custom/KLM4_40k.zip

pretrained_custom/KLM4_48k.zip

pretrained_custom/KLM_BeatMasterV1_48k.zip

pretrained_custom/KLM_BeatzForge_48k.zip

pretrained_custom/KLM_HFG49_32k.zip

pretrained_custom/KLM_HFG49_40k.zip

pretrained_custom/KLM_HFG49_48k.zip

pretrained_custom/Nezox_32k.zip

pretrained_custom/RIN_E3.zip

pretrained_custom/Rigel_32k_3890220.zip

pretrained_custom/Rigel_32k_fp32_2854856.zip

pretrained_custom/Rigel_Portuguese_50e.zip

pretrained_custom/Snowie-X-Rin-40k.zip

pretrained_custom/SnowieV3.1_32k.zip

pretrained_custom/SnowieV3.1_40k.zip

pretrained_custom/SnowieV3.1_48k.zip

pretrained_custom/Snowie_RuPretrain_48k_V1.2.zip

pretrained_custom/Snowie_RuPretrain_EnP.zip

pretrained_custom/UKA.zip

pretrained_custom/f0Ov2Super32k.zip

pretrained_custom/f0Ov2Super40k.zip

pretrained_custom/nanashi_anime_384e.zip

pretrained_custom/nanashi_anime_resized.zip

pretrained_custom/nanashi_v1.5.zip

pretrained_custom/nanashi_v1.7.zip

pretrained_custom/nanashi_v1.zip

pretrained_custom/nanashi_v2_base.zip

pretrained_custom/nanashi_v2_finetune.zip

pretrained_v1/D32k.pth

pretrained_v1/D40k.pth

pretrained_v1/D48k.pth

pretrained_v1/G32k.pth

pretrained_v1/G40k.pth

pretrained_v1/G48k.pth

pretrained_v1/SVC_f0D44k.pth

pretrained_v1/SVC_f0G44k.pth

pretrained_v1/f0D32k.pth

pretrained_v1/f0D40k.pth

pretrained_v1/f0D48k.pth

pretrained_v1/f0G32k.pth

pretrained_v1/f0G40k.pth

pretrained_v1/f0G48k.pth

pretrained_v2/D32k.pth

pretrained_v2/D40k.pth

pretrained_v2/D48k.pth

pretrained_v2/G32k.pth

pretrained_v2/G40k.pth

pretrained_v2/G48k.pth

pretrained_v2/RefineGAN_f0D32k.pth

pretrained_v2/RefineGAN_f0D40k.pth

pretrained_v2/RefineGAN_f0D48k.pth

pretrained_v2/RefineGAN_f0G32k.pth

pretrained_v2/RefineGAN_f0G40k.pth

pretrained_v2/RefineGAN_f0G48k.pth

pretrained_v2/SVC_f0D44k.pth

pretrained_v2/SVC_f0G44k.pth

pretrained_v2/f0D32k.pth

pretrained_v2/f0D40k.pth

pretrained_v2/f0D48k.pth

pretrained_v2/f0G32k.pth

pretrained_v2/f0G40k.pth

pretrained_v2/f0G48k.pth

speaker_diarization/base.en.pt

speaker_diarization/base.pt

speaker_diarization/large-v1.pt

speaker_diarization/large-v2.pt

speaker_diarization/large-v3-turbo.pt

speaker_diarization/large-v3.pt

speaker_diarization/medium.en.pt

speaker_diarization/medium.pt

speaker_diarization/small.en.pt

speaker_diarization/small.pt

speaker_diarization/speechbrain/classifier.ckpt

speaker_diarization/speechbrain/config.json

speaker_diarization/speechbrain/embedding_model.ckpt

speaker_diarization/speechbrain/hyperparams.yaml

speaker_diarization/speechbrain/mean_var_norm_emb.ckpt

speaker_diarization/tiny.en.pt

speaker_diarization/tiny.pt

uvr5_models/Demucs/04573f0d-f3cf25b2.th

uvr5_models/Demucs/5c90dfd2-34c22ccb.th

uvr5_models/Demucs/75fc33f5-1941ce65.th

uvr5_models/Demucs/92cfc3b6-ef3bcb9c.th

uvr5_models/Demucs/955717e8-8726e21a.th

uvr5_models/Demucs/d12395a8-e57c48e6.th

uvr5_models/Demucs/f7e0c4bc-ba3fe64a.th

uvr5_models/Demucs/hdemucs_mmi.yaml

uvr5_models/Demucs/htdemucs.yaml

uvr5_models/Demucs/htdemucs_6s.yaml

uvr5_models/Demucs/htdemucs_ft.yaml

uvr5_models/MDX/Kim_Inst.onnx ONNX

uvr5_models/MDX/Kim_Vocal_1.onnx ONNX

uvr5_models/MDX/Kim_Vocal_2.onnx ONNX

uvr5_models/MDX/Reverb_HQ_By_FoxJoy.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_1.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_2.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_3.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_HQ_1.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_HQ_2.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_HQ_3.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_HQ_4.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_HQ_5.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_Main.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Inst_full_292.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET-Voc_FT.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET_Crowd_HQ_1.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET_Inst_187_beta.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET_Inst_82_beta.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET_Inst_90_beta.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET_Main_340.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET_Main_390.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET_Main_406.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET_Main_427.onnx ONNX

uvr5_models/MDX/UVR-MDX-NET_Main_438.onnx ONNX

uvr5_models/MDX/UVR_MDXNET_1_9703.onnx ONNX

uvr5_models/MDX/UVR_MDXNET_2_9682.onnx ONNX

uvr5_models/MDX/UVR_MDXNET_3_9662.onnx ONNX

uvr5_models/MDX/UVR_MDXNET_9482.onnx ONNX

uvr5_models/MDX/UVR_MDXNET_KARA.onnx ONNX

uvr5_models/MDX/UVR_MDXNET_KARA_2.onnx ONNX

uvr5_models/MDX/UVR_MDXNET_Main.onnx ONNX

uvr5_models/VR/10_SP-UVR-2B-32000-1.pth

uvr5_models/VR/11_SP-UVR-2B-32000-2.pth

uvr5_models/VR/12_SP-UVR-3B-44100.pth

uvr5_models/VR/13_SP-UVR-4B-44100-1.pth

uvr5_models/VR/14_SP-UVR-4B-44100-2.pth

uvr5_models/VR/15_SP-UVR-MID-44100-1.pth

uvr5_models/VR/16_SP-UVR-MID-44100-2.pth

uvr5_models/VR/1_HP-UVR.pth

uvr5_models/VR/2_HP-UVR.pth

uvr5_models/VR/3_HP-Vocal-UVR.pth

uvr5_models/VR/4_HP-Vocal-UVR.pth

uvr5_models/VR/5_HP-Karaoke-UVR.pth

uvr5_models/VR/6_HP-Karaoke-UVR.pth

uvr5_models/VR/7_HP2-UVR.pth

uvr5_models/VR/8_HP2-UVR.pth

uvr5_models/VR/9_HP2-UVR.pth

uvr5_models/VR/UVR-De-Echo-Aggressive.pth

uvr5_models/VR/UVR-De-Echo-Normal.pth

uvr5_models/VR/UVR-DeEcho-DeReverb.pth

uvr5_models/VR/UVR-DeNoise-Lite.pth

uvr5_models/VR/UVR-DeNoise.pth

uvr5_models/mdx_c_configs/config_melbandroformer_inst.yaml

uvr5_models/mdx_c_configs/config_melbandroformer_inst_v2.yaml

uvr5_models/mdx_c_configs/config_melbandroformer_instvoc_duality.yaml

uvr5_models/mdx_c_configs/config_vocals_mel_band_roformer_kim.yaml

uvr5_models/mdx_c_configs/model1.yaml

uvr5_models/mdx_c_configs/model2.yaml

uvr5_models/mdx_c_configs/model3.yaml

uvr5_models/mdx_c_configs/modelA.yaml

uvr5_models/mdx_c_configs/modelB.yaml

uvr5_models/mdx_c_configs/model_2_stem_061321.yaml

uvr5_models/mdx_c_configs/model_2_stem_full_band.yaml

uvr5_models/mdx_c_configs/model_2_stem_full_band_2.yaml

uvr5_models/mdx_c_configs/model_2_stem_full_band_3.yaml

uvr5_models/mdx_c_configs/model_2_stem_full_band_8k.yaml

uvr5_models/mdx_c_configs/model_bs_roformer_ep_317_sdr_12.9755.yaml

uvr5_models/mdx_c_configs/model_bs_roformer_ep_368_sdr_12.9628.yaml

uvr5_models/mdx_c_configs/model_bs_roformer_ep_937_sdr_10.5309.yaml

uvr5_models/mdx_c_configs/model_mel_band_roformer_ep_3005_sdr_11.4360.yaml

uvr5_models/mdx_c_configs/sndfx.yaml