Use VITS and Opencpop to develop singing voice synthesis; Maybe it will VISinger.

Overview

Init

Use VITS and Opencpop to develop singing voice synthesis; Maybe it will VISinger.

本项目基于

https://github.com/jaywalnut310/vits
https://github.com/SJTMusicTeam/Muskits/
https://wenet.org.cn/opencpop/ 歌声数据

使用muskit数据预处理,获得初步数据

cd egs/opencpop/svs1/
./local/data.sh

VISinger_data
--lable
--midi_dump
--wav_dump

采样率转换

python wave_16k.py
--wav_dump
--wav_dump_16k

使用muskit将数据处理成vits的格式

1, 将lable进行拆分
python muskit/data_label_single.py

label_dump,midi_dump,wav_dump:一个文件一个标注

注意:label和lable的混用(两个单词都是对的)

VISinger_data
--label_dump
--midi_dump
--wav_dump
--wav_dump_16k

2, 将label和midi处理为frame对应的发音单元和音符(基音)
python muskit/data_format_vits.py
VISinger_data
--label_vits
--label_dump
--midi_dump
--wav_dump
--wav_dump_16k

3, 生成VITS需要的files,并分割为train和dev,test不需要(可以手动设计)
python muskit/data_format_vits.py

vits_file.txt 中的内容格式:wave path|label path|pitch path;

cp vits_file.txt VISinger/filelists/
cd VISinger/

python preprocess.py 分割为train和dev

VITS训练

cd VISinger
CUDA_VISIBLE_DEVICES=0 python train.py -c configs/singing_base.json -m singing_base 2>exit_error.log;cat exit_error.log
python vsinging_infer.py

使用16K节约内存,方便模型修改

编辑midi,然后测试

cd ../;python muskit/infer_midi.py;cd -;python vsinging_edit.py

LOSS值 MEL谱

样例音频

vits_singing_样例.wav

You might also like...
In this project, we develop a face recognize platform based on MTCNN object-detection netcwork and FaceNet self-supervised network.
In this project, we develop a face recognize platform based on MTCNN object-detection netcwork and FaceNet self-supervised network.

模式识别大作业——人脸检测与识别平台 本项目是一个简易的人脸检测识别平台,提供了人脸信息录入和人脸识别的功能。前端采用 html+css+js,后端采用 pytorch,

Official codebase used to develop Vision Transformer, MLP-Mixer, LiT and more.

Big Vision This codebase is designed for training large-scale vision models on Cloud TPU VMs. It is based on Jax/Flax libraries, and uses tf.data and

Use MATLAB to simulate the signal and extract features. Use PyTorch to build and train deep network to do spectrum sensing.

Deep-Learning-based-Spectrum-Sensing Use MATLAB to simulate the signal and extract features. Use PyTorch to build and train deep network to do spectru

Transfer style api - An API to use with Tranfer Style App, where you can use two image and transfer the style

Transfer Style API It's an API to use with Tranfer Style App, where you can use

Voice of Pajlada with model and weights.

Pajlada TTS Stripped down version of ForwardTacotron (https://github.com/as-ideas/ForwardTacotron) with pretrained weights for Pajlada's (https://gith

A voice recognition assistant similar to amazon alexa, siri and google assistant.
A voice recognition assistant similar to amazon alexa, siri and google assistant.

kenyan-Siri Build an Artificial Assistant Full tutorial (video) To watch the tutorial, click on the image below Installation For windows users (run th

An implementation of
An implementation of "Optimal Textures: Fast and Robust Texture Synthesis and Style Transfer through Optimal Transport"

Optex An implementation of Optimal Textures: Fast and Robust Texture Synthesis and Style Transfer through Optimal Transport for TU Delft CS4240. You c

this is a lite easy to use virtual keyboard project for anyone to use
this is a lite easy to use virtual keyboard project for anyone to use

virtual_Keyboard this is a lite easy to use virtual keyboard project for anyone to use motivation I made this for this year's recruitment for RobEn AA

A collection of easy-to-use, ready-to-use, interesting deep neural network models
A collection of easy-to-use, ready-to-use, interesting deep neural network models

Interesting and reproducible research works should be conserved. This repository wraps a collection of deep neural network models into a simple and un

Comments
  • couple of questions

    couple of questions

    Hello how are you ! very cool stuff you have here ,I can clearly see you love singing voice synthesis (SVS) from your forks and repos !! i wanted to ask is that a fully working Visingerr or is it a try from you to make it to sing , like can it be tested on a custom English data and have like results the same as or near the demo in the paper. Also do you have like other samples i can hear , i know that you tested it on opencpop that has almost 5.2 hours of singing data , and also in the paper they trained Visingerr for 600k iterations right ? how many iterations did you achieve on the opencpop to get the result linked below (vits_singing_样例.wav). to be honest i thought vits is data hungry like tacotron2 or fastspeech (aka needs a lot of data to get great results) , that opencpop result of your is so impressive for 5.2 hours data , i also wonder if you lowered the sample rate of opencpop from 44.1 KHz to 22KHz as i heard 44.1 KHz takes alot of time to train x10 the time needed.

    迫不及待地想知道你的消息 :)

    opened by dutchsing009 5
  • 问题

    问题

    python prepare/data_vits.py 输出 1,../VISinger_data/label_vits/XXX._label.npy|XXX_score.npy|XXX_pitch.npy|XXX_slurs.npy 2,filelists/vits_file.txt 内容格式:wave path|label path|score path|pitch path|slurs path;

    请问1 2这两步是怎么操作?

    opened by baipeng0110 3
  • 训练结果

    训练结果

    目前模型缺乏时长预测模型和基音预测模型; 训练语料中的句子修改歌词的效果;

    原歌词:雨淋湿了天空灰得更讲究

    https://user-images.githubusercontent.com/16432329/164953151-4c2513cb-f336-416b-8f04-604f13e63368.MP4

    修改歌词:你闹够了没有让我更难受

    https://user-images.githubusercontent.com/16432329/164953155-16c72670-cc89-40bc-99fe-42781c9dcdc0.MP4

    help wanted 
    opened by MaxMax2016 0
  • About release models and VISinger

    About release models and VISinger

    Hi

    This is a fantastic project that I have ever seen.

    Could you please share the released model? As on the inference step, it is said that "using the released model"

    Also, is there any plan to implement the VISinger model?

    Thank you!

    opened by shiyanpei0826 1
Owner
AmorTX
Speech
AmorTX
Few-shot Neural Architecture Search

One-shot Neural Architecture Search uses a single supernet to approximate the performance each architecture. However, this performance estimation is super inaccurate because of co-adaption among oper

Yiyang Zhao 38 Oct 18, 2022
Wider or Deeper: Revisiting the ResNet Model for Visual Recognition

ademxapp Visual applications by the University of Adelaide In designing our Model A, we did not over-optimize its structure for efficiency unless it w

Zifeng Wu 338 Dec 12, 2022
A solution to the 2D Ising model of ferromagnetism, implemented using the Metropolis algorithm

Solving the Ising model on a 2D lattice using the Metropolis Algorithm Introduction The Ising model is a simplified model of ferromagnetism, the pheno

Rohit Prabhu 5 Nov 13, 2022
Code to compute permutation and drop-column importances in Python scikit-learn models

Feature importances for scikit-learn machine learning models By Terence Parr and Kerem Turgutlu. See Explained.ai for more stuff. The scikit-learn Ran

Terence Parr 537 Dec 31, 2022
Age Progression/Regression by Conditional Adversarial Autoencoder

Age Progression/Regression by Conditional Adversarial Autoencoder (CAAE) TensorFlow implementation of the algorithm in the paper Age Progression/Regre

Zhifei Zhang 603 Dec 22, 2022
Instance-Dependent Partial Label Learning

Instance-Dependent Partial Label Learning Installation pip install -r requirements.txt Run the Demo benchmark-random mnist python -u main.py --gpu 0 -

17 Dec 29, 2022
Code for the ICML 2021 paper: "ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision"

ViLT Code for the paper: "ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision" Install pip install -r requirements.txt pip

Wonjae Kim 922 Jan 01, 2023
Curved Projection Reformation

Description Assuming that we already know the image of the centerline, we want the lumen to be displayed on a plane, which requires curved projection

夜听残荷 5 Sep 11, 2022
Official Implementation of "Third Time's the Charm? Image and Video Editing with StyleGAN3" https://arxiv.org/abs/2201.13433

Third Time's the Charm? Image and Video Editing with StyleGAN3 Yuval Alaluf*, Or Patashnik*, Zongze Wu, Asif Zamir, Eli Shechtman, Dani Lischinski, Da

531 Dec 20, 2022
An implementation of EWC with PyTorch

EWC.pytorch An implementation of Elastic Weight Consolidation (EWC), proposed in James Kirkpatrick et al. Overcoming catastrophic forgetting in neural

Ryuichiro Hataya 166 Dec 22, 2022
Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering

Graph ConvNets in PyTorch October 15, 2017 Xavier Bresson http://www.ntu.edu.sg/home/xbresson https://github.com/xbresson https://twitter.com/xbresson

Xavier Bresson 287 Jan 04, 2023
PyTorch 1.5 implementation for paper DECOR-GAN: 3D Shape Detailization by Conditional Refinement.

DECOR-GAN PyTorch 1.5 implementation for paper DECOR-GAN: 3D Shape Detailization by Conditional Refinement, Zhiqin Chen, Vladimir G. Kim, Matthew Fish

Zhiqin Chen 72 Dec 31, 2022
Read and write layered TIFF ImageSourceData and ImageResources tags

Read and write layered TIFF ImageSourceData and ImageResources tags Psdtags is a Python library to read and write the Adobe Photoshop(r) specific Imag

Christoph Gohlke 4 Feb 05, 2022
A hybrid framework (neural mass model + ML) for SC-to-FC prediction

The current workflow simulates brain functional connectivity (FC) from structural connectivity (SC) with a neural mass model. Gradient descent is applied to optimize the parameters in the neural mass

Yilin Liu 1 Jan 26, 2022
General purpose GPU compute framework for cross vendor graphics cards (AMD, Qualcomm, NVIDIA & friends)

General purpose GPU compute framework for cross vendor graphics cards (AMD, Qualcomm, NVIDIA & friends). Blazing fast, mobile-enabled, asynchronous and optimized for advanced GPU data processing usec

The Kompute Project 1k Jan 06, 2023
Project repo for the paper SILT: Self-supervised Lighting Transfer Using Implicit Image Decomposition

SILT: Self-supervised Lighting Transfer Using Implicit Image Decomposition (BMVC 2021) Project repo for the paper SILT: Self-supervised Lighting Trans

6 Dec 04, 2022
Codes to pre-train T5 (Text-to-Text Transfer Transformer) models pre-trained on Japanese web texts

t5-japanese Codes to pre-train T5 (Text-to-Text Transfer Transformer) models pre-trained on Japanese web texts. The following is a list of models that

Kimio Kuramitsu 1 Dec 13, 2021
Official implementation of the paper Chunked Autoregressive GAN for Conditional Waveform Synthesis

PyEmits, a python package for easy manipulation in time-series data. Time-series data is very common in real life. Engineering FSI industry (Financial

Descript 150 Dec 06, 2022
A modular domain adaptation library written in PyTorch.

A modular domain adaptation library written in PyTorch.

Kevin Musgrave 225 Dec 29, 2022
Jittor implementation of Recursive-NeRF: An Efficient and Dynamically Growing NeRF

Recursive-NeRF: An Efficient and Dynamically Growing NeRF This is a Jittor implementation of Recursive-NeRF: An Efficient and Dynamically Growing NeRF

33 Nov 30, 2022