一些网友或许在有需要的时候，会想要将音乐人声和伴奏分离开来，但是平时很少用到这方面的工具，于是从别的地方给大家搬回来点东西，这些都是在线分离工具，除了个别已经标注收费的之外，其余基本上都是免费的哦。Some netizens may want to separate the vocal and accompaniment of music when they need it, but they rarely use tools in this area, so they move back something for everyone from other places. These are all online separations. Except for some of the tools that have been marked as charged, the rest are basically free.
1. Just select the uploaded music, the other options are default, click start. After waiting for upload, website conversion and extraction, there will be a download link, and the directly downloaded file is the separated audio.
Among them, vocals.mp3 is audio with only human voice, and accompaniment.mp3 is audio with only background music.
If you use it for free, the uploaded song cannot exceed 10m, and you may need to wait in line.
2. The usage of this website is similar to the above, but it seems that you need to go online in a scientific way (this can only be done by yourself (￣▽￣)).
First fill in your email address, then upload the mp3 file, click Processing after uploading, and wait a few minutes for background processing, and then you will receive the processed audio compression package in the email address you filled in.
The content of the compressed package is also consistent with the above.
The first two songs are free, and the rest need to be paid at 50 cents each.
The above is an open source project based on Deezer. 3. Vocalremover (recommended)
Separate the sound in the song from the music with the help of a powerful AI algorithm.
Upload audio files, this is not limited to 10MB.
The effect is also very good, and it is completely free, it is recommended to use.
The following introduces several background sound separation websites developed by Chinese people. Fourth, Tuanzi AI is a website developed by Chinese people. The website is in Chinese, which is relatively more friendly. The usage method is basically the same as the above two stations, upload mp3, wait, and download the separated audio.
It is also based on the Python liuliushe.net open source project Spleeter? https://github.com/deezer/spleeter?, using ai technology to separate accompaniment, vocals, drums, bass and other instruments from any audio. Free users can listen to the website. 30s, you can use up to 5 songs, and you need to pay for more and full versions.
5. Tracking is not completely free. There is a free quota of 10 minutes per month, which is better than Tuanzi AI.
6. Sogou Sound Coffee can separate accompaniment, human voice and video screen from any audio/video, which can quickly provide material for video production and audio content production.
7. Silencing-Yunmao transcoding is similar to Sogou's voice coffee, supporting audio and video, and the maximum file size of ordinary users is 100M.
Then introduce some relatively small online accompaniment separation websites. 8. It can automatically extract and delete vocals from 2 kinds of mixed sound sources.
When uploading a sound source, you can download two sound sources, that is, a sound source that only extracts sound from the sound source and a sound source that only removes vocals.
9. You can separate the voice and background music for free by uploading songs directly.
10. Use artificial intelligence to separate music tracks, upload songs, and then extract accompaniment tracks and human voice tracks.
Upload the audio file after logging in with your Google account.
After completion, download the vocal and accompaniment audio files.
一、Ezstems
https://ezstems.com/
只需要选择上传的音乐，其他的选择默认，点击start。等待上传，网站转换和提取之后，会有一个下载链接，直接下载下来的文件就是分离好的音频。
其中，vocals.mp3是只有人声的音频，accompaniment.mp3则是只有背景音乐的音频。
免费的使用的话，上传的歌曲不能超过10m，而且很可能需要排队等候。
二、Melody
https://melody.ml/
这个网站和上面的用法差不多，但是貌似需要科学那个什么上网（这个只能自己想办法哈(￣▽￣)）。
先填写的是自己的邮箱，然后上传mp3文件，上传完成后点击Processing，然后等待几分钟后台处理，之后你填写的邮箱就会收到处理后的音频压缩包。
压缩包的内容也和上面的是一致的。
前两首音乐免费，之后的需要付费，每首50美分。
以上是基于Deezer的开源项目spleeter
三、Vocalremover（推荐）
https://vocalremover.org/
借助强大的AI算法将歌曲中的声音与音乐分离开来 。
上传音频文件，这个不限制10MB 。
效果也很不错，而且完全免费，推荐使用。
以下再介绍几款国人开发的背景声音分离网站
四、团子AI
https://dango.ai/
这个站是国人开发的，网站是中文的，相对更友好一些。使用方法和上面两个站基本一样，上传mp3，等待，下载分离好的音频。
也是基于 Python liuliushe.net 开源项目 Spleeter https://github.com/deezer/spleeter 做的，利用 ai技术从任何音频中分离出伴奏，人声，鼓点，贝斯以及其他乐器
免费用户可用网站试听30s，最多使用5首歌曲，需要更多和完整版本的则需要付费。
五、分轨
https://ifengui.com/
不是完全免费，每个月有10分钟免费额度，这点比团子AI强。
六、搜狗声咖
https://shengka.ai.sogou.com/
实现从任何音频/视频中分离出伴奏，人声以及视频画面，可快速为视频制作、有声内容生产提供素材。
七、消音 – 云猫转码
https://yunmaovideo.com/remove-audio/
和搜狗声咖差不多，支持音频和视频，
普通用户文件大小上限为 100M。
再介绍一些比较小众的伴奏在线分离网站
八、PhonicMind
https://phonicmind.com/
可以自动从2种混音声源中提取和删除人声。
上传声源时，您可以下载两个声源，即仅从声源中提取声音的声源和仅移除声乐的声源。
九、lalal
https://www.lalal.ai/
直接上传歌曲就可以免费将语音和背景音乐分离。
十、moises
https://moises.ai/
使用人工智能分离音乐音轨，上传歌曲，然后提取出伴奏音轨和人声音轨。
用谷歌账号登录后上传音频文件。
完成后下载人声和伴奏音频文件。
