在浏览器中进行 AI 分轨与音频清理

Stems ST-02 is a browser-based audio separation tool that isolates vocals, drums, bass and other instruments from any song. It runs Facebook's open-source Demucs model directly in your web browser, so you can split a track into individual stems without installing a plugin or paying upfront. It is aimed at DJs, producers, remixers and music learners who need clean stems for edits, mashups or practice.

这是编辑给出的评估,满分 5 分,不是用户打分的平均值。 本页的各方观点由 AI 根据产品信息归纳,不是真实用户的评价。
根据 OpenAITools.app 于 2026-11-11 核对的流量数据,Stems ST-02(stems.app)是一个收录在 AI 分轨工具下的 AI 平台,月访问量为 0,月度环比流量变化为 -145(-100%)。其主要地理市场为全球(— 流量份额),采用免费定价模式。按月流量计算的主要竞争替代方案包括 Splitter.ai(8.32万次访问/月),SplitSong.com(2335 次访问/月),LALAL.AI(196.34万次访问/月)。
不用装python直接在网页跑Demucs,太强了!
人声和伴奏分离得很干净。
免费还好用,良心网站。
大文件上传要等一会儿。
以下短评基于实时网页探查、定价机制与同类产品基准由 AI 综合模拟生成,旨在呈现不同使用场景下的常见槽点与亮点,非实名真实用户评价。
Demucs-based model splits any track into vocals, drums, bass and other, producing four clean audio files you can download individually.
Everything runs in the browser tab — no plugin installation, no account required to try a split.
Upload a track, wait for the model to finish, then preview and download each isolated stem.
Built for music makers: pull out an a cappella, grab a drum loop, or study the bassline of a reference track.
Powered by Facebook's open-source Demucs library, a state-of-the-art source separation model.
Stems ST-02 is a browser-based audio separation tool that isolates vocals, drums, bass and other instruments from any song. It runs Facebook's open-source Demucs model directly in your web browser, so you can split a track into individual stems without installing a plugin or paying upfront. It is aimed at DJs, producers, remixers and music learners who need clean stems for edits, mashups or practice.
Extract an a cappella or instrumental on the fly to build a live edit, mashup or transition.
Isolate a drum break or bassline to sample, chop and layer into a new production.
Remove vocals from a track so you can play along or practise a part without distraction.
Strip a reference mix into parts to study arrangement, balance and how elements sit together.
同一分类下同类 AI 工具的流量并排对比
| 排名 | AI 工具 | 月访问量 | 月度环比增长 | 直接流量占比 | 定价 | 头对头 |
|---|---|---|---|---|---|---|
| ● | Stems ST-02 | 0 | -100% | — | 免费 | 当前产品 |
| 1 | Splitter.ai | 8.32万 | -33.4% | 38% 直接流量 | 免费增值,免费试用 | 对比 Splitter.ai |
| 2 | SplitSong.com | 2335 | +9.7% | 40% 直接流量 | 免费增值 | 对比 SplitSong.com |
| 3 | LALAL.AI | 196.34万 | -12.1% | 33% 直接流量 | 免费增值,付费 | 对比 LALAL.AI |
| 4 | VocalRemover.co | 5.32万 | -2.5% | 27% 直接流量 | 免费增值 | 对比 VocalRemover.co |
| 5 | Vocal Remover Online | 6528 | -6.4% | 77% 直接流量 | 免费增值 | 对比 Vocal Remover Online |
| 6 | Moises | 448.65万 | +6.6% | 44% 直接流量 | 免费增值 | 对比 Moises |
在浏览器中进行 AI 分轨与音频清理
SplitSong.com is a browser-based AI stem splitter: upload a song or paste a YouTube link and it returns separate drums, instrumental, bass and vocal tracks you can download individually. It is built for quick karaoke prep, remix work, practice stems and sampling, with no software install required.
SplitSong.com is a browser-based AI stem splitter: upload a song or paste a YouTube link and it returns separate drums, instrumental, bass and vocal tracks you can download individually. It is built for quick karaoke prep, remix work, practice stems and sampling, with no software install required.
LALAL.AI is an online audio separation platform that uses AI and transformer-based neural networks to split music and video tracks into individual stems. It grew from a single-purpose vocal remover into a full audio suite covering vocal and instrumental isolation, multi-stem splitting, noise and echo cleanup, voice changing, and voice cloning.
LALAL.AI is an online audio separation platform that uses AI and transformer-based neural networks to split music and video tracks into individual stems. It grew from a single-purpose vocal remover into a full audio suite covering vocal and instrumental isolation, multi-stem splitting, noise and echo cleanup, voice changing, and voice cloning.

VocalRemover.co is an online AI vocal remover and audio splitter that separates any song into vocals, instrumentals, and individual stems, making karaoke tracks, acapellas, and remixes in a few simple steps.
VocalRemover.co is an online AI vocal remover and audio splitter that separates any song into vocals, instrumentals, and individual stems, making karaoke tracks, acapellas, and remixes in a few simple steps.
Vocal Remover Online (VocalRemoverOak) is a browser-based AI audio separation service that removes vocals, extracts accompaniment, isolates vocals and splits songs into stems from uploaded files, videos or streaming links.
Vocal Remover Online (VocalRemoverOak) is a browser-based AI audio separation service that removes vocals, extracts accompaniment, isolates vocals and splits songs into stems from uploaded files, videos or streaming links.
从任意歌曲中去除鼓声,在浏览器里直接跟练。
基于浏览器的 AI 人声分离工具,制作伴奏、卡拉OK 与清唱
在浏览器中分离或去除任意歌曲的人声