Overview
Create an instrumental for a sing-along, keep the vocal track for an edit, or separate more parts from an audio file with AI Vocal Remover. Under Extract, choose Instrumental Only, Vocals Only, or Both Tracks (Separate Files). Stem Separation offers 2-Stem, 4-Stem, and 6-Stem options. The configured output is WAV, though selections with multiple separate tracks download as a ZIP. Listen for vocal traces or missing instrumental detail before using the result. Online processing requires sign-in and an internet connection.
Extract an instrumental, a vocal track, or separate tracks from an audio file. Choose two, four, or six stems.
When to use
- Make an instrumental or karaoke practice track from a supported audio file by choosing Instrumental Only.
- Keep the vocal part for a cover or edit, or request the vocals and instrumental as separate tracks.
- Choose 4-Stem or 6-Stem when you want more parts than the vocal-and-music split.
How to use
Open AI Vocal Remover and use its linked workspace. For online processing, sign in and connect to the internet.
Upload one audio file, up to 100 MB. The listed input extensions are .mp3, .wav, .ogg, .flac, .m4a, .aac, and .wma; the extension list doesn’t establish codec or bit-depth support. Under Extract, choose Instrumental Only, Vocals Only, or Both Tracks (Separate Files). Under Stem Separation, choose 2-Stem (Vocals + Music), 4-Stem (Vocals/Drums/Bass/Other), or 6-Stem (Full Separation). The AI Model options are Fast and Quality; these are option labels, not a guarantee of speed or quality. Processing Passes ranges from 1 to 8, with a default of 1.
Use the workspace action to process the file, then inspect the result and verify any download. Listen to the beginning, a busy section, and the ending. For an instrumental, check for vocal traces; for separated stems, check whether parts remain in the wrong track. With 2-Stem separation, a single instrumental or vocal output downloads as WAV; Both Tracks (Separate Files), 4-Stem, and 6-Stem selections download as ZIP archives. Extract a ZIP before opening its tracks in an audio editor.
Settings reference
| Setting | Options | Default |
|---|---|---|
| Extract | instrumental: Instrumental Only vocals: Vocals Only both: Both Tracks (Separate Files) | instrumental |
| Stem Separation | 2-stem: 2-Stem (Vocals + Music) 4-stem: 4-Stem (Vocals/Drums/Bass/Other) 6-stem: 6-Stem (Full Separation) | 2-stem |
| AI Model | Fast: Fast Quality: Quality | Fast |
| Processing Passes | 1 – 8 | 1 |
Formats & limits
| Detail | Requirement |
|---|---|
| Input formats | .mp3, .wav, .ogg, .flac, .m4a, .aac, .wma |
| Output formats | WAV |
Access & cost
| Detail | Requirement |
|---|---|
| Processing mode | Online processing requires sign-in and an internet connection. |
| Online processing | Sign-in required; Internet connection required |
Examples & use cases
A singer preparing to practice over a song.
Make a karaoke practice track
An MP3 has lead vocals mixed with the music. The singer wants an instrumental to practice against and needs to check whether any voice remains in the passages they’ll use.
.mp3 to WAV
- Extract:
- instrumental
- Stem Separation:
- 2-stem
- Choose Instrumental Only and 2-Stem (Vocals + Music) for the vocal-and-music split.
- Listen to the beginning, a vocal-heavy passage, and the ending; check whether any remaining voice is acceptable for practice.
Expected outcome
A WAV instrumental result to review for vocal traces before using it as a practice track.
A musician learning a vocal line for a cover.
Keep the vocal part for cover practice
A WAV recording combines vocals and accompaniment. The musician wants to hear the vocal part separately, including in sections with prominent singing.
.wav to WAV
- Extract:
- vocals
- Stem Separation:
- 2-stem
- Choose Vocals Only with 2-Stem (Vocals + Music) for the vocal-and-music split.
- Check a quiet section and a section with prominent singing to see whether the vocal result suits the practice task.
Expected outcome
A WAV vocal result to review before using it for cover practice.
A music student studying an arrangement.
Study the listed parts in a mixed recording
A FLAC recording contains vocals, drums, bass, and other instruments. The student wants to examine those named parts rather than only a vocal-and-music split.
.flac to WAV
- Stem Separation:
- 4-stem
- Choose 4-Stem (Vocals/Drums/Bass/Other) because its label lists the parts needed for this study task.
- Listen to drum- and bass-heavy passages in the result and check whether the parts are separated as needed. This multi-track selection downloads as a ZIP; extract it before opening the tracks.
Expected outcome
A separated-track result delivered as a ZIP archive to extract and inspect before using it for study.
More ways to use it
- A karaoke host: They want an instrumental from a supported audio file for a sing-along session. An instrumental WAV result to listen through and assess for vocal traces before the session. Extract: instrumental, Stem Separation: 2-stem.
- A musician preparing a vocal edit: They need the vocal part from a mixed recording rather than an instrumental backing. A vocal WAV result to review for the sections needed in the edit. Extract: vocals, Stem Separation: 2-stem.
Troubleshooting
You can still hear a faint voice in the instrumental.
Try this: Select Instrumental Only and use 2-Stem (Vocals + Music) for the vocal-and-music split.
What to expect: The selected result is the instrumental output, but complete vocal removal isn’t guaranteed. Vocals can overlap with instruments or share effects such as reverb, so traces may remain and some instrumental detail may be removed. Listen to quiet sections and vocal-heavy passages to decide whether it suits your use.
You selected Instrumental Only but need the vocal and instrumental parts separately.
Try this: Select Both Tracks (Separate Files).
What to expect: The requested result changes from instrumental only to vocal and instrumental tracks as separate files; this selection downloads as a ZIP.
Privacy
Online processing uploads files. Completed results may remain available for later download. See Privacy Policy for current file-handling and retention details. Check the processing mode before starting.
Frequently asked questions
How do I remove vocals from a song?
Upload one supported audio file, choose Instrumental Only under Extract, and process it in the linked workspace. The configured output is WAV. Listen to the beginning, a busy section, and the ending; check for vocal traces before using it.
Can I keep the vocals instead of removing them?
Yes. Choose Vocals Only under Extract to get the vocal result. To get the vocals and instrumental as separate tracks, choose Both Tracks (Separate Files); that selection downloads as a ZIP.
Which audio files can I upload, and what is the file-size limit?
The listed input extensions are .mp3, .wav, .ogg, .flac, .m4a, .aac, and .wma. You can upload one file up to 100 MB. The extension list doesn’t establish codec or bit-depth support.
Why did I receive a ZIP instead of a WAV?
With 2-Stem separation, a single instrumental or vocal output downloads as WAV. Both Tracks (Separate Files), 4-Stem, and 6-Stem selections download as ZIP archives. Extract a ZIP to work with its separate tracks.
Why can I still hear a faint voice in the instrumental?
Vocals can overlap with instruments or share effects such as reverb, so separation may leave voice traces or remove some instrumental detail. Listen to quiet sections and vocal-heavy passages before deciding whether the instrumental suits your use.
Can I make a karaoke track with this vocal remover?
Yes. The tool is registered for creating instrumentals or karaoke tracks. Choose Instrumental Only under Extract for the vocal-and-music split, then review the WAV result for vocal traces before using it.
Can I separate drums, bass, and other parts too?
Yes. Stem Separation offers 4-Stem (Vocals/Drums/Bass/Other) and 6-Stem (Full Separation), as well as 2-Stem (Vocals + Music). The 4-Stem label names its parts; the 6-Stem label doesn’t specify individual parts.
Can I separate backing vocals from lead vocals?
The listed Extract choices are Instrumental Only, Vocals Only, and Both Tracks (Separate Files); they don’t list separate lead-vocal and backing-vocal outputs. Review the vocal result to see whether it meets your needs.
Can I remove vocals directly from a video file?
The listed inputs are audio extensions: .mp3, .wav, .ogg, .flac, .m4a, .aac, and .wma. The registered inputs don’t establish direct vocal removal from video.
Will removing vocals leave the instrumental’s sound quality unchanged?
The configured output is WAV, but that doesn’t guarantee unchanged sound quality or preserved detail. Separation may leave vocal traces or remove some instrumental detail. Listen to busy and vocal-heavy sections before deciding whether it works for your project.
Can I preview the separated audio before downloading it?
The workflow says to use the workspace action to process the file, inspect the result, and verify any download. It doesn’t specify whether a pre-download audio preview is available, so inspect the result in the workspace and check the downloaded file before using it.
Are uploaded files used to train AI models?
Online processing uploads files. Completed results may remain available for later download. See Privacy Policy for current file-handling and retention details. Check the processing mode before starting.
Does online vocal removal require an account?
This tool runs online and requires sign-in and an internet connection. Review the processing charge shown in the workspace before starting.
How long does processing take?
No processing-time estimate is provided. Use the workspace action to process the file, then inspect the result when it’s offered.