Report Bug / Feature Request
AI Music Generator AI Song Generator AI Music Video Maker AI Beat Maker AI Lyrics Generator AI Jingle Generator AI Karaoke Maker Vocal Remover More →

Instrumental Extractor

Commercial use OK 380+ models No watermark No sign-up needed
Model:
+ GPT-5, Claude, Gemini

Drag and drop an audio file here, or click to browse

Supports MP3, WAV, FLAC, OGG up to 50MB

No file selected
Upload a song to separate it into individual stems: vocals, drums, bass, and other (instruments). Each stem gets its own player and download button.
~1,000 tokens base + ~50/sec of audio (~3K for a 1-min track)
Separated Stems

What You Can Do

Extract Vocals

Isolate the vocal track from any song. Perfect for creating acapellas, remixes, or karaoke versions.

Get Instrumentals

Remove vocals to get a clean instrumental track. Use it for background music, covers, or practice sessions.

4 Separate Stems

Split into vocals, drums, bass, and other instruments. Each stem gets its own audio player and download button.

Multiple Formats

Upload MP3, WAV, FLAC, or OGG files up to 50MB. Works with any audio format and quality level.

Fast Processing

AI-powered stem separation runs on our GPUs. Most tracks are processed in under a minute.

Download Each Stem

Every separated stem is available for individual download. Get just the vocals, just the drums, or all four stems.

How to Use

Upload your audio

Drag and drop an audio file or click to browse. Supports MP3, WAV, FLAC, and OGG up to 50MB.

Click Remove Vocals

Hit the button and our AI will analyze the audio and separate it into individual stems using deep learning.

Preview each stem

Listen to the separated vocals, drums, bass, and other instruments individually. Each stem has its own audio player.

Download your stems

Download any or all of the separated stems. Use them for remixes, karaoke, music production, or practice.

Token Costs

ActionTokens per useOutput
Stem Separation~50/sec4 stems: vocals, drums, bass, other

Free.ai has 400+ AI tools beyond vocal removal.

Explore All Tools Get More Tokens
Advanced options
Result
Tokens running low. Get More Tokens
Want better results? Premium models (GPT-5, Claude, Gemini) deliver higher quality. View Plans

❤️ Love Free.ai? Tell your friends!

Sign up to get a referral link and earn 25,000 tokens per friend.

Want more? Sign up free for 5K tokens/day + 10K bonus
Sign Up Free

Processing your request...

Pull the instrumental track out of any song with AI. Drop in MP3 or WAV, get a clean instrumental in seconds. Powered by Meta Demucs (MIT). No sign up.

How to Use Instrumental Extractor

1
Enter your input

Type text, upload a file, or describe what you want. No account needed.

2
Click generate

Our AI processes your request in seconds using the best open-source models.

3
Download & share

Download, copy, or share your result. Free for personal and commercial use.

Use this tool via API

Automate this tool from your own code. OpenAI-compatible REST endpoint, Bearer-token auth, no extra SDK required. Token costs match the web interface.

API Documentation Get API Key 
curl -X POST https://api.free.ai/v1/music/generate/ \
  -H "Authorization: Bearer sk-free-..." \
  -H "Content-Type: application/json" \
  -d '{"prompt": "upbeat synthwave instrumental", "duration": 30}'

Related Free AI Tools

AI Music Generator
AI Song Generator
AI Music Video Maker
AI Beat Maker
AI Lyrics Generator
AI Jingle Generator
AI Karaoke Maker
Vocal Remover

Instrumental Extractor — FAQ

It pulls the instrumental track — drums, bass, melody, everything except vocals — out of any song you upload. Same Demucs source-separation model used by Moises, Splitter.ai, and PhonicMind, but free and unlimited.

We run Meta's Demucs (MIT license) on your MP3/WAV. It separates the audio into four stems — vocals, drums, bass, other — and we mix the three non-vocal stems back together. Output is a clean instrumental WAV.

MP3, WAV, M4A, FLAC, OGG, AIFF, and most other audio formats ffmpeg can decode. Max upload size is 50 MB (about 5 minutes of 128 kbps MP3 or 3 minutes of 320 kbps).

About 6-15 seconds per minute of audio on our A100. A typical 3-minute song extracts in 25-40 seconds end-to-end including upload + download.

Excellent on well-mixed studio tracks. Acceptable on dense pop / rock / hip-hop. Weaker on songs where the vocal sits very low in the mix or is heavily processed (heavy reverb, doubled vocals, vocoder). For those tracks, expect some residual vocal in the output.

Personal use (practice, sing-along, study) generally falls under fair use. Public performance, distribution, or commercial release of the extracted instrumental WITHOUT a mechanical or sync license is not allowed. You're responsible for how you use the output.

About 3,000 tokens per song, regardless of song length (it's per-request, not per-second, because Demucs is compute-heavy and we're budgeting for the worst-case 5-minute case). Free daily-pool tokens cover most users.

Yes — use /music/stem-splitter/ to get all four stems (vocals / drums / bass / other) as separate files in a ZIP, or /music/isolate-vocals/ to get just the acapella.

Same Demucs model, same quality. Moises charges $4/month for a small monthly quota. LALAL.AI and Splitter charge per-track. Free.ai is pay-with-tokens (or daily-free-pool), no monthly subscription, no upload caps.

WAV at 44.1 kHz, 16-bit stereo by default. MP3 download is also available via the format dropdown if you want a smaller file for sharing.

On the free tier one at a time. Signed-in users can queue multiple via /batch/ — drop a folder of MP3s, get back a folder of instrumentals. API is also available for bulk automation.

Yes — POST an audio file to /v1/audio/separate/ with `mode=instrumental`. Returns a job_id you poll for completion + the WAV URL when done. Docs at /api/.

Sign up free for 10,000 tokens

Create Free Account

No credit card required

How would you rate this tool?

Love Free.ai? Tell your friends!