How is the music extractor API different from a vocal remover?

A vocal remover is usually centered on vocals and no_vocals outputs. The music extractor API is centered on the music stem, so it is a better fit when your product wants the background music bed from a clip, ad, podcast segment, or user upload. Pick the API by the output your user expects. If the user says they need the soundtrack, send them to the music extractor API rather than a karaoke-style flow.

What output does the music extractor API return?

When the task succeeds, the response exposes downloadable audio URLs. If track labels are available, output.tracks can identify the music stem and the no_music remainder. That makes the music extractor API easier to wire into editors, moderation tools, and content libraries. Use those labels for download buttons, timeline lanes, and storage keys instead of relying on file order.

Can I choose the output format?

Yes. The music extractor API supports mp3, wav, flac, aac, and ogg output requests. If your app edits audio later, choose a format that matches your editor or storage pipeline instead of converting files twice.

How is music extractor API pricing calculated?

Pricing uses rounded-up source duration at 200 credits ($0.20) per started minute. A 20-second test clip is billed as one started minute, and a longer source scales by its rounded duration. The music extractor API uses the same async task pricing pattern as the other audio pages. Show the rounded estimate before long uploads so users know the cost before they start.

Where is the music extractor API reference?

The full parameter table and cURL example live at /docs/music-extractor. Create music extractor API jobs with POST /api/v1/audio/generations and poll status with GET /api/v1/tasks/:id. The docs show the model id, audio_url field, output formats, and task response shape. Use that page when you need exact request fields for backend validation or generated client code. The music extractor API docs are the source for accepted fields, pricing unit, and output mapping.

Music Extractor API

Extract background music from mixed audio. The music extractor API returns the music stem and a no_music remainder through one async task.

Price

est$0.2– $0.2per started minute

Try music extraction View music extractor docs

01playground

InputsPOST /api/v1/audio/generations

Source audio*

Drag audio here or click to uploadMP3 / WAV / AAC / OGG · max 20 MB · up to 1 file

MP3 / WAV / AAC / OGG

Splitter

Default auto

Dereverb

Optional

Output format

mp3, wav, flac, aac, or ogg

Outputidle

no output

Run Generate to see results.

02use cases

What you can build

Real workflows powered by this model.

music waveform isolated from mixed audio

01/case

Music extractor API for creator clips

Use the music extractor API when a clip combines speech, music, and room sound. Your app can keep the music stem for editing, cataloging, moderation, or reuse. The music extractor API works well for products that need the bed, not just a vocal split. It gives creator tools a clean way to save the soundtrack without asking users to export from an editor. The music extractor API is also useful when support teams need a repeatable backend job instead of manual audio cleanup.

Extract music

timeline with an extracted music bed ready for editing

02/case

Music extractor API for clean edit assets

Editors often need the bed without rebuilding the whole timeline. The music extractor API gives your workflow a downloadable music track and a no_music remainder from the same source. That pair lets a review tool compare the original mix against the extracted music bed. Teams can preview both outputs before approving a clip, archiving a sound bed, or sending the result to another workflow.

03/case

Batch-friendly music extractor API queues

Submit the music extractor API task, store the task id, and poll later. That flow works for creator tools, media archives, and review systems that process many clips at once. Each completed task returns URLs your app can store, preview, or hand to another service. The music extractor API lets backend queues process long clip libraries without keeping a browser tab open.

03pricing

Pricing

Credit-based — 1 credit = $0.001 USD. Pay only for completed generations.

Mode	Category	Unit	Price
Music extraction
└	Source audio	1 started minute	$0.2 200 credits

05why

Why reAPI

Why this music extractor API uses one surface

The music extractor API submits the music model directly instead of asking users to choose a generic stem. That keeps the page, playground, and docs aligned with one job. Less choice here is intentional because the user already came to extract music.

Music plus remainder

Completed tasks can return a music stem and a no_music track. Your app can save both, preview both, or send either output into the next step. Keep both files when users may later need the original speech without the bed or the bed without the speech. The music extractor API makes that pair available from one task.

Music extractor API output you can reuse

Send an audio_url and keep binary uploads out of your backend request body. The music extractor API handles processing asynchronously and returns downloadable output URLs. Your product can save those URLs with the task record or move them into its own storage after completion.

06quickstart

Integrate the music extractor API in three steps

step 01
Create an API key
Generate a reAPI key from workspace settings. Use it to call /api/v1/audio/generations from your backend.
Open
step 02
Submit a music extraction task
POST model audio-music-extractor with a public audio_url. Add encoder_format when your product needs a specific output container.
Open
step 03
Poll and store the tracks
GET /api/v1/tasks/:id until completed. Save output.audio_urls or use output.tracks to map the music and no_music files.
Open

07faq

Frequently asked questions

Common questions about this model.

The music extractor API is an async audio endpoint for extracting the music stem from mixed audio. It is designed for sources where music is mixed with voice or other content. A completed music extractor API task can include both music and no_music outputs for downstream editing.

09api reference

docs/api/music-extractor

API reference

Drop-in code and the full parameter table.

View full reference

curl https://reapi.ai/api/v1/audio/generations \
  -H "Authorization: Bearer $REAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"audio-music-extractor","audio_url":"https://cdn.example.com/mix.wav","encoder_format":"mp3"}'

start building

Ready to ship?

Try it in the playground or grab an API key to integrate now.

Try music extraction Get API key

What you can build

Real workflows powered by this model.

01/case