Comparison · Music & audio
Vocal Remover vs Music Extractor
Vocal Remover vs Music Extractor: one keeps the vocal track, the other the instrumental bed. Both take one audio URL and bill per minute. Which you need.
What each one accepts
| Vocal Remover | Music Extractor | |
|---|---|---|
| Vendor | Kadraris | Kadraris |
| Type | Audio | Audio |
| Modes | vocal-isolation | instrumental-extraction |
Rates and clip costs
| Model | Variant | Rate |
|---|---|---|
| Vocal Remover | - | $0.50 / min |
| Music Extractor | - | $0.50 / min |
Charged from a prepaid USD balance, no taxes added. A failed generation is $0.
These two are mirror images. Vocal Remover keeps the voice and drops everything else; Music Extractor keeps the music behind the voice. So the question is not which is better, but which half of the recording you want to hold on to.
When to pick Vocal Remover
- An acapella for a remix or mashup.
- A clean vocal for transcribing lyrics or timing them.
- Studying how harmonies and backing vocals are arranged.
When to pick Music Extractor
- A karaoke or backing track from a finished song.
- Recovering the music bed of a video whose project files are gone.
- Removing a voiceover so it can be recorded again in another language.
- Practice tracks for singers.
Differences you will notice
- Price. Vocal Remover costs $0.50 per minute, Music Extractor $0.50 per minute. Both bill per started minute of source audio.
- What you get. Vocal Remover returns one vocal file with lead and backing voices together. Music Extractor returns the instrumental bed.
- Typical sources. Vocal Remover is built around songs. Music Extractor also handles vlogs and podcasts where speech sits on top of music.
- Settings. Both take one field,
audio_url, and run with default settings. Dereverb, engine choice, lead/backing split and output formats other than the default are not available in Kadraris yet.
How to decide
Decide what stays. The voice: Vocal Remover. The music: Music Extractor. For karaoke you can run the same file through both. If you need drums and bass on separate tracks, neither is enough; use Multistem Splitter. Feed both the best-quality file you have, trim silence at the start and end, and host it at a public http(s) address with no login. The workflow is in remove vocals from a song with AI; rates on /pricing and in the cost calculator.
Questions
Can I pick the output format?
Not yet. Both return the default format of the engine.
Will either tool clean up echo on the vocal?
No. Dereverb exists in the engine but is not exposed, so room reverb stays in the result.
Am I free to publish what comes out?
Only with rights to the original recording. The tools process any audio you link, but that does not give you permission to release it.