How Stem Splitting Works
Stem Splitter separates a single audio file into cleaner individual parts so you can keep, remove, or reuse the elements you need. It is useful for isolating vocals, creating an instrumental, preparing vocals for other Vocalist.ai tools, or breaking a full mix into core instrument groups.
Stem Splitter can isolate vocals, instrumentals, reverb, or grouped mix elements depending on the mode you choose.
What You Can Do With It
- Isolate vocals from the a full mix for remixing, editing, or practice
- Create an instrumental from a full song
- Remove reverb from a vocal before sending it into Transform (best practice for a non-dry vocal)
- Split a mix into grouped stems for rough arrangement or production work
Available Modes
Vocals + Instrumental
This is the dedicated two-stem mode. It creates:
- Vocals
- Instrumental
Use this when your main goal is the cleanest possible vocal isolation or backing track removal.
Remove Reverb
This mode focuses on vocal cleanup. It creates:
- No Reverb
- Reverb Only
Use this when you have a vocal with ambience, room sound, or added effects and you want a drier vocal to process further.
6-Stem Mix Split
This mode breaks a full mix into broader musical groups. It creates:
- Bass
- Drums
- Other
- Vocals
- Guitar
- Piano
Use this when you want more production flexibility across a full song. Because it is separating more elements at once, the vocal stem here will usually be less precise than the dedicated Vocals + Instrumental mode.
How To Use Stem Splitter
- Open the Stem Splitter from your dashboard.
- Upload an audio file.
- Choose the mode that matches the result you want.
- Click Split Stems.
- Wait for processing to finish, then preview and download the generated outputs.
Choosing The Right Mode
- Choose Vocals + Instrumental for the best vocal isolation.
- Choose Remove Reverb when you need a cleaner, drier vocal for editing or transformation.
- Choose 6-Stem Mix Split when you need grouped stems from a whole mix. If you will be needing the vocals, we highly suggested running the dedicated vocal model for the vocal stem.
Best Results Tips
- Start with the cleanest source file you have.
- Full songs with heavy effects, stacked harmonies, distortion, or crowd noise are harder to separate cleanly.
- If you only need vocals or an instrumental, use the dedicated two-stem mode instead of the 6-stem mode.
- If your goal is voice transformation, try Remove Reverb first when the source vocal sounds wet or roomy.
Important Notes
- Output quality depends heavily on the source mix.
- Due to the nature of AI, results may be unexpected.
- Some bleed between stems can happen, especially in dense mixes.
When To Use Stem Splitter Before Transform
Stem Splitter is often the best prep step before using Transform. If your input contains backing music, room reverb, or other material competing with the lead vocal, splitting first can give you a cleaner source and a better final result.