Your community choir has a rehearsal video from last month, and the pianist is away this week. You want to rehearse the harmony line against the recorded accompaniment, but the video also carries the full choir singing. If you can separate vocals and music from video, you can make a rehearsal version where the recorded voices sit lower while the piano and the guide melody stay audible. This article walks through one evening rehearsal with that video as the source, stopping at a few checkpoints so you can decide what to adjust next.

ShengXi Vocal Remover lets you import a video from Photos or a local audio file and process it on the device. It separates vocals and accompaniment, offers fast and fine processing modes, and lets you preview a mix with independent vocal and accompaniment volume controls. The source files are not uploaded to a server. What the app produces depends on how the source was recorded, mixed, and compressed, so results differ from one file to the next. Also, the app does not give you rights to the music or video; make sure you have permission to process it.

Before You Press Anything: Define the Rehearsal Goal

The choir’s goal for this session is modest: a single practice track for the alto section, with the recorded altos lowered enough that the live singers can hear themselves, while the piano remains clear. Write that goal down in one sentence, because every later adjustment should serve it.

Next, listen to the source video once without touching any tool. Ask:

  • Is the accompaniment mostly one instrument, or a dense mix?
  • Is there talking, clapping, or room echo over the music?
  • Which part of the recording matters most for this rehearsal?

A clean recording with a simple backing track will usually separate more usefully than a noisy room recording with many overlapping voices. Knowing this in advance keeps expectations honest.

Import the Video and Choose a First Mode

Bring the rehearsal video into the app from Photos. The extraction, separation, and export all happen on the device, so nothing needs to be sent away first.

For the first pass, choose fast processing. The point is not the final rehearsal track; it is to hear quickly whether the separation idea works for this particular video. If the fast pass already gives a usable balance of voices and piano, you may not need more. If the voices smear into the piano or the accompaniment sounds thin, plan a second pass with fine processing.

Checkpoint 1: Play the first result from start to finish. Decide whether the problem is the separation itself or simply the volume balance. Those need different fixes.

Balance the Two Channels for the Alto Rehearsal

In the mixed preview, bring the vocal channel down and the accompaniment up. Listen for the moment the recorded altos stop covering the live singers but the guide melody is still recognizable. That balance is personal to your group; there is no single correct setting.

Try a short loop of the trickiest passage — for example, the entry after the bridge — and rehearse it twice with the current mix. Then adjust only one control at a time:

  1. If the live singers still cannot hear themselves, lower the vocal channel a little more.
  2. If the piano gets buried, raise the accompaniment instead of pushing the vocals to zero.
  3. If the entry is unclear, try fine processing rather than more volume changes.

Keep notes on which setting you landed on, because the same balance may not suit a different section of the piece.

Decide Whether to Reprocess or Accept the Result

Some source videos simply resist clean separation. If the voices and piano remain entangled after a fine pass, the recording itself may be the limit — a single microphone in a resonant room captures everything together. In that case, accept a rougher rehearsal track, or choose a different passage where the balance works better.

Also check the ends of the file. Separation can leave faint artifacts at the very start or during a spoken introduction. For a rehearsal track, trimming attention is often more useful than chasing perfection.

Checkpoint 2: Compare the fast result and the fine result side by side on the same 20 seconds. Choose the one that serves the singers, not the one that sounds more impressive in isolation.

Export, Share, and Store the Rehearsal Version

When the mix is settled, export it and share it with the alto section. Recent-history storage keeps earlier attempts available, so you can return to a previous balance if the new one does not work in the rehearsal room. When the session is over, use the cleanup option to remove takes you no longer need.

One final reminder before the file circulates: confirm the choir has permission to use this recording and the underlying music. The app processes audio but does not grant any rights to the material.

What This Evening Established

By the end of the session, you have one exported rehearsal track, a note about which processing mode and balance worked, and a clearer sense of which parts of the recording separate well and which do not. That knowledge is the real outcome. Next time the pianist is away, you can start from the previous settings, re-check them against the new video, and spend the saved time actually rehearsing.

Keep the loop simple: state the goal, take a fast pass, balance with one change at a time, decide whether a fine pass is worth it, then export and share. Repeat it and the workflow becomes routine.