How to Make a Mashup From Two Songs

To make a mashup from two songs, you pull the acapella (the isolated vocal) out of song A with a vocal remover, take the instrumental of song B, match their key and tempo, then layer the two together and export one track. The classic recipe is vocals of one song sitting on the beat of another. Vocal Cut's browser tools handle every preparation step — extraction, key and tempo matching, cleanup — on files you already have, so your audio never leaves your device; the final overlay of the two parts happens in a multitrack editor. Here is the full workflow.

1. Pick two songs that fit

A mashup lives or dies on song choice, not on mixing skill. The two tracks need to be musically compatible before you touch a single tool, and two things decide that: key and tempo.

Key is the set of notes a song is built around. If song A's vocal is in a key that clashes with song B's chords, the result sounds sour no matter how well it's aligned. You can shift one song to fix this (step 4), but small shifts sound far more natural than big ones, so pick a pair that's already close.

Tempo is the speed, measured in BPM (beats per minute). A vocal recorded at 90 BPM stretched onto a 128 BPM beat has to move a long way, and heavy stretching audibly degrades the sound. Songs within about 10 BPM of each other are the safest starting point.

The classic structure is the vocal of song A over the instrumental of song B. Choose the song with the vocal you love as your acapella source, and the song with the groove you want as your instrumental.

2. Get the acapella

An acapella is the vocal on its own: no drums, no bass, no instruments. You almost never get handed one, so you extract it from the finished record.

Open the AI vocal remover — it doubles as an acapella extractor — drop in song A, and keep the vocal output instead of the instrumental. The tool uses a neural model that recognizes which parts of the mix sound like a voice and rebuilds only those, so it works on nearly any commercial track regardless of how it was mixed.

Two honest notes on what to expect. First, it's fast and local, but not instant: separation runs on your own device's CPU, and the first time you use it the model downloads (tens of MB) before processing starts. After that it's cached, and a full song takes real time to process — seconds to a couple of minutes depending on your machine. Second, results vary by song. A clear lead vocal over a sparse arrangement comes out clean; a dense, heavily layered, effect-drenched vocal carries more artifacts. That isn't the tool failing — it's how much overlapping frequency content there is to untangle. See why some songs separate cleaner than others for what to look for.

If you want more control — say the acapella needs the backing vocals removed too — use the stem splitter to break the song into separate vocal, drums, bass, and other parts, then rebuild just the pieces you want. For a walkthrough focused purely on the vocal, see how to isolate vocals from a song.

3. Get the instrumental

The instrumental is the opposite stem: the full backing track with the vocal stripped out. Run song B through the same vocal remover, but this time keep the instrumental output. Full steps are in how to remove vocals from a song.

Now you have the two raw ingredients: song A's voice and song B's beat. The rest is making them agree.

4. Match the key

If the vocal and the instrumental are in different keys, they'll clash. Fix it by finding each song's key, then shifting one to meet the other.

Run both files through the key finder to read their keys. If they don't match, use the pitch changer to shift one — usually the acapella — up or down until it lands in the same key as the instrumental. Move by the smallest interval that works: a semitone or two is nearly transparent, while shifting by four or five semitones starts to sound unnatural.

Vocals are the most fragile thing to pitch-shift. Push too far and they take on a chipmunk or a hollow, underwater quality. How to pitch shift without ruining vocals covers how to keep the shift clean.

5. Match the tempo

Two songs in the same key still won't lock together if their beats drift apart. You need them at the same BPM.

Find each song's tempo with the BPM finder. To bring them together, use the speed changer, which changes tempo without changing pitch — critical here, because you already spent step 4 getting the key right and you don't want to undo it. An old-school turntable pitch-up speeds a song up and raises its key at the same time; modern time-stretching separates the two, which is exactly what you want. The relationship between those three properties is worth understanding, and how pitch, speed, and duration are related explains it.

Getting both key and tempo aligned is beatmatching plus key matching, the core DJ skill behind every clean mashup. Beatmatching and key matching explained breaks down why both matter together.

6. Align and layer

Layering two independent tracks so they play at the same time, then bouncing them to a single file, is the one step the single-purpose browser tools don't do for you. That final overlay happens in a multitrack editor or DAW: drop the prepared acapella on one track and the instrumental on another, then slide them until they lock. The goal is that the vocal's phrasing lands where you want it against the beat, and that downbeats agree — the first strong beat of the instrumental's bar should sit under the corresponding moment of the vocal.

Vocal Cut's job is to get each part ready to drop in. If a section starts in the wrong place, trim the lead-in silence or crop to the part you want with the audio cutter, and build an arrangement — intro, verse, chorus — by stitching sections end to end with the audio joiner before you import them.

7. Balance and polish

Once the two parts are locked, they need to sit together. Usually the raw acapella is either too loud or too quiet against the beat. Inside the editor, set the vocal so it's clearly present but not fighting the instrumental — the vocal should feel like it belongs on top, not pasted over. Then bounce the layered arrangement to a single file.

Bring that bounce back to Vocal Cut for the finishing touches. To even out the overall level, run it through the volume normalizer, which brings it to a consistent loudness. For a fuller, more finished sound — glue, tonal balance, and competitive loudness — run it through audio mastering as the last step.

8. Export

Export the polished mashup. Use WAV if you're keeping it for further editing, since it's lossless, or MP3 if you just want a shareable file. That's the whole chain: acapella from A, instrumental from B, key matched, tempo matched, layered, balanced, exported.

A note on copyright

A mashup is built entirely from other people's recordings, so it's copyrighted material by definition. Making one for practice, learning, or personal creative use is generally low-risk. Publishing, streaming, or monetizing it is a different matter and usually needs the rights holders' permission unless the source material was released for remixing. This isn't legal advice — copyright and fair use when sampling and remixing walks through where the lines tend to fall.

Troubleshooting

The vocal and beat sound phasey or hollow when combined. This is usually a timing issue: the two are close but not exactly aligned, so overlapping frequencies partially cancel. Nudge the vocal a few milliseconds until it snaps into focus.

The low end is muddy. Extracted acapellas often carry low-frequency bleed from the original beat, which then piles onto your new instrumental's bass. A gentle high-pass filter on the vocal, rolling off below roughly 80–100 Hz where voices carry little energy, clears space for the instrumental's low end.

You can still hear ghost instruments in the acapella. That's separation bleed, and it's worse on dense mixes. Start from the highest-quality source file you can find — a lossless WAV or FLAC gives the model far more to work with than a low-bitrate MP3 — and consider the stem splitter for a more surgical result.

The two songs slowly drift out of sync over time. Their tempos aren't truly matched, or one track has slight tempo variation. Re-check both with the BPM finder and re-align, working in shorter sections joined with the audio joiner if the drift is stubborn.

Frequently asked questions

Can you make a mashup for free? Yes. The prep tools — extracting the acapella, removing vocals for the instrumental, finding key and BPM, matching pitch and tempo, and polishing the final bounce — are free in your browser with no account and no upload limit, and the processing happens on your own device, so your audio never leaves it. You layer the two prepared parts together in a multitrack editor for the final mix.

Do the two songs need to be the same key and BPM? They need to end up matched, but they don't have to start that way. Use the key finder and BPM finder to measure both, then shift key with the pitch changer and tempo with the speed changer. Pick songs that are already close, though — small adjustments sound far more natural than large ones.

Why does the extracted acapella sound watery or full of artifacts? AI separation estimates the vocal rather than recovering a true studio stem, so dense, heavily processed, or low-quality mixes leave some smearing and bleed. Starting from a lossless file helps a lot, and sparse arrangements separate far more cleanly than busy ones. It's a limitation of the input, not a bug in the tool.

Is it legal to make a mashup? Making one for personal or creative practice is generally low-risk, but a mashup uses copyrighted recordings, so publishing, streaming, or monetizing it usually needs permission from the rights holders unless the material was released for remixing. This isn't legal advice — check the copyright guide for a fuller picture before you distribute anything.