Part of our guide to transcribing a full band.
A full band on one recording is the hardest thing to transcribe, because everything is happening at once and the instruments hide each other inside the mix. There are now two good ways through it. If you have the song as separate tracks, transcribing each stem on its own gives you the cleanest parts. If you only have the finished mix, Songscription's Multi-Instrument mode transcribes the whole band in one pass and hands you the multi-track result directly.
This guide covers both: why separate tracks give the cleanest read, how to transcribe each stem, what Multi-Instrument gives you from a mix, and how to assemble a real multi-instrument score. It is the separate-parts companion to our guide on condensing a whole band into one playable part.
Multi-Track vs a Mixed Recording
A mixed recording is the finished song: every instrument blended into a single stereo file, the way you hear it on a release. Multi-track audio keeps the instruments on separate tracks, or stems: a vocal stem, a bass stem, a drum stem, a keys stem, and so on. Stems let you hand each model a single, clean instrument, which is the condition under which transcription is most accurate. A mix asks the model to sort out the instruments as well as transcribe them, which is harder, but it is no longer a dead end: Multi-Instrument is built for exactly that job.
Why Stems Give the Cleanest Parts
In a mix, instruments occupy the same frequency space and mask each other: the bass sits under the keys' left hand, cymbals wash over everything, a pad blurs the harmony. That overlap is the main thing that lowers accuracy, the same masking problem discussed in transcribing multiple instruments at once. An isolated stem removes the competition. The bass model hears only bass; the piano model hears only piano, so the piano's polyphonic detail stays intact (our explainer on polyphonic piano transcription covers that detail). Cleaner input gives you cleaner notation.
Get or Make Your Stems
If you want the stem route, you get stems one of two ways, depending on whether you made the recording.
- Export them from the session. If you recorded or produced the song, every instrument is already on its own track in your DAW. Bounce each one to its own audio file and you have perfect stems.
- Separate them from the mix. If all you have is the finished song, a stem-separation tool can split it into approximate parts like vocals, bass, drums, and other. Separated stems are not perfectly clean, but they can sharpen a single part on a dense recording.
If you only have the mix and do not want to separate it, skip ahead: Multi-Instrument works from the mix directly.
Transcribe Each Track With the Right Model
With stems in hand, transcribe them one at a time, choosing the model that fits each instrument. Run the keys stem through the piano model, the bass stem through the bass model, the drum stem through the drum model, and so on. Each transcription is a clean, single-instrument read, with the hands split on the piano part and chord symbols where the harmony is clear. Our guides to transcribing a bass line and drums with AI cover the per-instrument settings, and producers working back toward MIDI will want our guide to turning audio into MIDI to recreate sounds. Start any of these from audio to sheet music.
Only Have the Mix? Use Multi-Instrument
Pick "Multi-Instrument" in the instrument picker and upload the mix as a file or a YouTube, TikTok, or Instagram link. Songscription transcribes every instrument in one pass and returns a multi-staff score with a separate part for each one it hears, from keys, guitars, and bass to strings, brass, saxophones, and drums, plus multitrack MIDI that opens in Ableton, Logic, FL Studio, or GarageBand. Keyboard tracks come split into left and right hands. Check each part against the recording and tidy rough spots in the built-in editor; accuracy is highest on clean recordings. A choir comes back as one voice track rather than separate soprano, alto, tenor, and bass parts. For the full walkthrough, see transcribing a full band recording, and for when to choose one pass over per-instrument, see one-pass vs per-instrument transcription. For MIDI only, start from audio to MIDI. If you only need the notes for your DAW, audio to multitrack MIDI covers that route.
Assemble the Parts, or Reduce to One
If you transcribed stems separately, export each part as MusicXML and combine them in a notation program, one instrument per staff, lined up bar by bar. If you used Multi-Instrument, the score is already assembled, and you can swap in a cleaner single-instrument transcription of any part that matters most. For a simpler outcome, keep only the parts you need, such as a rhythm-section chart, or use lead sheet mode for the vocal and chords. And if you just want something to play, piano cover mode condenses the whole mix into one playable part. Stems for the cleanest parts; Multi-Instrument for every part from a mix; one condensed score when you just want to play.
Frequently Asked Questions
Can I transcribe multi-track audio to sheet music?
Yes. If each instrument lives on its own track or stem, transcribe them one at a time with the model that fits each, then combine the results into a full score with a staff per part; this gives the cleanest parts because every model gets an isolated source. If you only have the mixed song, Songscription's Multi-Instrument mode transcribes a mixed recording in one pass and returns a multi-staff score and multitrack MIDI with a separate part for each instrument it hears.
Should I transcribe a mix or separate stems?
Use stems when you have them and need each part as clean as possible, because an isolated stem gives the model one instrument to read with nothing masking it. Transcribe the mix when you do not have stems. Multi-Instrument turns the mix straight into separate parts in one pass, and single-instrument mode can isolate any one instrument from the mix directly. Stems are optional in both cases; they are an accuracy boost, not a requirement.
How do I get a separate part for each instrument?
You have two routes. From the mix, choose Multi-Instrument in Songscription and upload the recording; you get a separate part for each instrument it hears in one pass, with every keyboard part split into left and right hands. From stems, transcribe each track on its own with the matching model, then combine the parts in a notation program by exporting MusicXML and lining them up bar by bar. Either way you end up with a multi-staff arrangement.
What if I only have the final mix, not the stems?
Use Multi-Instrument. It transcribes a mixed recording in one pass and gives you the multi-track result directly, as a multi-staff score plus multitrack MIDI for your DAW. Some parts may need cleanup in the built-in editor, and accuracy is highest on clean recordings. If one part needs to be as clean as possible, transcribe that instrument in single-instrument mode, which isolates it from the mix without stems. If you just want one playable part, use piano cover or lead sheet mode.
