Transcribe a song by ear

Isolate the part you are trying to write down and hear it without everything else on top.

Drop your song here

or click to choose · mp3, wav, flac, m4a · up to 50 MB

What do you want?

VocalsTake the singing out, or keep only the singing.You get 2 files:VocalsSong without vocalsDrumsGet the drums alone, or play along without them.You get 2 files:DrumsSong without drumsBassGet the bassline alone, or play along without it.You get 2 files:BassSong without bassEverythingEvery part on its own, and the song minus each part.You get 8 files:VocalsDrumsBassOther (guitars, keys, synths)Song without vocalsSong without drumsSong without bassSong without the rest

The hard part of transcription is rarely notation. It is hearing — picking one line out of a mix where four instruments are competing, and where the thing you want is buried under a cymbal or doubled by a synth an octave up.

Separating the recording first removes that obstacle. You still do the listening and the writing; you just do it on a recording where the part you care about is not fighting for space.

How this changes the work

Transcribing from a full mix involves a constant background task of separating parts in your head before you can even begin analysing them. That is genuinely useful ear training, and if training your ear is the goal, you should keep doing it the hard way.

If the goal is an accurate transcription — a chart for a band, an arrangement, a piece of analysis — then spending your attention on extraction rather than on the music is a poor trade. Separate first, and put the effort into getting the notes and rhythms right.

The most common use is the middle ground: work by ear from the full mix, and check the passages you are unsure about against the isolated stem.

What each stem is good for

Bass is the biggest single win. Basslines are the most commonly mis-transcribed part of any record simply because they are hard to hear, and an isolated bass stem removes all ambiguity about the notes.
Drums soloed make ghost notes and hi-hat articulation audible that no amount of careful listening to the full mix will reveal.
Vocals isolated expose the actual rhythm of a melody — how far behind the beat a phrase sits, and where the singer breathes.
Chords are the hard case, because guitar and keys arrive together in one stem. It still helps to have them without drums and bass, but it will not hand you a clean single instrument.

Questions

›Will this give me the notes or a score?

No. It separates audio — it does not detect pitch, generate notation or produce MIDI. The transcription is still yours to do, on a recording that makes it possible.

›Which option should I choose for transcribing?

Everything, so you have every part available. You rarely know in advance which stem will answer the question you are stuck on.

›Can I use these for teaching or coursework?

Separating a recording for study and analysis is ordinary private use. Distributing the separated files is a different matter and depends on the rights in the original recording.

Where it struggles

Harmony is where this helps least. Because guitars, keys and strings share a single stem, working out an exact chord voicing in a dense arrangement remains as difficult as it ever was. Rhythm-section transcription is transformed; harmonic transcription is only somewhat improved.