Ivory - A tool for transcribing piano recordings into editable notation

[ivoryapp]

Hello everyone!

My co-founder and I have spent the past while building Ivory (https://ivory-app.com), a tool for converting piano recordings into editable notation, and I wanted to introduce it here because this community is exactly who we built it for: people who play, study, and write at the piano.

The problem we set out to address is a practical one. Pianists who work at the instrument routinely capture ideas as audio, a phone recording, or a quick take, and then face the slow task of transcribing their playing by ear before they can read, study, or develop the material. Manual transcription of dense voicings, fast passagework, or pedalled textures can take longer than the playing itself.

Ivory is built specifically for that step.

What it does

• Converts a piano recording (MP3 or similar) into notation in the browser: nothing to install.
• Uses a model trained specifically for piano, rather than a general-purpose audio-to-MIDI converter.
• Produces editable output, exportable to MusicXML (Sibelius, MuseScore, Dorico), MIDI (any DAW), and PDF.

The intended workflow is to take a recording, transcribe it in Ivory, then export to your notation software and work from a readable draft rather than starting from nothing.

Piano is one of the harder instruments to transcribe automatically: overlapping voices, sustain pedal blurring note boundaries, fast runs, and up to ten simultaneous notes. General-purpose tools tend to struggle with exactly these cases. By narrowing the scope to piano, we handle chords, fast runs, and pedalling considerably better than the broader tools we benchmarked against.

Some honest limitations:

• Ivory is piano-focused, not intended for full multi-instrument or orchestral arrangements.
• The output is a first draft, so do expect to correct some rhythms, re-bar passages, and make voicing/enharmonic decisions yourself.
• Free tier is capped at 30 seconds of audio per transcription, and 20 transcriptions per month, enough to evaluate properly on real material.

We would really value feedback from the players, teachers, and students here, both on transcription quality and on how it fits your existing notation workflow. Happy to answer questions in this thread!

[essence]

interesting. I tried it out on a short recording of the start of op 109. It thought it was in Eb minor (my piano was out of tune) and difficult to see what the score was doing. But not exactly the easiest                                        music to transcribe.

On the midi playback, you could tell it was the 109.

In my experience, ai can nowadays be a useful starting point, and is improving a lot.