The newly-founded Keyboard Philharmonic is a music charity which aims to enlist music lovers to transcribe opera and classical music scores into a revisable, shareable, digital format, and then give those digital scores away for free. One way to think of it is as a way to bring Mozart and Beethoven’s scores into the digital age. I’m passionate about it, and I’m working hard to get it started.

Classical music and opera are an incredibly important part of our shared cultural heritage, worldwide, and especially in the West. Recordings and background music are how most of us experience this beauty. Some of us hear live performances. A lucky few of us perform opera or classical music ourselves, and it would be great if more did.  Music is an important part of a complete education, and it’s a shame that some schools have become so musically impoverished in recent decades.  But underneath the recordings, the concerts, the professional and the amateur musicians, are music scores. Written music notation, as part of a living cultural tradition, is a crucial reason why we have such a rich legacy of music dating back centuries.

Classical music and opera scores are almost always delivered in printed book form. This is a highly-refined and proven medium for passing music on to the future. But the book is a 16th-century technology for information delivery. Books are expensive to stock and distribute. They are heavy to carry to rehearsal. Economies of scale require large print runs, and make it harder and more expensive to distribute less-known works from the long tail of our heritage. In the 21st century, we deliver information in digital form. The cost to copy, stock, and distribute digital files is essentially zero. There are digital formats for text, still images, moving pictures, and for musical scores.  We can bring the benefits of digital culture to opera and classical music.

A first step is already occurring. Many projects, most notably the Petrucci Music Library / Internet Music Score Library Project (IMSLP), and the Choral Public Domain Library (CPDL), are collecting scanned images of music score books, usually as PDF files. They focus on scores published long enough ago to be in the public domain, free of copyright restrictions, and thus people can freely scan the scores and in turn donate the digital scan files to the public domain (or licence them freely via the Creative Commons or similar.  The untold thousands who have scanned and shared, and the projects like IMSLP and CPDL which have collected and shared onwards, are performing a hugely valuable service.

But these scanned image files are not enough.  Digital information is at its best in a revisable, reusable, symbolic form. It encourages sharing, revising, improving, collaborating. A scanned image of a page is static. It is big and clumsy. It is often hard to read, with dirt and marks making it harder to read. It is often formatted for different page size then the screens of the computers, tablets, and smart phones which the digital century uses. Transcribing the scores into a revisable digital score format allows many new possibilities. We can lay out the score anew, leaving behind the dirt and marks. Notes and lines in symbolic form will usually have much smaller file sizes than the scanned page image. Symbolic digital formats allow software to analyse and connect at the level of melody and harmony, not just of appearance. There exist good formats for music notation, chiefly MusicXML and the Music Encoding Initiative. They aren’t perfect, but they allow us to get started, and they will improve.

The problem is that transcribing music scores into digital notation is labour-intensive. A first pass converts the appearance of music score pages, be they a scanned image file or a physical book, into rough digital notation. Then there are several passes of proofreading against the original score. Then, correcting the layout of the digital score until it is as usable as its older sibling. There are many approaches to the first pass: hand-entering into notation software, connecting a piano keyboard to a computer and playing the notes in, or “Music OCR” by software that can recognise notes from a score image. Regardless, they will have errors, and proofreading passes are necessary. It’s fortunate that scores can be divided among many workers, a few pages to each: many hands make light work. PGDP (partnered with Project Gutenberg) has developed an excellent workflow for transcribing text books. They even have some music score transcription on their list.  The Keyboard Philharmonic will learn from them, and may do well just to recruit volunteers for their process. We will also set benchmarks for quality and editorial standards.

Transcribing music scores into revisable, shareable digital form, then giving those digital scores away, will have both short-term and long-term benefits. In the short term, it’s easy to see that transcribed and properly proofread and typeset scores will provide a clean alternative to the page image scans in collections like IMSLP and CPDL.  Since we give the scores away, we hope they and others will pick up the Keyboard Philharmonic reengravings. More sources for public domain scores means more accessibility. Imagine a secondary school music teacher, looking for an edition of a public domain score which they can use for rehearsals and for theory lessons without fear. Freely-licensed digital scores make that possible.  In the medium term, the digital scores will become a corpus, which will start to strengthen other innovations like digital music stands and tablet score readers, by providing content. In the long term, I believe the biggest benefits will be from enabling musicians to use digital tools to interact with music scores the way authors interact with digital texts, filmmakers with digital video, and software engineers with free software.

So, with all these excellent initiatives and projects already active, why the Keyboard Philharmonic?  It focusses on three distinctive elements.

First, the Keyboard Philharmonic is a vehicle for recruiting people and structuring their work for useful results. We hope to follow the example of Wikipedia. It set up a website, yes, but more importantly it set up an exciting goal that attracted people, and provided structures and social norms that let them work together and create something large and rich. The Keyboard Philharmonic aspires to be just as effective, for music scores.  We don’t aim to compete with repositories, but to supply them with more content. We don’t aim to invent new technologies, if existing technologies do the job.

Second, the Keyboard Philharmonic emphasises revisable, symbolic digital notation formats over scanned image formats. IMSLP, CPDL, and more and more libraries, have the scanned image angle covered. But few repositories are focussing on revisable-form digital scores with high quality and good editorial standards. We believe that if we can get the quality and the content right, the digital score corpus will enable all kinds of exciting innovation.

Third, the Keyboard Philharmonic aims to give its product away freely.  This has both business and ethical grounds. Not many people will want to volunteer their effort to make some company rich. But Wikipedia and many other projects demonstrate that people will bring huge energy, passion, and skill to making something which is free, and which makes the world better. The product must be free if we want to tap the free labour which we believe is awaiting a chance to help. That’s why we operate like a charity (whatever tax authorities eventually say about us). On an ethical level, the product must be free, because what is valuable about these scores is not our efforts at transcription, or the digital nature of the formats. It is the genius and beauty of the music itself. This is our shared heritage. Others have tried to fence off this beauty and heritage in some new form, and extract rents for access. That’s not what the Keyboard Philharmonic is about. We love this music, we respect this music, we want to share this music.

The Keyboard Philharmonic has exciting potential, but right now we are just in the beginning stages. The founders are doing a combination of trial transcriptions to explore techniques, and building the Keyboard Philharmonic structure, and getting the word out. There is not yet a website, just a rudimentary place-holder. But it’s not too early to start following our progress. If you are a pioneer, it’s not too early to start helping.   We welcome your interest. To find out more you can:

If you are a music lover, and want to bring our shared cultural heritage into the 21st century, please join us on this exciting journey.

[Update, Aug 2015: copy-editing.]

5 Comments »