lilypond-user
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Chords and what they mean


From: mskala
Subject: Re: Chords and what they mean
Date: Thu, 17 Sep 2015 21:10:47 -0500 (CDT)
User-agent: Alpine 2.20 (LNX 67 2015-01-07)

On Thu, 17 Sep 2015, Blöchl Bernhard wrote:
> <c es f g> I would call (always depending on the context) Cmadd4, F7sus2/C,
> D#6no5add2/C.

> Always consider the harmonic context!!!

What exactly does the "harmonic context" mean?  What would be specific
examples of contexts where it could make sense to call this set of notes
Cmadd4, and contexts where it would be better to call it F7sus2/C, etc.?

If we are hoping to teach a computer program, i.e. LilyPond, to assign
correct names to chords, then we have to really say what the
considerations are that lead one name to be correct over another.

I'd like to think it could be as simple as looking at the current key
signature for a clue, but I realize that's only likely to actually give
the right results in limited cases, and to highlight whatever errors
remain.  Maybe a smarter solution could involve a language model (hidden
Markov, context-free grammar, etc.) that could assign a likelihood to each
chord name for a set of notes depending on the ones before and after it -
like the standard techniques for determining which words in a sentence are
nouns and verbs and so on, even though any single word may be ambiguous.

It sure would be useful if there were a system of descriptive names for
sonorities *in isolation* that could be understood as giving a single name
to the set of notes not commenting on anything else except which notes are
and are not present, but I realize that's not the information
conventional chord names are intended to convey.

-- 
Matthew Skala
address@hidden                 People before principles.
http://ansuz.sooke.bc.ca/

reply via email to

[Prev in Thread] Current Thread [Next in Thread]