[Smt-talk] RN analyzers

[Dmitri Tymoczko]

Thanks Steve!

Last year I had a student working with the CC code -- he implemented a cool recursive system whereby the program went through the chorales and did a literal analysis.  It then went through the chorales again, and using the first-pass probability data, tried to improve its first-pass analysis -- asking "is this iii6 chord really a consonant suspension into the true harmony, which is V."  It then repeated this process 4 times -- by the end it could analyze the chorales better than a generic Princeton undergraduate, assuming it got itself in the right key.  (My kid didn't try to mess with CC's "key finding" algorithm.)

(Yes, there's apparently no end of undergraduates who want to build RN analyzers.)  Last year, CC couldn't really handle keyboard textures -- perhaps its improved in the meantime; I'll write to Rick and check this out.

My guy this year has built his own RN analyzer by scratch; I think it's already almost but not quite as good as CC ...

DT


On Feb 23, 2012, at 6:51 PM, Stephen Lett wrote:

> Dmitri,
> 
> Not sure if you've heard about Chorale Composer. I just contacted my friend that's been developing the software and I got this response after asking if the software had your requested capabilities:
> 
> "Yeah, Rick [Taube]'s "official" response to this was: 'most of it. it can import xml and the analysis is generic enough to work right now, but there is still the issue of non-harmonic tone analysis with multi tone "voices"'"
> 
> If interested, see link below
> http://camil.music.illinois.edu/software/choralecomposer/
> 
> 
> Steve _______________________________________________
> Smt-talk mailing list
> Smt-talk at lists.societymusictheory.org
> http://lists.societymusictheory.org/listinfo.cgi/smt-talk-societymusictheory.org

Dmitri Tymoczko
Associate Professor of Music
310 Woolworth Center
Princeton, NJ 08544-1007
(609) 258-4255 (ph), (609) 258-6793 (fax)
http://dmitri.tymoczko.com