if pitch tracking doesn't work, you should check out william brent's timbreID stuff,

J

On Tue, May 18, 2010 at 11:35 AM, Spencer Russell <spencer.f.russell@gmail.com> wrote:
On Tue, May 18, 2010 at 12:18 PM, Pagano, Patrick
<pat@digitalworlds.ufl.edu> wrote:
>
>
> Hi
>
> I have a project proposal that wants to track pitches from a guitar
> [non-midi] and then translate them into tablature that is readable by
> players. I was looking at fiddle and wondering perhaps if processingmight be
> of a help but I am posting her to hopefully get some pd feedback and perhaps
> a few ideas for implementing this project. Has anyone tried this kind of
> pitch tracking in pure data?
>

Hi Patrick,

I believe that there are quite a few people using pitch tracking for
various things, and I believe that the current best practice for
general pitch tracking is to use sigmund~, as Frank pointed out.

Pitch tracking, especially multiphonic pitch tracking is still an
active research area and not an easy problem to solve, but if you're
restricting yourself to a single instrument you can boost your
accuracy by taking into account the harmonic profile of the
instrument.

Check out the slides at
http://www.ee.columbia.edu/~dpwe/e4896/lectures/E4896-L08.pdf
for some good pitch-tracking info.

-spencer

_______________________________________________
Pd-list@iem.at mailing list
UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list



--
Jaime E Oliver LR

www.jaimeoliver.pe

858 750 0924 (cel)
858 202 1522 (home)
9168 Regents Rd. Apt. G
La Jolla, CA 92037
USA