Re: [Rosegarden-devel] Pitch-tracker: release time?

[Nicholas B. <n.j...@el...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


On 20 Jun 2006, at 4:45 pm, D. Michael 'Silvan' McIntyre wrote:

> ...
> They play into the microphone, and it compares their timing, 
> intonation, and
> gross pitch against a baseline, reporting results, and giving them a 
> score
> for how well they completed the exercise.  It's intended to encourage 
> them to
> keep whacking at something until they truly master it, and to give them
> rewarding positive feedback when they do.
>
That's exactly what the pitch tracker gets used for! The notation view 
chugs away and the singer sings (or whatever), then the error graph 
appears underneath it. It does rely on the timing being right though. 
Which is to say, if you don't keep up, it just compares you with a 
different note and the graph tells you you're wrong. This is mostly 
because it's aimed at conservatory standard musicians, and these guys' 
sight reading is beyond belief. They read music like I read English. 
Some of them can look at a string quartet score for example, and just 
hear it 100% accurately. Or even more complex things. Sit them at a 
piano and put a full orchestral score in front of them, and they just 
play it (or something which conveys the essence of it at least). It 
comes out of having to hire a musician for a symphony orchestra and 
expecting them to play a piece of new music with minimal rehearsal.

I've been told by some freelance woodwind players I know that in fact, 
due to the extreme tightfistedness or arts funding in the UK and the 
consequent lack of rehearsal time, UK players are in demand in Europe 
when they need an orchestral member fast (illness, whatever) because 
their sight-reading is so good!

> I haven't been paying the slightest attention to your microtonal 
> foolery (my
> microtones are not intentional! :D ), but if your pitch tracker could
Same here. Funny how open strings on the 'cello keep drifting out of 
tune :)
> conceivably be adapted at little expense to serve a similar purpose, I 
> think
> you just captured my interest!  I've been thinking about trying this 
> software
> under WINE and/or digging up a copy of Windows.  I think it would 
> benefit
> *me*, because the local community band's schedule is grossly 
> incompatible
> with mine, unless I change careers entirely, and so I'm stuck playing 
> with my
> young son, and no one else.  The blind leading the blind, there.  We 
> could
> use help in the form of an impartial analysis.
Beat tracking is a whole different thing. For the pitch tracking part 
though, it should work. You can also record your performance to an 
audio track and then to a post performance review (you select the MIDI 
and the audio, then ask for the graph. Dougie's manual explains how it 
works if memory serves)

>
> Maybe there's something to all that.  Though doing a full-on 
> Rosegarden-based
> computer assisted band method/trainer for Linux seems a bit 
> farfetched, and
> insanely ambitious for a man with very little ambition.  Maybe there's 
> a
> collaborative spin-off project in there somewhere.  Rosegarden K12 or
> something.  I surely couldn't do a decent job of it alone, but I think 
> we do,
> as a project, have an interest in being used for educational purposes. 
>  Most
> particularly in poor countries, where something that kind of mostly 
> works is
> far better than something you can't remotely afford.
>
We've been handing out Knoppix-based CDs at all opportunities for 
people to try, in an attempt to ween them away from certain software 
monopolies I could mention. Works for a lot of them!

Nick/.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (Darwin)

iD8DBQFEmB76Fo+kGmUnzkQRAv+kAJ9C857KYIwr7mewGH3qWwz4qquHIgCggNqB
PN8ouzNuihuxTpDmA77CqBk=
=9eP4
-----END PGP SIGNATURE-----