[PD] rank, beginner (actually haven't even begun) with a question

Andy Farnell padawan12 at obiwannabe.co.uk
Tue Apr 22 04:08:52 CEST 2008


On Mon, 21 Apr 2008 16:05:47 -0400
"Jeffrey Atkisson" <jeffrey.atkisson at gmail.com> wrote:

> Ok so this can happen in real time?


Yes, well depending on the available CPU, all objects like filters
in Pd work in real-time.

> I would want to do it with any note or pair of notes.
> I am talking about the possibility of using this concept to tune a piano.
> I imagine having an application that aurally re-enforces certain aspects of
> the notes
> making it easier to hear what you need to while tuning.


Sounds plausible.

> A big part of tuning involves analyzing two notes played together;
> what the tuner is actually listening to are those note's <coincident
> partials>.

Remember that two different notes can share coincident partials,
hence harmony/consonance. This is especially true of strings that
vibrate in all harmonic modes.


> In equal temperament the frequency of those coinciding harmonics turn out to
> differ by a few hertz which creates beating, as I'm sure you would know.
> If one was able to aurally isolate that beating it would make tuning a lot
> easier.

The beating will be at a low frequency, just a few Hertz. It may be more 
difficult to isolate these low frequencies and you wouldn't be able to 
directly hear them as they are below the frequency threshold for being
heard as sound. You could however use them to modulate a third source
like noise.



> So is it feasible to have an application that does this fairly automatically
> from interval to interval as one tunes?  It would have to be able to isolate
> that beating as it is changed by tuning the notes in real time.

I think so. Perhaps some of the guys who understand pitch perception better
will have something to say here. It seems like you need to track the fundamental
so Hardoffs suggestion of employing [fiddle~] is good. 


> 
> I know I'm not explaining this very concisely but I would appreciate it in
> someone was willing to hash over this with me.

Hope you are okay with this posted back on list as I think you will get
many more useful responses now the context is clear.



> 
> Jeff
> 
> On Mon, Apr 21, 2008 at 3:34 PM, Andy Farnell <padawan12 at obiwannabe.co.uk>
> wrote:
> 
> >
> >
> >
> > Hi Jeff,
> >
> > Something as simple as a narrow band filter could achieve
> > what you describe if you know which piano note you will play.
> > Otherwise you can do it with FFT and isolate a particular
> > harmonic.
> >
> >
> >
> > On Mon, 21 Apr 2008 15:21:24 -0400
> > "Jeffrey Atkisson" <jeffrey.atkisson at gmail.com> wrote:
> >
> > > Hi,
> > > can someone please enlighten me re: what I am calling real time DSP.
> > > How possible would it be to create following scenario.
> > > 1. create real world acoustic sound (e.g. piano note)
> > > 2. mike that sound
> > > 3. monitor sound over headphones
> > > 4. analyze and process according to analysis in real time (immediate).
> > > for example
> > > play note
> > > isolate per instructions an upper partial/harmonic in that note
> > > and equalize (or whatever you would call it) to have just that partial
> > be
> > > heard.
> > > So that in real time you are hearing just that harmonic.
> > >
> > > Thanks in advance if anyone is willing to answer this.
> > >
> >
> >
> > --
> > Use the source
> >
> > _______________________________________________
> > PD-list at iem.at mailing list
> > UNSUBSCRIBE and account-management ->
> > http://lists.puredata.info/listinfo/pd-list
> >
> 


-- 
Use the source




More information about the Pd-list mailing list