Reply by July 11, 20062006-07-11
```Andreas Huennebeck wrote:
> Not a reference, but source code from my libAHmath library:
>
>
> The function ReduceSpectrum::smoothLogXScale() in ReduceSpectrum.c does
> smoothing for any factor.
>
> bye
> Andreas

Andreas, thanks! I'll dig in.

Andrew

```
Reply by July 11, 20062006-07-11
```a@gaydenko.com wrote:

> Does anybody know a ref(s) to free description of fractional octave
> smoothing?

Not a reference, but source code from my libAHmath library:

The function ReduceSpectrum::smoothLogXScale() in ReduceSpectrum.c does
smoothing for any factor.

bye
Andreas
--
Andreas H&#2013266172;nnebeck | email: acmh@gmx.de
----- privat ---- | www  : http://www.huennebeck-online.de
Fax/Anrufbeantworter: 0721/151-284301
GPG-Key: http://www.huennebeck-online.de/public_keys/andreas.asc
PGP-Key: http://www.huennebeck-online.de/public_keys/pgp_andreas.asc

```
Reply by July 10, 20062006-07-10
```robert bristow-johnson wrote:
> Google is your friend.  try it!
>
> http://www.wcl.ee.upatras.gr/audiogroup/Publications/Full%20Papers/asme.pdf
>
> i read the abstract and i don't even know what it is about.
>
> r b-j

Agree, Google is my friend :-) But the article has nothing cobcrete,
except for a ref to the P.Hatziantoniou,J.Mourjopoulos paper.

At any case, I have found inside DRC code something like dtetrmining a
window width for a smotthing. Now, my python code looks  this way:

def foClasicSmooth( x, octaveFraction = 1.0/3.0 ):
y = []
rightFactor = math.pow(2.0, octaveFraction / 2.0 )
leftFactor =  1.0 / rightFactor

for n in range(len(x)):
left = long(n * leftFactor)
right = long(n * rightFactor)
y.append( sum( x[left : right + 1] ) / (right - left + 1) )

return y

It works fine for me.

```
Reply by July 9, 20062006-07-09
```a@gaydenko.com wrote:
> Does anybody know a ref(s) to free description of fractional octave
> smoothing?

```Does anybody know a ref(s) to free description of fractional octave