[mirtoolbox] mircepstrum delays
- From: Hendrik Schreiber <hs@xxxxxxxxxxxx>
- To: mirtoolbox@xxxxxxxxxxxxx
- Date: Fri, 17 Sep 2010 11:50:12 +0200
Hi Guys,
I'm currently looking at the mircepstrum code and have a couple of questions:
The documentation says, the min delay is 0s - the code says it's 0.0005s - I
guess that's a (documentation) bug, right?
I'm unfortunately not very familiar with cepstrums - is there some literature
(you could point me to) that explains why I would like to use delays when
creating the cepstrum rather than when interpreting it? This might also be a
nice addition for the docs/code-comments.
Also, I'm wondering, why does one want to specify a max delay? Isn't that given
through the sample frequency and the size of the window?
Thanks a lot for your insights,
-hendrik
Other related posts:
