> Am 22.02.2016 um 17:01 schrieb Dario Sanfilippo :
>
> I'll try studying autocorrelation more and see if I can implement a new
> algorithm or combine it to the one I already have.
Dario,
you need to be careful with polyphonic material plus noise. The autocorrelation
will have much lower pe
Thanks, Corey, Steffan and Risto.
I'll try studying autocorrelation more and see if I can implement a new
algorithm or combine it to the one I already have.
ZCR alone doesn't seem to work unless for specific contexts. Would you
agree on that? For example, a 10kHz sine and a 2kHz bandwidth noise c
On February 22, 2016 at 12:13:39 pm +01:00, Corey K wrote:
> I don't have any links on the use of autocorrelation in this context, and I
> don't even know if it would work. My basic thought, however, was that the
> autocorrelation of white noise should be zero at all time lags other than 0.
Auto correlation does work in that context. Therefore you just calculate the
ratio of the peak at the pitch period and the 0ms peak (IOW normalize to
energy).
Steffan
> On 22.02.2016|KW8, at 12:13, Corey K wrote:
>
> I don't have any links on the use of autocorrelation in this context, a
These properties are true, if you have only noise or only signal. In case of a
mixture, also the described properties mix and this “torpedoes” that approach.
So, an FFT with a subsequent processing like floor estimation (connect a line
thru all floors between peaks) and peak estimation (connect
I don't have any links on the use of autocorrelation in this context, and I
don't even know if it would work. My basic thought, however, was that the
autocorrelation of white noise should be zero at all time lags other than
0. Pitched signals, on the other hand, should have peaks at multiples of
th
Thank you so much for the explanation, Ethan.
Best,
Dario
On 21 February 2016 at 23:47, Ethan Duni wrote:
> Not a purely time-domain approach, but you can consider comparing sparsity
> in the time and Fourier domains. The idea is that periodic/tonal type
> signals may be non-sparse in the time
Not a purely time-domain approach, but you can consider comparing sparsity
in the time and Fourier domains. The idea is that periodic/tonal type
signals may be non-sparse in the time domain, but look sparse in the
frequency domain (because all of the energy is on/around harmonics).
Similarly, trans
Hello.
Corey: I'm honestly not so familiar with auto-correlation; I'm aware that
it is implemented for pitch-detection but I didn't know about those other
features; would you have a reference or link to a document I could check
out?
Evan: I get your point; in my case I was following more of a low
wouldn't using varying ZCR be defeated by frequency modulated or bell tones?
One could also craft a very noisy signal with a perfectly periodic ZCR.
James McCartney
> On Feb 19, 2016, at 04:49, Dario Sanfilippo
> wrote:
>
> Hello everybody.
>
> Following on a discussion about cheap/time-do
Noise is an elusive concept. One way of thinking about it is that the real
signal can be decomposed into a sum of two theoretical signals, comprising
desirable and undesirable information. What isn't signal is noise; and
what isn't noise is signal -- so our definitions or models for these must
be
I haven't researched this at all, so take the following with a grain of
salt. But, how about looking at different features of the auto-correlation
(e.g., flatness, peakiness, ...)?
On Fri, Feb 19, 2016 at 1:49 PM, Dario Sanfilippo <
sanfilippo.da...@gmail.com> wrote:
> Hello everybody.
>
> Follow
Hello everybody.
Following on a discussion about cheap/time-domain spectral centroid
estimators, I thought it could have been interesting to also discuss
time-domain noisiness estimators.
I think that a common approach is the FFT-based spectral flatness
algorithm. In the time-domain, zero-crossin
13 matches
Mail list logo