Hi,

I'm planning to generate a sin tone in a given frequnecy and length then apply fade-in effect to beginning of this tone and fade-out effect to end of this tone. And finally save this tone to wav file. Is it possible with pure data? Is there any example?

Hi,

How can i add externals to my android or is project which use pd-android and pd-ios. I need add timreID external and iemlib external to my project.

Thanks

Hi,

I'm looking for echo detection with pd. I found http://puredata.info/downloads/timbreid which performs cepstrum analysis. But i need some examples and autocepstrum methods. Is there anybody who has experience with echo detection.

Hi,

Is there any external object for cepstrum analysis? or any antorher way to make a cepstrum analysis?

Hi

pd-for-android doesn't load iemlib as externals. How can i add iemlib objects to my android project

Hi,

I want to find peak frequency in a frequency interval like 2000hz-3000hz from adc.
i tried :

[adc~]
|
[lop~] --- [3000]
|
[hip~] --- [2000]
|
[sigmund~ -npeak 5 peaks]
|
[route 0 1 2 3 4 5]
|
[unpack 0 0 0 0 0]
|
[value]

but i couldn't get success.

How can i do this?

thanks

I have a very simple sigmund patch.

As [adc~]---[sigmund~ pitch]---[number].

For example, with this patch number is 71.2 for 500hz external noise.
How can i convert this 71.2 pitch value to frequency. What is this 71.2 value?

thanks

Hi

I want to do a patch which detects pitch frequency from adc which is greater than 15000hz. Is there any sample for this situation? How can i do this patch?

Thanks for your quick reply, i need also autocepstrum (autocorrelation of cepstrum), how can i do this with tId? And do you know how can i detect echo's with this library?

Thank you very much for your answer, i used [hp10_butt~] and it filters audio. But now i have problems detecting peaks. I have peak freqs on 2500hz and 2600hz, they occurs seperately in a random time, i want only find these peaks when they occur. Sigmund detects many peaks between 2000-3000hz but it cannot always catch my expected peaks(2500hz and 2600hz). Is there any another efficient way of doing this? Thanks

Isn't there any comment or answer, i don't want making my custom fft patch for finding peaks, [sigmund~] works very well

@katjav said:

@ryback3 said:

i need pitch detection for high frequencies which is greater than 17000

Sorry, the technique used in [helmholtz~] (autocorrelation) is very precise but not for high frequencies. Certainly not 17 KHz. Pitch of acoustic instruments rarely go above 1 KHz, that's [helmholtz~] default limit but it can go to 2 KHz using message maxfreq. If you want higher, better use an fft based detector like [sigmund~] or [fiddle~].

Katja

I tried sigmund~ and set -maxfreq to 22000 and npts to 1024 but i couldn't get pitch value which are greater than 15000 hz, what is the problem.

I tried sigmund~ and set -maxfreq to 22000 but i can not get pitch value which are greater than 15000 hz, what is the problem. I tried also fiddle~ it was also failed.

Hi,

i tried [helmholtz~] and it works great but i need pitch detection for high frequencies which is greater than 17000. I couldn't update [helmholtz~] for detecting high frequencies. How can i disable lo pass filter and detect high frequencies with [helmholtz~]