[Clam-devel] Developing a peak detection monitor

Xavier Amatriain xavier at create.ucsb.edu
Wed Apr 18 11:57:36 PDT 2007


Hi JCIMS, I see you are speeding up ;)

What I would recommend is to look at the "Creating a minimal Processing
Object Tutorial" at the wiki:

http://clam.iua.upf.edu/wikis/clam/index.php/Creating_a_minimal_processing_object

Once you get that rolling and you have your Processing object in Network
Editor, implementing what you want will be very easy. If I understand
right you just need to measure peaks over the time-domain waveform,
right? If so, having a simple implementation will be just a matter of
adding 4-5 lines to your Do method.

On Wed, 2007-04-18 at 18:40 +0000, JCIMS wrote:
> Hi Folks!
>  
> I am brand new to CLAM and have found it to be very adaptable.  I am
> currently using it as the analysis tool for a software-based
> tachometer that i'm developing.  The basic approach is to pass the
> input audio (live or from a file) through the Fundamental analyis and
> let the ControlPrinter spit the fundamental frequency out to standard
> output.  This fundamental frequency is very close to the actual RPM of
> the device (in Hz) and has been very useful for rough (+/- 5%)
> estimates.
>  
> However, i'm moving into areas where i'm measuring velocity and
> acceleration, and have found that this approach is not adequate.  I
> have found, through converting WAV files to ASCII and perl scripts,
> that measuring peak to peak times is much more accurate and precise.
> I would like to try to implement this in CLAM, but don't know where to
> start.
>  
> First of all, i guess i would like to know if there is existing code
> that can identify peak values over a given threshold, and whether that
> code can pass the values in the form of a timestamp or sample #.  If
> not, i'm going to have to roll my own.
>  
> I just wonder if any of you have advice for how i should tackle this.
> It will be my first C++ project in about 5 years, and my first Qt
> project ever.  I did look through some of the existing code in SVN and
> feel like i have a reasonable shot at making something useful...but
> would love some input!
>  
> Thanks!
>  
>  
> _______________________________________________
> Clam-devel mailing list
> Clam-devel at llistes.projectes.lafarga.org
> https://llistes.projectes.lafarga.org/cgi-bin/mailman/listinfo/clam-devel
-- 
/*********************************
 *       Xavier Amatriain        *
 *  Associate Director - MATi    *
 *  Research Director - CREATE   *
 *    UCSB, Santa Barbara CA     *
 *      1-(805)- 893 83 52       *
 ********************************/





More information about the clam-devel mailing list