Spectrum seems different when porting from 0.10 to 1.0

Victor henri nadaeck at hotmail.com
Tue Nov 4 10:04:02 PST 2014


Hello 

I did some tests. It appears the problem is connected with the Band-Pass/Band-reject filter (audiochebband). When I remove it, everything behaves with gst-1.0 as with gst-0.10.

Has this plugin changed? Can I find an example of code using this properly somewhere for I don't see it in 
gst-plugins-good/tests/examples/spectrum?

Thank you for your help

Victor



> 
> My app displays harmonics of the sound in 3D, thanks to the 'spectrum' element.
> When I use the microphone of the computer as the audio source, with Gstreamer-0.10, I have a clear view of the harmonics.
> When I do the same with Gstreamer-1.0, the sensitivity seems much higher, and I have a huge "background" noise contamining the spectrum data, especially in the lower frequencies (but also in the rest of the spectrogram), wich makes the ratio sound/background noise much lower. I can lower the sensitivity of the microphone but in that case, I loose the interesting display...
> 
> When I do an analysis of an audio file (either a flac, wac, mp3... or a file recorded by the app), it works fine.
> 
> I join two pictures to show the difference, when pronouncing differents voyels; you must see higher and lower harmonics, according to wich voyel is said; on the first pic, when gst-0.10 is used, you can see a very clear drawing; on the second pic, when gst-1.0 is used, although the sensitivity of the mic is lessened, the ratio sound/background noise is much worse.
> 
> I guess the spectrum element has been rewritten. What is possible to change? Is it possible to get more information on how this plugins works?
 
This looks like a regression in spectrum. It wasn't really rewritten,
but it was refactored a bit and IIRC the API changed slightly too.
 
Take a look at the example in gst-plugins-good/tests/examples/spectrum
to see if you're still using it correctly. If you do, please open a bug
report at http://bugzilla.gnome.org with a testcase to reproduce the
different behaviours. Thanks!
 
-- 
Sebastian Dröge, Centricular Ltd · http://www.centricular.com

_______________________________________________
gstreamer-devel mailing list
gstreamer-devel at lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/gstreamer-devel 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/gstreamer-devel/attachments/20141104/1e96d4da/attachment.html>


More information about the gstreamer-devel mailing list