Forums

Audio- Signal to Interference Ratio measurement

Started by shal January 27, 2009
Hello Sir,

I am trying to run a lab test, where I am capturing an audio signal
(running through computer speaker sampled at 48kHz) with a microphone.
Besides the main audio signal there are other cluttering/ background
noises( laughing, people talking).

Is there a way I can measure the SIR using a piece of hardware. Or else
how do I do it.

I wish to parameterize my room setting and hence SIR measurement is
important

Thanks!
Shal
On 27 Jan, 18:08, "shal" <shalini_pr...@yahoo.co.in> wrote:
> Hello Sir, > > I am trying to run a lab test, where I am capturing an audio signal > (running through computer speaker sampled at 48kHz) with a microphone. > Besides the main audio signal there are other cluttering/ background > noises( laughing, people talking). > > Is there a way I can measure the SIR using a piece of hardware. Or else > how do I do it.
I'd at least make a recording of only the background noise, with no signal present, and then use this as basis an SIR estimate. The noises you describe are not very stable, so you may or may not get accurate estimates for the 'true' SIR. What other options you have depend on the type of signal you use for your measurements. If the signal is very different from the noise (like very sharp, preferably known, transients or narrow spectrum lines at known frequencies) you can use this knowledge about the signal to come up with estimates for the SIR. Rune
shal wrote:
> Hello Sir, > > I am trying to run a lab test, where I am capturing an audio signal > (running through computer speaker sampled at 48kHz) with a microphone. > Besides the main audio signal there are other cluttering/ background > noises( laughing, people talking). > > Is there a way I can measure the SIR using a piece of hardware. Or else > how do I do it. > > I wish to parameterize my room setting and hence SIR measurement is > important
Why not run the test at night when the room is quiet? Jerry -- Engineering is the art of making what you want from things you can get. &#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;&#2013266095;
>On 27 Jan, 18:08, "shal" <shalini_pr...@yahoo.co.in> wrote: >> Hello Sir, >> >> I am trying to run a lab test, where I am capturing an audio signal >> (running through computer speaker sampled at 48kHz) with a microphone. >> Besides the main audio signal there are other cluttering/ background >> noises( laughing, people talking). >> >> Is there a way I can measure the SIR using a piece of hardware. Or
else
>> how do I do it. > >I'd at least make a recording of only the background noise, with >no signal present, and then use this as basis an SIR estimate. >The noises you describe are not very stable, so you may or may >not get accurate estimates for the 'true' SIR. > >What other options you have depend on the type of signal you use >for your measurements. If the signal is very different from the >noise (like very sharp, preferably known, transients or narrow >spectrum lines at known frequencies) you can use this knowledge >about the signal to come up with estimates for the SIR. > >Rune >
______________ Thanks Rune, I guess I will go ahead with your idea. The tape of random noises most likely to be present is created. I will this signal for calculations. Apart, I am thinking of measuring Sound pressure level (in dB) of Main audio signal and Noise. This definilty tells nothing about frequency that might be interfereing, but do give the relative power of each signal when the microphone does the capture.