1
$\begingroup$

I'm trying to localize a sound by using an array of microphones, the localization is done based on the sum and delay algorithm. I'm trying to to do the same in machine learning. I don't have much knowledge about machine learning but I think it can be done using linear regression using stochastic gradient ascend, I'm using a sine wave as input, I have to optimize the delay to and find out at which delay I get the max gain at the output. can anyone provide an example for linear regression using stochastic gradient ascend

$\endgroup$

closed as unclear what you're asking by MBaz, Stanley Pawlukiewicz, lennon310, Peter K. Jan 31 at 20:16

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ The machine can search for identical sounds on the different audio tracks and pass them to the localization equation. The search for identical sound is "machine recognition of audio signals / waveforms. it's going to be identifying things. you may find that the shazam variety of algorythm is the most efficient for recognizing audio.. it transforms signal into FFT make a fingerprint of the FFT graph peaks.... I would search for machine comparison of similar signals from different tracks... The localization is very easy, it's just an equation of the time differences. $\endgroup$ – com.prehensible Jan 31 at 8:28
  • $\begingroup$ So, perhaps research sound ID and send results to the time comparison equation. a lot has been written and many libraries and research are available for automatic sound recognition, you will have divide the microphone signals into soundclip snippets and compare across all different tracks for same one. $\endgroup$ – com.prehensible Jan 31 at 8:30
  • $\begingroup$ I'm a new·bie in this field don't know that much machine learning. So what are you suggesting me to do? $\endgroup$ – Alwyn Jan 31 at 8:51
  • $\begingroup$ It's a very cool topic and very relevant to current technologies, so it's worth studying. I would have searched for "comparing sound machine learning" and "audio recognition machine learning", search for pages related to a variety of any search words that you can think of in the search phase. first result is this: medium.com/@awjuliani/… $\endgroup$ – com.prehensible Feb 1 at 16:16
  • $\begingroup$ intro science page: analyticsvidhya.com/blog/2017/08/… .... google.com/… $\endgroup$ – com.prehensible Feb 1 at 16:30