Subject: Re: Subtracting regularly repeating sounds From: Kyle Jasmin <kyle.jasmin.11@xxxxxxxx> Date: Thu, 20 Sep 2012 13:51:14 -0400 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>--bcaec517a9c4041c7a04ca25c553 Content-Type: text/plain; charset=ISO-8859-1 Dear all I've tried Bryan's repeated pattern extraction technique -- it worked pretty well, although it removed some of the speech. If you're curious about the results, the input and output files are here. https://www.dropbox.com/sh/nrj4j7mmefhf36w/WBrN8KmlhS Next I'll try Jort's approach. Thanks for the advice. Best Kyle On Thu, Sep 20, 2012 at 12:54 PM, Matt Flax <flatmax@xxxxxxxx> wrote: > The exemplar work looks like a good approach, as Jort has provided > Matlab code, I would imagine that it could be readily tested for your > particular case. > > Another approach to think about, if you have a section of isolated noise > in your audio recording, is the pioneering work by Ephraim & Malah. > They produced the statistical method of spectral noise estimation and > removal ... the trick in your case would be to make sure that you are using > an analysis window which is long enough to be able to capture the low > frequency components of your noise source. > > You can find a SPDemo application in the downloads section on their web > page : > http://sipl.technion.ac.il/ > > Matt > > > On 09/20/2012 12:19 AM, Jort Gemmeke wrote: > > Dear Kyle, > > I dont have experience with MRI scanner sounds in particular, but it > seems like such a structured interfering source should be a good match for > the exemplar-based source separation techniques I have been using over the > past few years. You can find more information on my website, > www.amadana.nl , as well as a demo (matlab) and some links to related > techniques that may be of use. > > Feel free to contact me directly if you think this can be of use, then > we can give it a try. > > Jort Gemmeke > Postdoctoral Researcher > KU Leuven, ESAT-PSI Speech group > www.amadana.nl > > > Date: Wed, 19 Sep 2012 16:59:06 -0400 >> From: Kyle Jasmin <kyle.jasmin.11@xxxxxxxx> >> Subject: Subtracting regularly repeating sounds >> >> --bcaec501654d005be504ca1447c5 >> Content-Type: text/plain; charset=ISO-8859-1 >> >> Dear all, >> >> I wonder if anyone has successfully removed the sound of a helium pump >> from >> auditory recordings made in an MRI scanner. It occurs regularly (every >> 1s), >> and changes over time (Praat thinks it is a formant with downward >> trajectory). Any advice is appreciated on subtracting regularly repeating >> sounds would be appreciated. >> >> With thanks, >> >> Kyle Jasmin >> Speech Communication Laboratory, UCL Institute of Cognitive Neuroscience >> & Laboratory of Brain and Cognition, National Institute of Mental Health >> >> > --bcaec517a9c4041c7a04ca25c553 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Dear all=A0<div><br></div><div>I've tried Bryan's repeated pattern = extraction technique -- it worked pretty well, although it removed some of = the speech. If you're curious about the results, the input and output f= iles are here.=A0<a href=3D"https://www.dropbox.com/sh/nrj4j7mmefhf36w/WBrN= 8KmlhS">https://www.dropbox.com/sh/nrj4j7mmefhf36w/WBrN8KmlhS</a></div> <div><br></div><div>Next I'll try Jort's approach. Thanks for the a= dvice.</div><div><br></div><div>Best</div><div>Kyle<br><br><div class=3D"gm= ail_quote">On Thu, Sep 20, 2012 at 12:54 PM, Matt Flax <span dir=3D"ltr">&l= t;<a href=3D"mailto:flatmax@xxxxxxxx" target=3D"_blank">flatmax@xxxxxxxx= org</a>></span> wrote:<br> <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p= x #ccc solid;padding-left:1ex"> <div text=3D"#000000" bgcolor=3D"#FFFFFF"> <div>The exemplar work looks like a good approach, as Jort has provided Mat= lab code, I would imagine that it could be readily tested for your particul= ar case.<br> <br> Another approach to think about, if you have a section of isolated noise in= your audio recording, is the pioneering work by Ephraim & Malah.<br> They produced the statistical method of spectral noise estimation and remov= al ... the trick in your case would be to make sure that you are using an a= nalysis window which is long enough to be able to capture the low frequency= components of your noise source.<br> <br> You can find a SPDemo application in the downloads section on their web pag= e :<br> <a href=3D"http://sipl.technion.ac.il/" target=3D"_blank">http://sipl.techn= ion.ac.il/</a><span class=3D"HOEnZb"><font color=3D"#888888"><br> <br> Matt</font></span><div class=3D"im"><br> <br> On 09/20/2012 12:19 AM, Jort Gemmeke wrote:<br> </div></div> <blockquote type=3D"cite"><div class=3D"im"> Dear Kyle, <div><br> </div> <div>I dont have experience with MRI scanner sounds in particular, but it s= eems like such a structured interfering source should be a good match for t= he exemplar-based source separation techniques I have been using over the p= ast few years. You can find more information on my website, <a href=3D"http://www.amadana.nl" target=3D"_bl= ank"> www.amadana.nl</a> , as well as a demo (matlab) and some links to related t= echniques that may be of use.</div> <div><br> </div> <div>Feel free to contact me directly if you think this can be of use, then= we can give it a try.</div> <div><br> </div> <div>Jort Gemmeke</div> <div>Postdoctoral Researcher</div> <div>KU Leuven, ESAT-PSI Speech group</div> <div><a href=3D"http://www.amadana.nl" target=3D"_blank">www.amadana.nl</a>= </div> <div><br> </div> </div><div><br> <div class=3D"gmail_quote"> <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p= x #ccc solid;padding-left:1ex"><div class=3D"im"> Date: =A0 =A0Wed, 19 Sep 2012 16:59:06 -0400<br> From: =A0 =A0Kyle Jasmin <<a href=3D"mailto:kyle.jasmin.11@xxxxxxxx" ta= rget=3D"_blank">kyle.jasmin.11@xxxxxxxx</a>><br> Subject: Subtracting regularly repeating sounds<br> <br> --bcaec501654d005be504ca1447c5<br> Content-Type: text/plain; charset=3DISO-8859-1<br> <br></div><div class=3D"im"> Dear all,<br> <br> I wonder if anyone has successfully removed the sound of a helium pump from= <br> auditory recordings made in an MRI scanner. It occurs regularly (every 1s),= <br> and changes over time (Praat thinks it is a formant with downward<br> trajectory). Any advice is appreciated on subtracting regularly repeating<b= r> sounds would be appreciated.<br> <br> With thanks,<br> <br> Kyle Jasmin<br> Speech Communication Laboratory, UCL Institute of Cognitive Neuroscience<br= > & Laboratory of Brain and Cognition, National Institute of Mental Healt= h<br> <br> </div></blockquote> </div> </div> </blockquote> <br> </div> </blockquote></div><br></div> --bcaec517a9c4041c7a04ca25c553--