Re: Optimal sweep duration for BRIR measurements (Trevor Agus )


Subject: Re: Optimal sweep duration for BRIR measurements
From:    Trevor Agus  <t.agus@xxxxxxxx>
Date:    Tue, 28 Jul 2015 09:53:04 +0100
List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>

This is a multi-part message in MIME format. --------------080004090108030700050505 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: quoted-printable Dear list, There's a conundrum in this discussion that: (1) increasing the duration of the sine-wave sweep increases the=20 signal-to-noise ratio (which seems intuitively true) versus (2) increasing the duration of the sine-wave sweep does not affect the=20 signal-to-noise ratio (as John noted; which is literally true if the=20 "signal" is the sweep, whose level is unaffected by duration). It's a fun paradox, and I don't want to take away from anyone's pleasure=20 by my stab at a resolution... (Potential spoiler alert.) Is it that the SNR of interest is the impulse-response-to-noise, only=20 over its relatively short time period? As such, a large amount of the=20 energy of the noise (in a long sweep) would be outside of the IR's=20 expected time period (after the convolution stage that 'extracts' the=20 IR), but all the energy of the sine sweep (and its reverberation) would=20 be kept within this time period. So the SNR is constant (if you consider=20 the full duration) yet increased (if you focus on the time period of the=20 IR). Or is there more to it? All the best, Trevor Brian FG Katz wrote: > > Dear John, > > As others have pointed out, increasing the length of the sweep > increases your signal-to-noise ratio. For large room acoustics, we > typically use sweeps of 20 to 40 seconds. This of course depends on > the size of your room, its reverberation time, and the power of your > source. The weaker the source, the longer the sweep. You should > basically do a test and see what SNR you get. If you need more, and > your measurement chain is at its limit, the only option is longer > sweep. Longer sweeps will not help much with impulsive interruptions, > while averaging will.With a single sweep, basically that freq-range > during the noise is lost. If you are just measuring from the RIR, this > may not be an issue, as measurement parameters are often in > octave-bands, but this is not true for auralization usage; corrupt > data is corrupt data, though I haven't gone through a thorough study > of this actual case. > > The sweep should definitely be longer than the RT, for room > measurements. Then, don't forget that you need to record the sweep > length PLUS the RT or more if your SNR is better than 60dB! > > We have also compared averaging repeated sweeps vs. longer sweeps. > Avoiding the recent developments in overlapping sweep processing, this > basic repetition approach 'requires' the decay of the first sweep to > finish before you launch the second sweep. As such, 3x20 second sweeps > take longer than 1 x 60 second sweep, due to the additional pauses. Of > course, without repetitions, you have no backup in case something goes > wrong, like a door slam or something. So, I tend to use repeated > sweeps and take the best 1. > > We do all our processing in MatLab, and have never had an issue > (within the last 20 years) of processing long sweeps in real halls. > > As you are considering BRIR, and not anechoic HRTFs, you are subject > to the same conditions. If you want to convolve the BRIR directly, you > will need to ensure that the SNR is sufficiently high that the > noise-floor is not audible as a late reverb part of the BRIR. Some > noise extension methods exist from older studies on basic auralization > and scale model RIR auralizations. I cannot imagine a 2s sweep for > BRIR unless you are measuring an office of other room with <1 sec RT. > I think there is a fault in your correlation-based analysis for > reliability, and there are too many factors to consider in comparing > BRIR of different lengths. First, examine your SNR. > > Regarding distortion of the source, this should be an issue for the > processing element (different from burning out you speaker). This is > because one of the strengths of the sweep method (when done correctly) > is that any harmonic distortion components of a higher frequency that > the excitation signal at the time or folded back to BEFORE the direct > sound after deconvolution by the excitation signal (as that frequency > has yet to be generated). We presented a through work on this feature, > extending Farina's earlier works, to general conditions: > > M. R=E9billat, R. Hennequin, E. Corteel, and B. Katz, "Identification o= f > cascade of Hammerstein models for the description of nonlinearities in > vibrating devices," J. Sound and Vibration, vol. 330, pp. 1018--1038, > 2011, (doi:10.1016/j.jsv.2010.09.012). > > This method lets you actually extract and analyse the different > harmonic distortions (THD etc.) as well as allowing for the modelling > of non-linear responses. > > Cheers, > > Brian > > -- > > Brian FG Katz, Ph.D, HDR > > Resp. Groupe Audio & Acoustique > > LIMSI - CNRS > Rue John von Neumann > Campus Universitaire d'Orsay, B=E2t 508 > 91405 Orsay cedex > > France > > Phone. + 33 (0)1 69 85 80 67 - Fax. + 33 (0)1 69 85 80 88 > http://www.limsi.fr <http://www.limsi.fr/> web_group: > http://www.limsi.fr/Scientifique/aa/ web_theme: > http://www.limsi.fr/Scientifique/aa/thmsonesp/ > > *De :*AUDITORY - Research in Auditory Perception > [mailto:AUDITORY@xxxxxxxx *De la part de* Anders Tornvig > Christensen > *Envoy=E9 :* lundi 27 juillet 2015 09:07 > *=C0 :* AUDITORY@xxxxxxxx > *Objet :* Re: Optimal sweep duration for BRIR measurements > > Hello John, > > The sweep method at a single frequency is an approximation to a > steady-state measurement with a pure tone. Longer sweeps give higher > signal to noise ratio per sweep because it spends more time per > frequency. > > Short repeated sweeps (but not shorter than the length of the impulse > response) are good, if time-varying or sudden noise that doesn't > average out is likely to contaminate the measurement. > > Sweep duration (and "rate" in general) also matters if the system > (room in your case) is nonlinear, time-variant, or both, but that's > another discussion. > > Something is wrong with your implementation if the temporal offset of > the impulse responses you measure depends on the sweep duration. You > should be able to check this by connecting the output of your sound > card directly to its input. Also note that wrongly measured or wrongly > computed impulse responses may be very reproducible in terms of > correlation. > > Best, > Anders > > PhD student in acoustics > Aalborg University, Denmark > > -----------------------------------------------------------------------= - > > *From:*AUDITORY - Research in Auditory Perception > [AUDITORY@xxxxxxxx on behalf of John Culling > [CullingJ@xxxxxxxx > *Sent:* Friday, July 24, 2015 5:25 PM > *To:* AUDITORY@xxxxxxxx <mailto:AUDITORY@xxxxxxxx> > *Subject:* Optimal sweep duration for BRIR measurements > > Dear all, > > Basic Q... > > Does anyone have insight into the optimum sweep duration using > Farina's method > > for measuring room impulses responses? > > More detailed background... > > We are planning to make an extensive series of measurements, and in > preparation have > > been testing the method using different sweep durations. One way to > check the method > > is to correlate the impulses respones from repeated measurements or > those generated > > with different durations. To our surprise short sweeps (1-2 seconds) > appear to give more > > reliable results (repeated sweeps correlate, r>0.98) than longer ones. > Comparing sweeps > > of different durations is a little trickier, because we find a > temporal offset that reduces > > the correlation and can only be partially overcome by using > cross-correlation. Nonetheless, > > it is apparent that durations from 1 second upwards correlate well, > while going below one > > second leads to reliable IRs, but ones that are inaccurate when > compared with those from > > longer sweep durations. > > Our surprising conclusion is that ~2s should be fine, but Farina > refers to an ISO standard that > > recommends very long sweeps (Farina has an example of 50s) to help > overcome noise. > > This seems an unintuitive rationale to us, since longer sweeps should > increase both the > > signal energy captured and the noise energy, and the method does not > involve averaging > > as far as I understand. Longer durations should help address brief > interupting sounds, but > > I am unsure if that it what was the idea. In the presence of > continuous noise, we did not > > notice any improvement in the IRs produced by longer sweeps. > > The nascent plan is to take >1 short sweep for each measurement and > reject IRs that > > that don't correlate well with another. > > Any insights/advice appreciated, > > John. > > Prof. John Culling > > School of Psychology, Cardiff University > > Tel: +44 (0)29 2087 4556 > > > > Yr Athro John Culling > > Yr Ysgol Seicoleg, Prifysgol Caerdydd > > Ff=F4n : +44 (0)29 2087 4556 > Brian FG Katz <mailto:brian.katz@xxxxxxxx> > 27 July 2015 17:46 > > Dear John, > > As others have pointed out, increasing the length of the sweep=20 > increases your signal-to-noise ratio. For large room acoustics, we=20 > typically use sweeps of 20 to 40 seconds. This of course depends on=20 > the size of your room, its reverberation time, and the power of your=20 > source. The weaker the source, the longer the sweep. You should=20 > basically do a test and see what SNR you get. If you need more, and=20 > your measurement chain is at its limit, the only option is longer=20 > sweep. Longer sweeps will not help much with impulsive interruptions,=20 > while averaging will.With a single sweep, basically that freq-range=20 > during the noise is lost. If you are just measuring from the RIR, this=20 > may not be an issue, as measurement parameters are often in=20 > octave-bands, but this is not true for auralization usage; corrupt=20 > data is corrupt data, though I haven't gone through a thorough study=20 > of this actual case. > > The sweep should definitely be longer than the RT, for room=20 > measurements. Then, don't forget that you need to record the sweep=20 > length PLUS the RT or more if your SNR is better than 60dB! > > We have also compared averaging repeated sweeps vs. longer sweeps.=20 > Avoiding the recent developments in overlapping sweep processing, this=20 > basic repetition approach 'requires' the decay of the first sweep to=20 > finish before you launch the second sweep. As such, 3x20 second sweeps=20 > take longer than 1 x 60 second sweep, due to the additional pauses. Of=20 > course, without repetitions, you have no backup in case something goes=20 > wrong, like a door slam or something. So, I tend to use repeated=20 > sweeps and take the best 1. > > We do all our processing in MatLab, and have never had an issue=20 > (within the last 20 years) of processing long sweeps in real halls. > > As you are considering BRIR, and not anechoic HRTFs, you are subject=20 > to the same conditions. If you want to convolve the BRIR directly, you=20 > will need to ensure that the SNR is sufficiently high that the=20 > noise-floor is not audible as a late reverb part of the BRIR. Some=20 > noise extension methods exist from older studies on basic auralization=20 > and scale model RIR auralizations. I cannot imagine a 2s sweep for=20 > BRIR unless you are measuring an office of other room with <1 sec RT.=20 > I think there is a fault in your correlation-based analysis for=20 > reliability, and there are too many factors to consider in comparing=20 > BRIR of different lengths. First, examine your SNR. > > Regarding distortion of the source, this should be an issue for the=20 > processing element (different from burning out you speaker). This is=20 > because one of the strengths of the sweep method (when done correctly)=20 > is that any harmonic distortion components of a higher frequency that=20 > the excitation signal at the time or folded back to BEFORE the direct=20 > sound after deconvolution by the excitation signal (as that frequency=20 > has yet to be generated). We presented a through work on this feature,=20 > extending Farina's earlier works, to general conditions: > > M. R=E9billat, R. Hennequin, E. Corteel, and B. Katz, "Identification o= f=20 > cascade of Hammerstein models for the description of nonlinearities in=20 > vibrating devices," J. Sound and Vibration, vol. 330, pp. 1018--1038,=20 > 2011, (doi:10.1016/j.jsv.2010.09.012). > > This method lets you actually extract and analyse the different=20 > harmonic distortions (THD etc.) as well as allowing for the modelling=20 > of non-linear responses. > > Cheers, > > Brian > > -- > > Brian FG Katz, Ph.D, HDR > > Resp. Groupe Audio & Acoustique > > LIMSI - CNRS > Rue John von Neumann > Campus Universitaire d'Orsay, B=E2t 508 > 91405 Orsay cedex > > France > > Phone. + 33 (0)1 69 85 80 67 - Fax. + 33 (0)1 69 85 80 88 > http://www.limsi.fr <http://www.limsi.fr/> web_group:=20 > http://www.limsi.fr/Scientifique/aa/ web_theme:=20 > http://www.limsi.fr/Scientifique/aa/thmsonesp/ > > *De :*AUDITORY - Research in Auditory Perception=20 > [mailto:AUDITORY@xxxxxxxx *De la part de* Anders Tornvig=20 > Christensen > *Envoy=E9 :* lundi 27 juillet 2015 09:07 > *=C0 :* AUDITORY@xxxxxxxx > *Objet :* Re: Optimal sweep duration for BRIR measurements > --------------080004090108030700050505 Content-Type: multipart/related; boundary="------------050307080404060203060206" --------------050307080404060203060206 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit <html><head> <meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type"> </head><body bgcolor="#FFFFFF" text="#000000">Dear list,<br> <br> There's a conundrum in this discussion that:<br> (1) increasing the duration of the sine-wave sweep increases the signal-to-noise ratio (which seems intuitively true)<br> versus<br> (2) increasing the duration of the sine-wave sweep does not affect the signal-to-noise ratio (as John noted; which is literally true if the "signal" is the sweep, whose level is unaffected by duration).<br> <br> It's a fun paradox, and I don't want to take away from anyone's pleasure by my stab at a resolution... (Potential spoiler alert.)<br> <br> Is it that the SNR of interest is the impulse-response-to-noise, only over its relatively short time period? As such, a large amount of the energy of the noise (in a long sweep) would be outside of the IR's expected time period (after the convolution stage that 'extracts' the IR), but all the energy of the sine sweep (and its reverberation) would be kept within this time period. So the SNR is constant (if you consider the full duration) yet increased (if you focus on the time period of the IR). Or is there more to it?<br> <br> All the best,<br> <br> Trevor<br> <br> Brian FG Katz wrote:<br> <blockquote type="cite"><br> Dear John,<br> <br> As others have pointed out, increasing the length of the sweep <br> increases your signal-to-noise ratio. For large room acoustics,&nbsp; we <br> typically use sweeps of 20 to 40 seconds. This of course depends on <br> the size of your room, its reverberation time, and the power of your <br> source. The weaker the source, the longer the sweep. You should <br> basically do a test and see what SNR you get. If you need more, and <br> your measurement chain is at its limit, the only option is longer <br> sweep. Longer sweeps will not help much with impulsive interruptions, <br> while averaging will.With a single sweep, basically that freq-range <br> during the noise is lost. If you are just measuring from the RIR, this <br> may not be an issue, as measurement parameters are often in <br> octave-bands, but this is not true for auralization usage; corrupt <br> data is corrupt data, though I haven&#8217;t gone through a thorough study <br> of this actual case.<br> <br> The sweep should definitely be longer than the RT, for room <br> measurements. Then, don&#8217;t forget that you need to record the sweep <br> length PLUS the RT or more if your SNR is better than 60dB!<br> <br> We have also compared averaging repeated sweeps vs. longer sweeps. <br> Avoiding the recent developments in overlapping sweep processing, this <br> basic repetition approach &#8216;requires&#8217; the decay of the first sweep to <br> finish before you launch the second sweep. As such, 3x20 second sweeps <br> take longer than 1 x 60 second sweep, due to the additional pauses. Of <br> course, without repetitions, you have no backup in case something goes <br> wrong, like a door slam or something. So, I tend to use repeated <br> sweeps and take the best 1.<br> <br> We do all our processing in MatLab, and have never had an issue <br> (within the last 20 years) of processing long sweeps in real halls.<br> <br> As you are considering BRIR, and not anechoic HRTFs, you are subject <br> to the same conditions. If you want to convolve the BRIR directly, you <br> will need to ensure that the SNR is sufficiently high that the <br> noise-floor is not audible as a late reverb part of the BRIR. Some <br> noise extension methods exist from older studies on basic auralization <br> and scale model RIR auralizations. I cannot imagine a 2s sweep for <br> BRIR unless you are measuring an office of other room with &lt;1 sec RT. <br> I think there is a fault in your correlation-based analysis for <br> reliability, and there are too many factors to consider in comparing <br> BRIR of different lengths. First, examine your SNR.<br> <br> Regarding distortion of the source, this should be an issue for the <br> processing element (different from burning out you speaker). This is <br> because one of the strengths of the sweep method (when done correctly) <br> is that any harmonic distortion components of a higher frequency that <br> the excitation signal at the time or folded back to BEFORE the direct <br> sound after deconvolution by the excitation signal (as that frequency <br> has yet to be generated). We presented a through work on this feature, <br> extending Farina&#8217;s earlier works, to general conditions:<br> <br> M. R&eacute;billat, R. Hennequin, E. Corteel, and B. Katz, &#8220;Identification of <br> cascade of Hammerstein models for the description of nonlinearities in <br> vibrating devices,&#8221; J. Sound and Vibration, vol. 330, pp. 1018&#8211;1038, <br> 2011, (doi:10.1016/j.jsv.2010.09.012).<br> <br> This method lets you actually extract and analyse the different <br> harmonic distortions (THD etc.) as well as allowing for the modelling <br> of non-linear responses.<br> <br> Cheers,<br> <br> Brian<br> <br> --<br> <br> Brian FG Katz, Ph.D, HDR<br> <br> Resp. Groupe Audio &amp; Acoustique<br> <br> LIMSI - CNRS<br> Rue John von Neumann<br> Campus Universitaire d'Orsay, B&acirc;t 508<br> 91405 Orsay cedex<br> <br> France<br> <br> Phone. + 33 (0)1 69 85 80 67 - Fax. + 33 (0)1 69 85 80 88<br> <a class="moz-txt-link-freetext" href="http://www.limsi.fr">http://www.limsi.fr</a> <a class="moz-txt-link-rfc2396E" href="http://www.limsi.fr/">&lt;http://www.limsi.fr/&gt;</a> web_group: <br> <a class="moz-txt-link-freetext" href="http://www.limsi.fr/Scientifique/aa/">http://www.limsi.fr/Scientifique/aa/</a>&nbsp;&nbsp;&nbsp;&nbsp; web_theme: <br> <a class="moz-txt-link-freetext" href="http://www.limsi.fr/Scientifique/aa/thmsonesp/">http://www.limsi.fr/Scientifique/aa/thmsonesp/</a><br> <br> *De :*AUDITORY - Research in Auditory Perception <br> [<a class="moz-txt-link-freetext" href="mailto:AUDITORY@xxxxxxxx">mailto:AUDITORY@xxxxxxxx</a>] *De la part de* Anders Tornvig <br> Christensen<br> *Envoy&eacute; :* lundi 27 juillet 2015 09:07<br> *&Agrave; :* <a class="moz-txt-link-abbreviated" href="mailto:AUDITORY@xxxxxxxx">AUDITORY@xxxxxxxx</a><br> *Objet :* Re: Optimal sweep duration for BRIR measurements<br> <br> Hello John,<br> <br> The sweep method at a single frequency is an approximation to a <br> steady-state measurement with a pure tone. Longer sweeps give higher <br> signal to noise ratio per sweep because it spends more time per <br> frequency.<br> <br> Short repeated sweeps (but not shorter than the length of the impulse <br> response) are good, if time-varying or sudden noise that doesn't <br> average out is likely to contaminate the measurement.<br> <br> Sweep duration (and "rate" in general) also matters if the system <br> (room in your case) is nonlinear, time-variant, or both, but that's <br> another discussion.<br> <br> Something is wrong with your implementation if the temporal offset of <br> the impulse responses you measure depends on the sweep duration. You <br> should be able to check this by connecting the output of your sound <br> card directly to its input. Also note that wrongly measured or wrongly <br> computed impulse responses may be very reproducible in terms of <br> correlation.<br> <br> Best,<br> Anders<br> <br> PhD student in acoustics<br> Aalborg University, Denmark<br> <br> ------------------------------------------------------------------------<br> <br> *From:*AUDITORY - Research in Auditory Perception <br> [<a class="moz-txt-link-abbreviated" href="mailto:AUDITORY@xxxxxxxx">AUDITORY@xxxxxxxx</a>] on behalf of John Culling <br> [<a class="moz-txt-link-abbreviated" href="mailto:CullingJ@xxxxxxxx">CullingJ@xxxxxxxx</a>]<br> *Sent:* Friday, July 24, 2015 5:25 PM<br> *To:* <a class="moz-txt-link-abbreviated" href="mailto:AUDITORY@xxxxxxxx">AUDITORY@xxxxxxxx</a> <a class="moz-txt-link-rfc2396E" href="mailto:AUDITORY@xxxxxxxx">&lt;mailto:AUDITORY@xxxxxxxx&gt;</a><br> *Subject:* Optimal sweep duration for BRIR measurements<br> <br> Dear all,<br> <br> Basic Q&#8230;<br> <br> Does anyone have insight into the optimum sweep duration using <br> Farina's method<br> <br> for measuring room impulses responses?<br> <br> More detailed background&#8230;<br> <br> We are planning to make an extensive series of measurements, and in <br> preparation have<br> <br> been testing the method using different sweep durations. One way to <br> check the method<br> <br> is to correlate the impulses respones from repeated measurements or <br> those generated<br> <br> with different durations. To our surprise short sweeps (1-2 seconds) <br> appear to give more<br> <br> reliable results (repeated sweeps correlate, r&gt;0.98) than longer ones. <br> Comparing sweeps<br> <br> of different durations is a little trickier, because we find a <br> temporal offset that reduces<br> <br> the correlation and can only be partially overcome by using <br> cross-correlation. Nonetheless,<br> <br> it is apparent that durations from 1 second upwards correlate well, <br> while going below one<br> <br> second leads to reliable IRs, but ones that are inaccurate when <br> compared with those from<br> <br> longer sweep durations.<br> <br> Our surprising conclusion is that ~2s should be fine, but Farina <br> refers to an ISO standard that<br> <br> recommends very long sweeps (Farina has an example of 50s) to help <br> overcome noise.<br> <br> This seems an unintuitive rationale to us, since longer sweeps should <br> increase both the<br> <br> signal energy captured and the noise energy, and the method does not <br> involve averaging<br> <br> as far as I understand. Longer durations should help address brief <br> interupting sounds, but<br> <br> I am unsure if that it what was the idea. In the presence of <br> continuous noise, we did not<br> <br> notice any improvement in the IRs produced by longer sweeps.<br> <br> The nascent plan is to take &gt;1 short sweep for each measurement and <br> reject IRs that<br> <br> that don't correlate well with another.<br> <br> Any insights/advice appreciated,<br> <br> John.<br> <br> Prof. John Culling<br> <br> School of Psychology, Cardiff University<br> <br> Tel: +44 (0)29 2087 4556<br> <br> <br> <br> Yr Athro John Culling<br> <br> Yr Ysgol Seicoleg, Prifysgol Caerdydd<br> <br> Ff&ocirc;n : +44 (0)29 2087 4556<br> </blockquote> <br> <br> <blockquote style="border: 0px none;" cite="mid:27281_1438056636_55B700BC_27281_245_1_006e01d0c88b$bbeac900$33c05b00$@xxxxxxxx" type="cite"> <div style="margin:30px 25px 10px 25px;" class="__pbConvHr"><div style="display:table;width:100%;border-top:1px solid #EDEEF0;padding-top:5px"> <div style="display:table-cell;vertical-align:middle;padding-right:6px;"><img photoaddress="brian.katz@xxxxxxxx" photoname="Brian FG Katz" src="cid:part1.06010402.07030506@xxxxxxxx" name="compose-unknown-contact.jpg" height="25px" width="25px"></div> <div style="display:table-cell;white-space:nowrap;vertical-align:middle;width:100%"> <a moz-do-not-send="true" href="mailto:brian.katz@xxxxxxxx" style="color:#737F92 !important;padding-right:6px;font-weight:bold;text-decoration:none !important;">Brian FG Katz</a></div> <div style="display:table-cell;white-space:nowrap;vertical-align:middle;"> <font color="#9FA2A5"><span style="padding-left:6px">27 July 2015 17:46</span></font></div></div></div> <div style="color:#888888;margin-left:24px;margin-right:24px;" __pbrmquotes="true" class="__pbConvBody"><meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type"> <meta content="Microsoft Word 14 (filtered medium)" name="Generator"><!--[if !mso]><style>v\:* {behavior:url(#default#VML);} o\:* {behavior:url(#default#VML);} w\:* {behavior:url(#default#VML);} .shape {behavior:url(#default#VML);} </style><![endif]--><style>&lt;!-- /* Font Definitions */ @xxxxxxxx {font-family:Calibri; panose-1:2 15 5 2 2 2 4 3 2 4;} @xxxxxxxx {font-family:Tahoma; panose-1:2 11 6 4 3 5 4 4 2 4;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {margin:0cm; margin-bottom:.0001pt; font-size:11.0pt; font-family:"Calibri","sans-serif";} a:link, span.MsoHyperlink {mso-style-priority:99; color:#0563C1; text-decoration:underline;} a:visited, span.MsoHyperlinkFollowed {mso-style-priority:99; color:#954F72; text-decoration:underline;} p {mso-style-priority:99; margin:0cm; margin-bottom:.0001pt; font-size:12.0pt; font-family:"Times New Roman","serif";} p.MsoAcetate, li.MsoAcetate, div.MsoAcetate {mso-style-priority:99; mso-style-link:"Texte de bulles Car"; margin:0cm; margin-bottom:.0001pt; font-size:8.0pt; font-family:"Tahoma","sans-serif";} p.msochpdefault, li.msochpdefault, div.msochpdefault {mso-style-name:msochpdefault; margin:0cm; margin-bottom:.0001pt; font-size:12.0pt; font-family:"Calibri","sans-serif";} span.emailstyle17 {mso-style-name:emailstyle17; font-family:"Calibri","sans-serif"; color:windowtext;} span.EmailStyle20 {mso-style-type:personal-reply; font-family:"Calibri","sans-serif"; color:#1F497D;} span.TextedebullesCar {mso-style-name:"Texte de bulles Car"; mso-style-priority:99; mso-style-link:"Texte de bulles"; font-family:"Tahoma","sans-serif";} .MsoChpDefault {mso-style-type:export-only; font-size:10.0pt;} @xxxxxxxx WordSection1 {size:612.0pt 792.0pt; margin:70.85pt 70.85pt 70.85pt 70.85pt;} div.WordSection1 {page:WordSection1;} --&gt;</style><!--[if gte mso 9]><xml> <o:shapedefaults v:ext="edit" spidmax="1027" /> </xml><![endif]--><!--[if gte mso 9]><xml> <o:shapelayout v:ext="edit"> <o:idmap v:ext="edit" data="1" /> </o:shapelayout></xml><![endif]--><div class="WordSection1"><p class="MsoNormal"><span style="color:#1F497D">Dear John,<o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D"><o:p>&nbsp;</o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">As others have pointed out, increasing the length of the sweep increases your signal-to-noise ratio. For large room acoustics, &nbsp;we typically use sweeps of 20 to 40 seconds. This of course depends on the size of your room, its reverberation time, and the power of your source. The weaker the source, the longer the sweep. You should basically do a test and see what SNR you get. If you need more, and your measurement chain is at its limit, the only option is longer sweep. Longer sweeps will not help much with impulsive interruptions, while averaging will.With a single sweep, basically that freq-range during the noise is lost. If you are just measuring from the RIR, this may not be an issue, as measurement parameters are often in octave-bands, but this is not true for auralization usage; corrupt data is corrupt data, though I haven&#8217;t gone through a thorough study of this actual case. &nbsp;<o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB"><o:p>&nbsp;</o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">The sweep should definitely be longer than the RT, for room measurements. Then, don&#8217;t forget that you need to record the sweep length PLUS the RT or more if your SNR is better than 60dB!<o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB"><o:p>&nbsp;</o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">We have also compared averaging repeated sweeps vs. longer sweeps. Avoiding the recent developments in overlapping sweep processing, this basic repetition approach &#8216;requires&#8217; the decay of the first sweep to finish before you launch the second sweep. As such, 3x20 second sweeps take longer than 1 x 60 second sweep, due to the additional pauses. Of course, without repetitions, you have no backup in case something goes wrong, like a door slam or something. So, I tend to use repeated sweeps and take the best 1. <o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB"><o:p>&nbsp;</o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">We do all our processing in MatLab, and have never had an issue (within the last 20 years) of processing long sweeps in real halls. <o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB"><o:p>&nbsp;</o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">As you are considering BRIR, and not anechoic HRTFs, you are subject to the same conditions. If you want to convolve the BRIR directly, you will need to ensure that the SNR is sufficiently high that the noise-floor is not audible as a late reverb part of the BRIR. Some noise extension methods exist from older studies on basic auralization and scale model RIR auralizations. I cannot imagine a 2s sweep for BRIR unless you are measuring an office of other room with &lt;1 sec RT. I think there is a fault in your correlation-based analysis for reliability, and there are too many factors to consider in comparing BRIR of different lengths. First, examine your SNR. <o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB"><o:p>&nbsp;</o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">Regarding distortion of the source, this should be an issue for the processing element (different from burning out you speaker). This is because one of the strengths of the sweep method (when done correctly) is that any harmonic distortion components of a higher frequency that the excitation signal at the time or folded back to BEFORE the direct sound after deconvolution by the excitation signal (as that frequency has yet to be generated). We presented a through work on this feature, extending Farina&#8217;s earlier works, to general conditions:<o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB"><o:p>&nbsp;</o:p></span></p><p style="margin-left:35.4pt" class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">M. R&eacute;billat, R. Hennequin, E. Corteel, and B. Katz, &#8220;Identification of cascade of Hammerstein models for the description of nonlinearities in vibrating devices,&#8221; J. Sound and Vibration, vol. 330, pp. 1018&#8211;1038, 2011, (doi:10.1016/j.jsv.2010.09.012).<o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB"><o:p>&nbsp;</o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">This method lets you actually extract and analyse the different harmonic distortions (THD etc.) as well as allowing for the modelling of non-linear responses. &nbsp;<o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB"><o:p>&nbsp;</o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">Cheers,<o:p></o:p></span></p><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">Brian <o:p></o:p></span></p><div><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB">--<o:p></o:p></span></p><p class="MsoNormal"><span style="font-size:10.0pt;color:#1F497D" lang="EN-GB">Brian FG Katz, Ph.D, HDR<o:p></o:p></span></p><p class="MsoNormal"><span style="font-size:10.0pt;color:#1F497D">Resp. Groupe Audio &amp; Acoustique<o:p></o:p></span></p><p class="MsoNormal"><!--[if gte vml 1]><v:shapetype id="_x0000_t75" coordsize="21600,21600" o:spt="75" o:preferrelative="t" path="m@xxxxxxxx@xxxxxxxx@xxxxxxxx@xxxxxxxx@xxxxxxxx@xxxxxxxx@xxxxxxxx@xxxxxxxx" filled="f" stroked="f"> <v:stroke joinstyle="miter" /> <v:formulas> <v:f eqn="if lineDrawn pixelLineWidth 0" /> <v:f eqn="sum @xxxxxxxx 1 0" /> <v:f eqn="sum 0 0 @xxxxxxxx" /> <v:f eqn="prod @xxxxxxxx 1 2" /> <v:f eqn="prod @xxxxxxxx 21600 pixelWidth" /> <v:f eqn="prod @xxxxxxxx 21600 pixelHeight" /> <v:f eqn="sum @xxxxxxxx 0 1" /> <v:f eqn="prod @xxxxxxxx 1 2" /> <v:f eqn="prod @xxxxxxxx 21600 pixelWidth" /> <v:f eqn="sum @xxxxxxxx 21600 0" /> <v:f eqn="prod @xxxxxxxx 21600 pixelHeight" /> <v:f eqn="sum @xxxxxxxx 21600 0" /> </v:formulas> <v:path o:extrusionok="f" gradientshapeok="t" o:connecttype="rect" /> <o:lock v:ext="edit" aspectratio="t" /> </v:shapetype><v:shape id="Image_x0020_2" o:spid="_x0000_s1026" type="#_x0000_t75" style='position:absolute;margin-left:-.25pt;margin-top:.45pt;width:66.55pt;height:34.6pt;z-index:251659264;visibility:visible;mso-wrap-style:square;mso-width-percent:0;mso-height-percent:0;mso-wrap-distance-left:9pt;mso-wrap-distance-top:0;mso-wrap-distance-right:9pt;mso-wrap-distance-bottom:0;mso-position-horizontal:absolute;mso-position-horizontal-relative:text;mso-position-vertical:absolute;mso-position-vertical-relative:text;mso-width-percent:0;mso-height-percent:0;mso-width-relative:page;mso-height-relative:page'> <v:imagedata src="imap://3048314@xxxxxxxx:143/fetch%3EUID%3E/INBOX%3E8726?header=quotebody&part=1.1.2&filename=image001.png" o:title="" croptop="22737f" cropbottom="8693f" /> <w:wrap type="square"/> </v:shape><![endif]--><!--[if !vml]--><img name="image.jpg" v:shapes="Image_x0020_2" src="cid:part2.04070804.09090909@xxxxxxxx" align="left" height="46" hspace="12" width="89"><!--[endif]--><span style="font-size:10.0pt;color:#1F497D">LIMSI - CNRS<br></span><span style="font-size:10.0pt;color:#1F497D">Rue John von Neumann<br>Campus Universitaire d'Orsay, B&acirc;t 508<br>91405 Orsay cedex</span><span style="font-size:10.0pt;color:#1F497D"> <o:p></o:p></span></p><p class="MsoNormal"><span style="font-size:10.0pt;color:#1F497D" lang="EN-US">France<o:p></o:p></span></p><p class="MsoNormal"><span style="font-size:10.0pt;color:#1F497D" lang="EN-US">Phone.&nbsp;</span><span style="font-size:10.0pt;color:#1F497D" lang="EN-GB">+&nbsp; 33 (0)1 69 85 80 67 - Fax.&nbsp; +&nbsp; 33 (0)1 69 85 80 88</span><span style="color:#1F497D" lang="EN-GB"><br></span><span style="color:#1F497D" lang="EN-US"><a moz-do-not-send="true" target="_blank" href="http://www.limsi.fr/"><span style="font-size:10.0pt;color:navy" lang="EN-GB">http://www.limsi.fr</span></a>&nbsp;&nbsp;&nbsp; </span><span style="font-size:10.0pt;color:#1F497D" lang="EN-GB">web_group: </span><span style="font-size:10.0pt;color:#1F497D"><a moz-do-not-send="true" href="http://www.limsi.fr/Scientifique/aa/"><span style="color:blue" lang="EN-GB">http://www.limsi.fr/Scientifique/aa/</span></a></span><span style="font-size:10.0pt;color:#1F497D" lang="EN-GB">&nbsp;&nbsp;&nbsp;&nbsp; web_theme: </span><span style="font-size:10.0pt;color:#1F497D"><a moz-do-not-send="true" href="http://www.limsi.fr/Scientifique/aa/thmsonesp/"><span style="color:blue" lang="EN-GB">http://www.limsi.fr/Scientifique/aa/thmsonesp/</span></a></span><span style="font-size:10.0pt;color:#1F497D" lang="EN-GB"><o:p></o:p></span></p></div><p class="MsoNormal"><span style="color:#1F497D" lang="EN-GB"><o:p>&nbsp;</o:p></span></p><div><div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm"><p class="MsoNormal"><b><span style="font-size:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;">De&nbsp;:</span></b><span style="font-size:10.0pt;font-family:&quot;Tahoma&quot;,&quot;sans-serif&quot;"> AUDITORY - Research in Auditory Perception [<a class="moz-txt-link-freetext" href="mailto:AUDITORY@xxxxxxxx">mailto:AUDITORY@xxxxxxxx</a>] <b>De la part de</b> Anders Tornvig Christensen<br><b>Envoy&eacute;&nbsp;:</b> lundi 27 juillet 2015 09:07<br><b>&Agrave;&nbsp;:</b> <a class="moz-txt-link-abbreviated" href="mailto:AUDITORY@xxxxxxxx">AUDITORY@xxxxxxxx</a><br><b>Objet&nbsp;:</b> Re: Optimal sweep duration for BRIR measurements<o:p></o:p></span></p></div></div><p class="MsoNormal"><o:p>&nbsp;</o:p></p></div></div> </blockquote> </body></html> --------------050307080404060203060206 Content-Type: image/jpeg; x-apple-mail-type=stationery; name="compose-unknown-contact.jpg" Content-Transfer-Encoding: base64 Content-ID: <part1.06010402.07030506@xxxxxxxx> Content-Disposition: inline; filename="compose-unknown-contact.jpg" /9j/4AAQSkZJRgABAQEARwBHAAD/2wBDAAEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEC AQEBAQEBAgICAgICAgICAgICAgICAgICAgICAgICAgICAgL/2wBDAQEBAQEBAQICAgICAgIC AgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgL/wAAR CAAZABkDAREAAhEBAxEB/8QAGAAAAwEBAAAAAAAAAAAAAAAABgcICQr/xAA0EAABAwMCAgUK BwAAAAAAAAACAQMEBQYRABITIQcUMUF2CBUXIjI2N0JRtVRWkZOV0dL/xAAYAQEAAwEAAAAA AAAAAAAAAAADAAEEAv/EACQRAAICAAQGAwAAAAAAAAAAAAABAhEDMrHREyExM0FxgfDx/9oA DAMBAAIRAxEAPwDuEt+gW/ULet6oVC3rfqNQqFv0OfPn1GhUqfOmzZtKZlS5UqZMaNwzNwiJ VIl7eXLCaZIGwBl3TY8epPx2+jy2ZNPjvkwc9uhW8j7nCPhvOsQliYIeS7cvCpp8o50qwrC4 v3lsNSDbdmTEhvs2tahxpfV3WnmbbozJEw/gwdadbYExVRXKEKoSdvJcaOSqxE7/AAiX0gXx +a69/JSf9alIlste0VzaNpeFrcT9KKymotyiaZ0KRCnzacoE7Kjzn4gi2KqUh3jqDHDHv4mR UfruTWlMzlVUKIVNp9GguEJnAh0+IZjyAiisgyRDnu5azS8miKqjOTVkKqS/psG37fo1Fbab eg25b8eZPeFJBBJSjMG5HjMeyihnaauZwe4OGiju13GAcpOwBeN+U8/IkGbsiS8b7ryogmbz hbyc9REROfZhERO5ETShjPtvpGqTUyLErytS4siSwx5x2tRH4hPOI0DkjZtaJtFxuVEbIUUi yeNujlBUJGbJN6nM/Cyf2Hf60YgjvKA+NPSP4gT7axpcPtr51YWJnYn9dnAQWl722p4ot37y zqnlfp6FrqbwawG8/9k= --------------050307080404060203060206 Content-Type: image/jpeg; name="image.jpg" Content-Transfer-Encoding: base64 Content-ID: <part2.04070804.09090909@xxxxxxxx> Content-Disposition: inline; filename="image.jpg" /9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAMCAgICAgMCAgIDAwMDBAYEBAQEBAgGBgUGCQgK CgkICQkKDA8MCgsOCwkJDRENDg8QEBEQCgwSExIQEw8QEBD/2wBDAQMDAwQDBAgEBAgQCwkL EBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBD/wAAR CAAuAFkDAREAAhEBAxEB/8QAHwAAAQUBAQEBAQEAAAAAAAAAAAECAwQFBgcICQoL/8QAtRAA AgEDAwIEAwUFBAQAAAF9AQIDAAQRBRIhMUEGE1FhByJxFDKBkaEII0KxwRVS0fAkM2JyggkK FhcYGRolJicoKSo0NTY3ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqDhIWG h4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXGx8jJytLT1NXW19jZ2uHi4+Tl 5ufo6erx8vP09fb3+Pn6/8QAHwEAAwEBAQEBAQEBAQAAAAAAAAECAwQFBgcICQoL/8QAtREA AgECBAQDBAcFBAQAAQJ3AAECAxEEBSExBhJBUQdhcRMiMoEIFEKRobHBCSMzUvAVYnLRChYk NOEl8RcYGRomJygpKjU2Nzg5OkNERUZHSElKU1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6goOE hYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3uLm6wsPExcbHyMnK0tPU1dbX2Nna4uPk 5ebn6Onq8vP09fb3+Pn6/9oADAMBAAIRAxEAPwD9UKACgDM0fXNH1+z/ALR0XV7TUrXc8fn2 sySxh0OHG9O9OpBx3Abr+vad4Z0HUPEeuXAgsNKtZLy7l2btkUab3P5Cs6abAzfh54+8MfFP wZpvj7wZeyXejaxB9os55IXidkP+w4zTcpx6AdPSUpvoAVN6i6AJ+FL2k10AiuLiKzhe4uJN kMKb3c1q5XlYDkPhX8VfBnxk8JR+NvAGpPfaTNPJbCaSB4G82N9jjZIM1piINRsO52tQhBQA ZHrQHkGaE77AfNX/AAT6QJ+ztaqn/Qw67/6cJa9DMYck0vJfqPoe3fEi+k0z4feJ9TgEfn2e j3lxGHTegdIHccVxYde8hHyJZ/Hb4x3Pwu/Z28C/D3UtC0XxP8VrGR7rW59NSS3tfs8Qkk2W 8f7v95XsVqFNSlJ7Lp8k/wBR9CT45eJf2v8A9n2y8N6lJ8dvDvi1PFetW/huOG+8LR2n2We5 P7ufMcnITHSqwtPD17+7sGxpfGTUf2wv2d/h7dfGPWPj14e8X2OhyW8mpaJP4WjtI545LiOP ZHJF+8/5aVzU1QqvlSsBH8ev2kfHOh/E3SfCMHxb0L4ReG7/AMMW+t2ut6poX9pf2jcSSfPb o/8Aq4/LGK6qODpSjdq7u/63X9MbPbvg/wDtAfCr4oafb+E/D3xe8PeMfEtlp6HUTaDynunR B5lwkB5CF+w4FcFXCSpS5nGyFYyf2MPFeseN/gDovijXzaf2heXuopI1raJBGRFeSxx/u4+P uIKyxy5NBHvFcqAKAOI+LTXdv4A1S507eZ7cR3A2df3bh/6V4+fV62FwvPRXqd+VuKxKrPqd D4f1W113R7XU7R90N1Cjqfw612ZbVp1cLGvTd7oxxtKWDxLgz5l8JfAn9qn4LJqHhX4M/Eb4 f3fhCTUri/sbfxHptx9rtPPlMjpvt+JOXODXv1cVTr2c1qc92d34c8NftSa3Y+J9A+LviD4d z6bq2hXdhZf2BaXcc8d3ImxJJPN48vmTp7VyuUE04hc4zw3+yh430e4/Z7ubjxLor/8ACobS 4g1XZ5n+lPJEI82/yf8AoeK6K2OU1Jd/8kgO7/aa+CGv/GzSfBeneH9Y06xfw74t0/X5/tok PmW9uTvjj8v+M5+lThMWqN/P/gg3c1f2oPhNrfxx+CPiP4YeHdRsbHUdYW3EE96JPIj8u4jk +fy+T/q6wwtdU58zEcz8QPA/7SF0+i6B8Orz4UT6Bp2k29vInifTLue5+2RjY8kYj+Ty8bP1 ro9vGLcle7fcbZzHw8/Zx+LrfGnQvi98VtZ+H1v/AMIvY3lpptj4P0mS1897iPZIZ5JP6V0V MdGVN00t++v6jR6Z+zP8J9Z+CPwh0z4d+INSsb29sLu8neey3+WRPcSSLjfz/wAtK8zFT9tK 6JPWayAKAKOo6fBqNnLp9wm+K4jMbj2rGajeVOrqpFUp+xSiuh418P8AxLJ8L/Es/wAOfFco jsJZ9+l3T/cUP0izXwmVYmpkGYyyetrGeqf4n1+YYJZjgFmEN1o0e51+gK3Q+OegU3Ll3AKh xcgCjkaAKEuXYAou2AVUroApwd0AUAFADSRjNCSpqzIqKzucn4/+Hui+PdO+z30Wy4j5gnT7 8Zrxsxy36+7rR9z08Dj3h3Z/D1R5pYar8VfhTiw1HTJvEWjR/wCrki+/HH7vXzNLDZzw++aT 9rE+jxc8mzWK+rLlkddYfHLwjqEaLcWesWk//PJ7CRyP++Aa9vCZ/h8YuXFwcH8zxamRYmF3 SkrHo8FxHNClxH9yRN9fRqzWh47TTsySmIKACgAoAKACgAoAaMAY7UfGipLmDAC+1HwIxlHl GEo6fMmRTg+dXFSqODsisdLsCRKLOAN6+WK5cRDD01zTgmbVZ4irpzl6uhbaC2CmAUAFABQA UAFABQB//9k= --------------050307080404060203060206-- --------------080004090108030700050505--


This message came from the mail archive
/var/www/postings/2015/
maintained by:
DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University