[AUDITORY] Speech Intelligibility Index vs. Direct Measurement ("Richard M. Warren" )

Subject: [AUDITORY] Speech Intelligibility Index vs. Direct Measurement From: "Richard M. Warren" <rmwarren@xxxxxxxx> Date: Fri, 8 Nov 2013 14:39:12 -0600 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY> --Apple-Mail-55--238456964 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 The Speech Intelligibility Index (SII) [1] procedure has been used for = many years to determine the intelligibility of passbands heard singly = and in combination. The SII is a complex multi-step procedure that = initially cancels contributions from filter slopes by combining = intelligibility judgments of partially masked high-pass and low-pass = speech having a series of cut-off frequencies. This data is used to = calculate the band importance values, which then can be used to produce = band intelligibility estimates. Studebaker and Sherbecoe (1991) [2] = employed a commercially available broadband recording of W-22 word lists = to produce band importance estimates that are listed in Table B.3 of the = SII. Recently, J.M. Kates (2013) [3] used the data reported by = Studebaker and Sherbecoe to produce an improved method used to determine = SII importance values and intelligibility estimates shown in his Figures = 1 and 3.=20 But rather than trying to improve the SII procedure, Warren, Bashford, = and Lenz (2011) [4] proposed a direct measurement alternative: The = Rectangular Passband Intelligibility (RPI) procedure. We employed the = same copy of the commercially available recording of W-22 lists employed = by Studebaker and Sherbecoe. Instead of generating high-pass and = low-pass speech, we used high-order FIR filtering to produce effectively = vertical slopes for passbands at different center frequencies. = Listeners were then able to use these rectangular passbands to obtain = direct intelligibility measurements of the same passbands heard singly = and in the combinations reported using the SII procedure. The relative = intelligibilities obtained were in general similar using the two = different procedures as can be seen in the Warren et al. Figures 5 and = 6. Also, the RPI procedure is not limited to word lists: It had been = used previously with sentences (Warren, Bashford, & Lenz, 2005) [5]. The RPI has several advantages over the SII procedure. These include: = (1) the intrinsic advantage of direct measurement over estimation; (2) = the RPI is simpler to use, eliminating the requirement for measuring = intelligibility of high-pass and low-pass speech heard at various = signal/noise ratios. In addition, there is no need for calculating the = intermediate stage of importance values followed by the requirement of a = transfer function in order to obtain intelligibility estimates; (3) the = RPI procedure is the only one that can obtain passband intelligibility = judgments under quiet conditions. This makes it possible to quantify = the decrement produced by extraneous sounds and distortions. =20 REFERENCES=20 [1.] ANSI S3.5, 1997, Reaffirmed, 2012. =93Methods for the Calculation = of the Speech Intelligibility Index,=94 (American National Standards = Institute, New York). [2.] Studebaker, G.A., & Sherbecoe, R.L. =93Frequency-Importance and = Transfer Functions for Recorded CID W-22 Word Lists,=94 Journal of = Speech and Hearing Research, 1991, Vol. 34, 427-438.=20 [3.] Kates, J.M. =93Improved Estimation of Frequency Importance = Functions,=94 Journal of the Acoustical Society of America, 2013, Vol. = 134, EL459-EL464. [4.] Warren, R.M., Bashford, J.A., Jr., & Lenz, P.W. =93An Alternative = to the Computational Speech Intelligibility Index Estimates: Direct = Measurement of Rectangular Passband Intelligibilities,=94 Journal of = Experimental Psychology: Human Perception and Performance, 2011, Vol. = 37, 296 - 302. [5.] Warren, R.M., Bashford, J.A., Jr., & Lenz, P.W. =93Intelligibilities= of 1-Octave Rectangular Bands Spanning The Speech Spectrum When Heard = Separately and Paired.=94 Journal of the Acoustical Society of America, = 2005, Vol. 118, 3261-3266. =20 Richard M. Warren=20 Research Professor=20 and Distinguished Professor Emeritus Department of Psychology University of Wisconsin-Milwaukee PO Box 413 Milwaukee, WI 53201 (414) 229-5328 --Apple-Mail-55--238456964 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=windows-1252 <html><head></head><body style=3D"word-wrap: break-word; = -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">    <p class=3D"MsoNormal"><font = class=3D"Apple-style-span" face=3D"'Times New Roman'"><span = class=3D"Apple-style-span" style=3D"font-size: 16px;">The Speech = Intelligibility Index (SII) [1] procedure has been used for many years to determine the intelligibility of passbands heard singly and in combination. The SII is = a complex multi-step procedure that initially cancels contributions from = filter slopes by combining intelligibility judgments of partially masked = high-pass and low-pass speech having a series of cut-off frequencies.  This data = is used to calculate the band importance values, which then can be used to produce band = intelligibility estimates.  Studebaker and Sherbecoe (1991) [2] employed a commercially available broadband recording of W-22 = word lists to produce band importance estimates that are listed in Table B.3 = of the SII.  Recently, J.M. Kates (2013) [3] used the data reported by Studebaker and Sherbecoe to produce an improved method = used to determine SII importance values and intelligibility estimates shown in = his Figures 1 and 3. </span></font></p><p class=3D"MsoNormal"><font = class=3D"Apple-style-span" face=3D"'Times New Roman'"><span = class=3D"Apple-style-span" style=3D"font-size: 16px;">But rather than = trying to improve the SII procedure, Warren, Bashford, and Lenz (2011) [4] = proposed a direct measurement alternative:<span style=3D"mso-spacerun:yes">  = </span>The Rectangular Passband Intelligibility (RPI) procedure.<span = style=3D"mso-spacerun:yes">  </span>We employed the same copy of = the commercially available recording of W-22 lists employed by Studebaker and = Sherbecoe.<span style=3D"mso-spacerun:yes">  </span>Instead of = generating high-pass and low-pass speech, we used high-order FIR filtering to produce effectively vertical = slopes for passbands at different center frequencies.<span = style=3D"mso-spacerun:yes">  </span>Listeners were then able to use these rectangular passbands to = obtain direct intelligibility measurements of the same passbands heard singly = and in the combinations reported using the SII procedure.<span = style=3D"mso-spacerun:yes">  </span>The relative intelligibilities obtained were in general similar = using the two different procedures as can be seen in the Warren et al. Figures = 5 and 6.<span style=3D"mso-spacerun:yes">  </span>Also, the RPI procedure = is not limited to word lists:<span style=3D"mso-spacerun:yes">  </span>It = had been used previously with sentences (Warren, Bashford, & Lenz, 2005) = [5].</span></font></p><p class=3D"MsoNormal"><font = class=3D"Apple-style-span" face=3D"'Times New Roman'"><span = class=3D"Apple-style-span" style=3D"font-size: 16px;">The RPI has = several advantages over the SII procedure.<span style=3D"mso-spacerun:yes">  = </span>These include: (1) the intrinsic advantage of direct measurement over = estimation; (2) the RPI is simpler to use, eliminating the requirement for measuring intelligibility of high-pass and low-pass speech heard at various = signal/noise ratios.<span style=3D"mso-spacerun:yes">  </span>In addition, there = is no need for calculating the intermediate stage of importance values followed by the = requirement of a transfer function in order to obtain intelligibility estimates; (3) = the RPI procedure is the only one that can obtain passband intelligibility = judgments under quiet conditions.<span style=3D"mso-spacerun:yes">  </span>This = makes it possible to quantify the decrement produced by extraneous sounds and = distortions.</span></font></p><p class=3D"MsoNormal"><o:p = style=3D"font-size: 16px;"><font class=3D"Apple-style-span" face=3D"'Times= New Roman'"> </font></o:p></p><p class=3D"MsoNormal"><font = class=3D"Apple-style-span" face=3D"'Times New Roman'"><span = class=3D"Apple-style-span" style=3D"font-size: = 16px;">REFERENCES </span></font></p><p class=3D"MsoNormal" = style=3D"mso-pagination:none;mso-layout-grid-align:none; text-autospace:none"><font class=3D"Apple-style-span" face=3D"'Times New = Roman'"><span class=3D"Apple-style-span" style=3D"font-size: 16px;">[1.] = ANSI S3.5, 1997, Reaffirmed, 2012. =93Methods for the Calculation of the = Speech Intelligibility Index,=94 (American National Standards Institute, New = York).</span></font></p><p class=3D"MsoNormal"><font = class=3D"Apple-style-span" face=3D"'Times New Roman'"><span = class=3D"Apple-style-span" style=3D"font-size: 16px;">[2.] Studebaker, = G.A., & Sherbecoe, R.L.<span style=3D"mso-spacerun:yes">  </span>=93Frequency-Importance= and Transfer Functions for Recorded CID W-22 Word Lists,=94 <u>Journal of Speech and = Hearing Research</u>, 1991, Vol. 34, 427-438. </span></font></p><p = class=3D"MsoNormal"><font class=3D"Apple-style-span" face=3D"'Times New = Roman'"><span class=3D"Apple-style-span" style=3D"font-size: 16px;">[3.] = Kates, J.M. =93Improved Estimation of Frequency Importance Functions,=94 = <u>Journal of the Acoustical Society of America</u>, 2013, Vol. 134, EL459-EL464.</span></font></p><p class=3D"MsoNormal"><font = class=3D"Apple-style-span" face=3D"'Times New Roman'"><span = class=3D"Apple-style-span" style=3D"font-size: 16px;">[4.] Warren, R.M., = Bashford, J.A., Jr., & Lenz, P.W. <span style=3D"mso-spacerun:yes"> </span>=93An = Alternative to the Computational Speech Intelligibility Index Estimates: Direct Measurement = of Rectangular Passband Intelligibilities,=94 <u>Journal of Experimental = Psychology: Human Perception and Performance</u>, 2011, Vol. 37, 296 - = 302.</span></font></p><p class=3D"MsoNormal"><font = class=3D"Apple-style-span" face=3D"'Times New Roman'"><span = class=3D"Apple-style-span" style=3D"font-size: 16px;">[5.] Warren, R.M., = Bashford, J.A., Jr., & Lenz, P.W.<span style=3D"mso-spacerun:yes">  = </span>=93Intelligibilities of 1-Octave Rectangular Bands Spanning The Speech Spectrum When Heard = Separately and Paired.=94<span style=3D"mso-spacerun:yes">  </span><u>Journal = of the Acoustical Society of America</u>, 2005, Vol. 118, = 3261-3266.<o:p></o:p></span></font></p><p class=3D"MsoNormal"><font = class=3D"Apple-style-span" face=3D"'Times New Roman'"><span = class=3D"Apple-style-span" style=3D"font-size: = 16px;"><o:p> </o:p> </span></font></p> <div> <div style=3D"font-size: 13px; "><div><font class=3D"Apple-style-span" = size=3D"4"><span class=3D"Apple-style-span" style=3D"font-size: 14px; = "><b><br class=3D"Apple-interchange-newline">Richard M. = Warren</b></span></font> </div><div>Research = Professor </div><div>  and Distinguished Professor = Emeritus</div><div>Department of Psychology</div><div>University of = Wisconsin-Milwaukee</div><div>PO Box 413</div><div>Milwaukee, WI =  53201</div></div><div style=3D"font-size: 13px; ">(414) = 229-5328</div><div><br></div><br class=3D"Apple-interchange-newline"> </div> <br></body></html>= --Apple-Mail-55--238456964--

This message came from the mail archive
/var/www/postings/2013/
maintained by:

DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University