Subject: Re: [AUDITORY] Everyone knows that ... From: Pierre Divenyi <pdivenyi@xxxxxxxx> Date: Mon, 1 Jul 2013 22:56:46 -0700 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>> This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. --B_3455564213_315182 Content-type: text/plain; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable Matt is right. Woods DL, Yund EW, Herron TJ & Ua Cruadlaoich MA (2010), JAS= A 127(3), 1609-23 also looked at the threshold SNR of individual consonants i= n CVC syllables presented in speech-spectrum noise, although they did not compare their consonant thresholds with those of vowels alone. On 7/1/13 9:31 PM, "Matt Winn" <mwinn83@xxxxxxxx> wrote: Stuart, Phatak and Allen (2007) looked at English consonants and a small number of vowels. That might be a good place to start. Most of the literature I know looks only at consonants (sometimes only at consonant manner & place contrasts). Phatak, S. A., and Allen, J. B. (2007), =B3Consonant and vowel confusions in speech-weighted noise,=B2 J. Acoust. Soc. Am. 121, 2312=AD2316. =20 If I may add my two cents though, I think that the comparison between Cs an= d Vs in this situation is not exactly a fair one, since they are inherently different in intensity. Traditionally, we define SNR over the entire stimulus, but a single CVC syllable at a given SNR can be broken down into two Cs that have a much less-favorable SNR than the vowel nucleus. So, at a nominal SNR, we are actually presenting the vowel at a much more favorable SNR. That ought to account for some of the robustness of vowels in noise. =20 On the other hand, if you=B9re simply interested in which elements of the signal are robust at a given long-term SNR, none of that matters. Matt On Mon, Jul 1, 2013 at 5:59 AM, Stuart Rosen <s.rosen@xxxxxxxx> wrote: > the identification of consonants is more affected by speech-shaped noise = than > vowels, but can someone please provide a citation to a paper that discuss= es > this issue thoroughly? A classic paper would be ideal. >=20 > Thanks for any help. >=20 > Yours - Stuart >=20 > P.S. I have been looking around for a while. >=20 > P.P.S. Surely Jont will let me know that it is all in Fletcher's book! --B_3455564213_315182 Content-type: text/html; charset="ISO-8859-1" Content-transfer-encoding: quoted-printable <html><head></head><body style=3D"word-wrap: break-word; -webkit-nbsp-mode: s= pace; -webkit-line-break: after-white-space; color: rgb(0, 0, 0); font-size:= 16px; font-family: Calibri, sans-serif; "><div>Matt is right. Woods DL, Yun= d EW, Herron TJ & Ua Cruadlaoich MA (2010), JASA 127(3), 1609-23 al= so looked at the threshold SNR of individual consonants in CVC syllables pre= sented in speech-spectrum noise, although they did not compare their consona= nt thresholds with those of vowels alone.</div><div><br></div><span id=3D"OLK_= SRC_BODY_SECTION"><div><div>On 7/1/13 9:31 PM, "Matt Winn" <<a href=3D"mail= to:mwinn83@xxxxxxxx">mwinn83@xxxxxxxx</a>> wrote:</div></div><div><br><= /div><div dir=3D"ltr"><p class=3D"">Stuart,</p><p class=3D"">Phatak and Allen (200= 7) looked at English consonants and a small number of vowels. That might be a good place to start. Most of the li= terature I know looks only at consonants (sometimes only at consonant manner= & place contrasts).</p>Phatak, S. A., and Allen, J. B. (2007), “C= onsonant and vowel confusions in speech-weighted noise,” J. Acoust. So= c. Am. 121, 2312–2316.<br><p class=3D""> </p><p class=3D"">If I may a= dd my two cents though, I think that the comparison between Cs and Vs in this situation is not exactly a fair one, s= ince they are inherently different in intensity. Traditionally, we define SNR ov= er the entire stimulus, but a single CVC syllable at a given SNR can be broken= down into two Cs that have a much less-favorable SNR than the vowel nucleus= . So, at a nominal SNR, we are actually presenting the vowel at a much more favor= able SNR. That ought to account for some of the robustness of vowels in noise. <= /p><p class=3D""> </p><p class=3D"">On the other hand, if you’re simp= ly interested in which elements of the signal are robust at a given long-term SNR, none of that matters. </p><p class=3D"">Matt</p></div><div class=3D"gmail_extra"><br><br><di= v class=3D"gmail_quote">On Mon, Jul 1, 2013 at 5:59 AM, Stuart Rosen <span dir= =3D"ltr"><<a href=3D"mailto:s.rosen@xxxxxxxx" target=3D"_blank">s.rosen@xxxxxxxx= .uk</a>></span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"margin:0= 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">the identification of= consonants is more affected by speech-shaped noise than vowels, but can som= eone please provide a citation to a paper that discusses this issue thorough= ly? A classic paper would be ideal.<br><br> Thanks for any help.<br><br> Yours - Stuart<br><br> P.S. I have been looking around for a while.<br><br> P.P.S. Surely Jont will let me know that it is all in Fletcher's book!<br><= /blockquote></div><br></div></span></body></html> --B_3455564213_315182--