From - Tue Sep 19 09:26:46 2000 Return-Path: Received: from mail-blue.research.att.com (mail-blue.research.att.com [135.207.30.102]) by hermes.research.att.com (8.8.7/8.8.7) with ESMTP id EAA13554 for ; Tue, 19 Sep 2000 04:53:41 -0400 (EDT) Received: by mail-blue.research.att.com (Postfix) id E667E4CE19; Tue, 19 Sep 2000 04:56:14 -0400 (EDT) Delivered-To: jba@research.att.com Received: from [130.227.121.254] (gateway.widex.com [130.227.121.254]) by mail-blue.research.att.com (Postfix) with SMTP id 129624CE15 for ; Tue, 19 Sep 2000 04:56:13 -0400 (EDT) Received: from mail.widex.com by [130.227.121.254] via smtpd (for H-135-207-30-102.research.att.com [135.207.30.102]) with SMTP; 19 Sep 2000 08:56:13 UT Received: from MAH ([89.0.1.10]) by mail.widex.dk (Lotus Domino Release 5.0.1b (Intl)) with ESMTP id 2000091910534765:1175 ; Tue, 19 Sep 2000 10:53:47 +0100 From: Martin Hansen Message-ID: <14791.14391.460000.326530@widex.dk> Date: Tue, 19 Sep 2000 10:56:07 +0100 MIME-Version: 1.0 To: Jont Allen Subject: forwarded message from Rick Schmiedt X-Mailer: VM 6.76 under Emacs 20.4.1 X-MIMETrack: Itemize by SMTP Server on WIDEX_DK_02/DK/WIDEX(Release 5.0.1b (Intl)|30 September 1999) at 19-09-2000 10:53:47, Serialize by Router on WIDEX_DK_02/DK/WIDEX(Release 5.0.1b (Intl)|30 September 1999) at 19-09-2000 10:53:53, Serialize complete at 19-09-2000 10:53:53 X-Priority: 3 (Normal) Content-Transfer-Encoding: 7bit Content-Type: multipart/mixed; boundary="GKvxEj9f0V" X-Mozilla-Status: 8001 X-Mozilla-Status2: 00000000 X-UIDL: Hi^"!&4i!!iT`!!<0#"! --GKvxEj9f0V Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Content-Description: message body text Dear Jont, you asked me to let you know about any answers from Rick Schmiedt. He answered right away and sent me a longer list of references to his research which I can share with you: Regards, Martin P.S. I still don't see your point in the AI argument. > > Now, how do you calculate the signal-to-noise ratio, if you don't have > > acces to both the signal power spectrum and the noise power spectrum? > > I dont think that is the point. It isnt the SNR that predicts the > words. That is, it is not meaningful to think of the SNR as the > measure of the sounds. The SNR predicts the phoneme score, but not the > actual sound. Since the spectrum can change without changing the SNR, > it follows that the spectrum is not the significant measure of the > sounds. > > Take the case where the SNR is 30. Then the score will be maximum (i.e., > 98.5% correct). > Any filter that does not change the SNR will not effect the final score. > > Having a way to measure the SNR is not the point. Are we talking about the same "SNR" thing? I am considering the SNR in narrow frequency bands. How can you change the spectrum of speech without changing the SNR in some bands, if you don't equally change the noise spectrum? (The "SNR" I'm refering to is the ratio of the band-specific energy of speech relative to the energy of (all) masking noise(s), including across-band masking effects of noise and self-masking effects of speech. In the calculation procedure for AI/SII, the speech spectrum and the noise spectrum are taken as input and a kind of "SNR" is then calculated from these two measures.) --GKvxEj9f0V Content-Transfer-Encoding: quoted-printable Content-Type: message/rfc822 Content-Description: forwarded message MIME-Version: 1.0 Content-type: text/plain; charset=iso-8859-1 ---------- start of forwarded message ---------- From: Rick Schmiedt To: Martin Hansen , Rick Schmiedt Subject: Re: presbyacusis Date: Thu, 14 Sep 2000 16:23:07 -0400 Dear Dr. Hansen: Thank you for your interest in my work. Our group has had a large prog= ram grant studying human and animal presbyacusis for the last 15 years, and= one of the main findings is how much the endocochlear potential plays a rol= e in age-related hearing loss. Below are some articles for your reference. Also are some abstracts that reflect our current findings. We have been talking up the role of the EP for a long while. I guess I= needed to finally convince Jont that there might be something to it -- = he is now our best salesman on the topic! Best regards, Rick Schmiedt Schulte BA, Schmiedt RA (1992). Lateral wall Na,K-ATPase and endocochl= ear potentials decline with age in quiet-reared gerbils. Hear Res 61, 35-46. Schmiedt RA (1993). Cochlear Potentials in quiet-aged gerbils: Does = the aging cochlea need a jump start? In: R.T. Verrillo, ed. Sensory Research Multimodal Perspectives. New York= ; Erlbaum and Associates, 91-103. Boettcher FA, Mills JH, Norton BN, Schmiedt RA (1993). Age-related cha= nges in evoked potentials of gerbils. II. Response latencies. Hear Res 71, 146-156. Boettcher FA, Mills JH, Dubno JR, Schmiedt RA (1995). Masking of auditory brainstem responses in young and aged gerbils. Hear Res 89, 1-13. He N, Schmiedt RA (1996). Effects of Aging on the fine structure of th= e 2f1-f2 acoustic distortion product. J Acoust Soc Amer 99, 1002-1015. Gratton MA, Schmiedt RA, Schulte BA (1996). Age-related decreases in endocochlear potential are associated with vascular abnormalities in the stria vascularis. Hear Res 94, 116-124. Hellstrom LI, Schmiedt RA (1996). Frequency selectivity in single-fib= er and whole-nerve tuning curves in young and aged gerbils. J Acoust Soc Amer 100, 3275-3285. Schmiedt RA, Mills JH, Boettcher FA (1996). Age-related loss of activ= ity of auditory-nerve fibers. J Neurophysiol 76, 2799-2803. Schmiedt RA (1996). Effects of aging on potassium homeostasis and the endocochlear potential in the gerbil cochlea. Hear Res 102, 125-132. Gratton MA, Smyth BJ, Lam C, Boettcher FA, Schmiedt RA (1997). Decli= ne in the endocochlear potential corresponds with decreased Na,K-ATPase activity in the lateral wall o= f quiet-aged gerbils. Hear Res 108, 9-16. Some abstracts: Schmiedt, R.A. "Old before its time -- modeling the presbyacusic = ear in a young animal". Association for Research in Otolaryngology, St. Petersbu= rg, FL, February, 1997. Schmiedt, R.A., Spicer, S.S., and Schulte, B.A. "Morphologic and= physiologic changes in gerbil cochleas exposed to chronic infiltration of furosemide i= nto perilymph." Association for Research in Otolaryngology, St. Petersburg, Fl. February, 1999. Okamura, H.-O., Schmiedt, R.A., Spicer, S., and Schulte, B.A. "Morphologic and Physiologic changes in gerbil cochleas induced by chronic exposure= to furosemide." Association for Research in Otolaryngology, St. Petersburg, Fl. February, 2000. Schmiedt, R.A., Lang, H., and Okamura, H.-O. "Model of metabolic presbyacusis: Neural responses under conditions of chronically-low endocochlear potentials induced with furosemide." Association for Research in Otolaryngol= ogy, St. Petersburg, Fl. February, 2000. Schmiedt, R.A. and Lang, H. "Functional changes in the ear with old = age: A review." J. Acoust. Soc. Am.:107:2797, 2000. (Invited.) Martin Hansen wrote: > Dear Richard Schmiedt, > > Jont Allen pointed me to an interesting piece of work you did. It is= > your research about the origin of presbyacusis, which according to > you, if I understand it right, might be explained by a loss of > endocochlear potential. Jont Allen told me about experiments where > the presbyacusis hearing loss could be reversed by an external > intercellular increase of the voltage which can lift the thresholds > back to normal level. > > He pointed me to the abstract in JASA, Vol 107, #5, page 2797, May 20= 00. > > As I'm doing research on audiology and signal processing in a hearing= > aid company, I find this work very interesting. I would therefore li= ke > to ask you if you have published further details on this research, > e.g., in an article. I would very much appreciate if you could direc= t > me towards more information on this topic. > > Kind regards, > Martin Hansen > > ----------------------------------------------------------------- > Martin Hansen, PhD Toepholm & Westermann ApS > AUDLAB > Email : m.hansen@widex.dk Ny Vestergaardsvej 25 > Phone : (+45) 44 35 57 43 DK-3500 Vaerloese > Fax : (+45) 44 35 56 03 DENMARK > -----------------------------------------------------------------= ---------- end of forwarded message ---------- --GKvxEj9f0V--