Subject: searching article, From: bert hubert <ahu(at)DS9A.NL> Date: Mon, 24 May 2004 23:50:25 +0200Dear researchers, I'm doing research on (very) small vocabulary speaker independent untrained speech recognition [1] and left the well travelled path of FFT for simulated oscillators. Faced with the 'Cochlear Paradox' [2] of conflicting goals for Q (frequency specificity) and Tau (relaxation time), I'm looking for biological inspiration. In trying to get into phase with the field, I find a lot of work from differing timeperiods, some frankly contradicting. It appears the CA controversy has died out and that the cochlea is now widely regarded as being 'active', but other issues appear to be less balanced. So what I'm looking for is a recommended reference to an article summarizing recent findings regarding the 'active cochlea'. I'm less interested in the full biological gory details than in the mechanical workings. I also read http://www.bcm.tmc.edu/oto/research/cochlea/Volta/16.html and I'm wondering if this reflects the state of the art. Somewhat older papers also mention gammatone filterbanks, are these still hot? Thank you for your time. I don't have a lot to offer in return, but some of you may be able to benefit from http://ds9a.nl/gcc-simd/ 'SIMD and other techniques for fast numerical programming with gcc'. Bert Hubert [1] think digits, plus, minus, times, equals etc etc, 20 words perhaps [2] http://www.vimm.it/cochlea/cochleapages/theory/sndproc/paradox.htm -- http://www.PowerDNS.com Open source, database driven DNS Software http://lartc.org Linux Advanced Routing & Traffic Control HOWTO