Subject: Re: *cochlea*: inner ear models in Python From: Marek Rudnicki <marek.rudnicki@xxxxxxxx> Date: Fri, 22 Aug 2014 16:46:21 +0200 List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>--=-=-= Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Hi Lowel, "Dr. Lowel P. O'Mard" <lowel.omard@xxxxxxxx> writes: > Hi Etienne and Everybody, > > Just to remind everybody that the Development System for Auditory Modelli= ng > (DSAM) RunDSAMSim application also provides a Python interface, and allows > complete access to all of the models available in DSAM. No doubt > "Cochlea*" is a worthy competitor to the DSAM RunDSAMSim application, as = it > seems to have similar functionality. The RunDSAMSim python interface lin= ks > directly to the C/C++ DSAM library. Thank you for reminding about DSAM. It's a very reach library of auditory models. I used to work with it and you can probably still find some DSAM components in the old commits. In my opinion, those two libraries complement each other, rather than compete, because of the following reasons: 1. DSAM and *cochlea* implement different models. Perhaps, at the moment there is not even a single intersection. DSAM includes: - Zilany and Bruce (2006, 2007) - Meddis et al. (2001) - Patterson et al. (2001) *cochlea* includes: - Zilany et al. (2009, 2014) - Holmberg (2007) - Meddis (MAP, externally) 2. DSAM models have higher granularity, which makes it easier to tweak them. Models in *cochlea* are integrated in a single unit and are easier to use on the higher level: sound in, spikes out. 3. DSAM implements higher stages of auditory processing (McGregor model). *cochlea* does not implement any neurons. So far, we used Brian and NEURON simulators for this purpose. I have included DSAM and other implementations of inner ear models in a small list on the website [1]. If you have a reference to any other implemented model, please let me know and I'll be happy to include it. Best regards Marek Rudnicki [1] https://github.com/mrkrd/cochlea#other-implementations > DSAM provides a flexible environment for both novice and advanced users to > run auditory models on all platforms (Windows =E2=84=A2, Ubuntu, Fedora, = Mac OS, > etc.). It provides immediate access to auditory models such as the Zilany > and Bruce (2006, 2007), Meddis et al. (2001) auditory nerve models and the > Auditory Image Model (AIM) by Patterson et al. (2001). In all it provides > seven different published auditory nerve models using both linear and > non-linear basilar membrane frequency selectivity, including the DRNL and > Gamma-Chirp models. It also provides several neural cell models including > an ultra-fast implementation of the Hudgkin-Huxley neural cell which > operates at 80% the speed of a simple point neuron model (the McGregor > Model). Other auditory models, analysis functions, threaded processing and > sound file support are also provided. A 320 page manual is available for > DSAM that provides detailed information on how simulations can be created > and controlled. > In 2012 DSAM joined the ever growing band of scientists who are turning to > Python as their analysis and visualisation programming language. Because > the standard Python =E2=80=9Cctypes=E2=80=9D foreign function library is = used to create the > interface DSAM can be run on any platform for which Python is available. > Self-installing packages (Windows[tm], Ubuntu and Fedora) for RunDSAMSim , > SAMS , the DSAM_SDK and other source archives are available from the > =E2=80=9CDownloads=E2=80=9D page of the DSAM website: http://dsam.org.uk.= The > =E2=80=9CApplications=E2=80=9D page provides information on the afore men= tioned application > packages. > > Sincere regards, > > ...Lowel. --=-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- iQIcBAEBAgAGBQJT91e+AAoJEP+xXflpqb4mgusQAKUccXy3ZQsgxUI+ofCFl9X9 Hs+nNcZAsoukGlAaCe6VJHos9tkkXWHjYnfPLqEz/fMZtE4qT4PdjRSMFlhF37IR oHvDDC8ZSHgEdIcLK2Fd/8YCirAyIEQle0rrlna1FshA1wcYsbvfa3FzaX8UQehH 38oNzxfU2BfZ4hzD30uCX6T/MbJdjf24wRS1du+RRCgZpviVjSTzFdb8VpPZk/k1 UDXYMi1CQ4ubTppOS4klfIKKQcFbTrYT1PBLC59TwRXXynZF9/RouG5VvDF0w/XW Z5yKdRGt3PWpJE81bg1bKiWMfa2CxaxjkSRy99KW3lfhec4dmX71Eps4tGisu/Le wyu8JNLsnOUYAy2yXV1iGNiV1Y+Ycd9o+rX4I/03arYa+MYVkA31clVbd4QQ35dD IqyAvYoOfnlwUwUN2g1qK1H4DWIFcJQegph2YN+mZW6heOb6khoLoTvNSqxXrhdp /+uUgeQMKlI3OubORpPLRuc/U65Bv01Pq33Noq2VCbYLeO7sTwpRDr57WwzSmBPS i/ERdvBvVn4851HM2v0k72q3CBRcw5KazKwAahdJrPD9yJ0bEgsTAGUhFlvkIdHg 7Nv1zJmfUYMc03l5Uq9mYqOGIoRf0MVgZ83DjhBIxsc59mgUbNrZKBUnqAphWVd5 Fr/9iqSeiKj1sf4I6++o =/1tT -----END PGP SIGNATURE----- --=-=-=--