Resume

Noam Amir

Noam Amir
My own home page!

Who I am:

Dr. Noam Amir
Lecturer
Department for Communications Disorders - Speech, Language and Hearing
Sackler Faculty of Medicine
Tel Aviv University
---------------------------------
tel: +972-3-5349817 ext. 109
fax: +972-3-5352868
noama@post.tau.ac.il

What you'll find here:

My research interests	Research projects I'm working on	Student projects done under my supervision
Publications	Courses I teach
My Resume	Contacting me	Something I do 30 days a year - the VAV pag

The SEDI project

The student zone

DSP
at a DISTANCE

Matlab
GUI's
for all

Research Interests

I'm interested in lots of things, among them speech, acoustics and signal processing. To be more specific:

Prosody and its interaction with emotions
Musical acoustics - brass instruments, didjeridus, reconstructing wind instrument bores from their acoustical properties.
Acoustic waveguides: solving the exact wave equation in varying cross section waveguides

Research projects

Unfortunately I don't have the time to pursue all the projects I'd like to. So some projects are under way, and some are waiting.

Formant corrected pitch shifting:

I recently became interested in this subject, following an article about it by Keith Lent, in CMJ. I found another one about it by Bristow-Johnson in JAES. Reading these two set me rolling, and I reached the conclusion that they both evade a rigorous treatment of the problem. My own conclusions were that an exact algorithm for formant corrected pitch shifting doesn't exist! Read all about it in the paper I submitted to JAES (also zippped postscript).

I'd like to do some further work on the subject - like examining the degree to which various musical instruments keep their formants fixed in different registers. This is waiting.

Extracting emotions from speech:

This is something I've been working on for some time now, together with Dr. Shmuel Ron. Actually it was his idea, and he's patented some applications for it. The idea is to take a speech signal and try to find the emotional state of the speaker, given a neutral recording as a baseline. This kind of information is present in the prosody, as opposed to the actual textual information in the signal. We have a prototype working in real time on a pentium, but there is a lot left to improve. We're working on it.

Other projects waiting for me:

Multimodal inverse scattering
Masking (and models of masking) of speech and complex tones by noise
JND thresholds for various components of prosody

Interesting student projects I'm supervising or have supervised:

Term papers in the course "Speech Processing":

Term papers in the course "Advanced Signal Processing":

Publications

        2. N. Amir, G. Rosenhouse, U. Shimony
             "Input impedance of musical horns and the 'horn function'"
             Applied Acoustics, Vol.38, 1993 (pp.15-35)

        3. B. Gazengel, J. Gilbert, N. Amir
             "From the measured input impedance to the synthesized signal: where are the traps?"
            Acta Acustica, Vol.3, 1995 (pp.445-472)

        4. N. Amir, G. Rosenhouse, U. Shimony
             "A discrete model for tubular acoustic systems with varying cross sections -
             the direct and inverse problems. Part 1: theory"
             Acustica, Vol.81, No.5, 1995 (pp.450-462)

         5. N. Amir, G. Rosenhouse, U. Shimony
             "A discrete model for tubular acoustic systems with varying cross sections -
             the direct and inverse problems. Part 2: experiments"
             Acustica, Vol.81, No.5, 1995 (pp.463-479)

         6. N. Amir, G. Rosenhouse, U. Shimony
             "Losses in tubular acoustic systems - theory and experiment"
             Acustica, Vol.82, No.1, 1996 (pp.1-8)

        7. V. Pagneux, N. Amir, J. Kergomard
             "A study of wave propagation in varying cross section waveguides
             using modal decomposition: Part 1: Theory and Validation"
             Jour. Acoust. Soc. Am., Vol.100, No.4, 1996 (pp.2034-2048)

        8. N. Amir, R. Starobinsky
             "Finding the eigenmodes of two dimensional cavities with two axes of symmetry"
             Acustica, Vol.82, No.6, 1996 (pp.811-822)

        9. N. Amir, V. Pagneux, J. Kergomard
             "A study of wave propagation in varying cross section waveguides
             using modal decomposition: Part 2: Experiments"
             Jour. Acoust. Soc. Am., Vol.101, No.5, 1997 (pp.2504-2517)

         10. N. Amir, H. Matzner, S. Shtrikman
             "Study of the acoustics of a flanged cylindrical pipe using singular basis
            functions" Jour. Acoust. Soc. Am., Vol.107, No.2, 2000 (pp.714-724)

         11. J. Kemp, D. Campbell, N. Amir
             "Multimodal radiation impedance of a rectangular duct terminated in an infinite baffle"
             Acustica., Vol.87, No.1, 2001 (pp.11-15)

There a few more actually, which will get to here one day...

Courses I teach

Acoustics for speech and hearing
Electricity and electroacoustics
Acoustic and electroacoustic instrumentation
Speech and acoustics lab
Signal processing for speech and hearing (graduate)
Basic signal processing for medical sciences (graduate)

Something I do (well did, actually) 30 days a year - the VAV page

Most Israeli men are away for about 30 days a year. Here's where I go.

So how did I get to be what I am today? This is my personal history:

Education:

D.Sc. in EE, Technion - Israel Institute of Technology, 1993

"Acoustic analysis and shape reconstruction of brass wind instruments"

M.Sc. in EE, Technion - Israel Institute of Technology, 1989

"Segmentation of EEG signals during sleep"

B. Sc. in EE, Technion - Israel Institute of Technology, 1985

Experience

Postdoc

I did my postdoc at the Laboratoire d'Acoustique de l'Universite du Maine ( I wish they had a website). A fine lab with some 40 researchers, in nice city well known for it's racing circuit - Le Mans. I spent two years there, from August '93 to August '95, truly unforgettable. Industry

I worked for IBM Science & Technology in Haifa, Israel, first as an undergrad student and then as a research fellow, between 1984 and 1986. I was in the speech processing group, run by Yoav Medan, working on a speech compression algorithm. It worked.

Contact Information