Re: Harmonic Analysis on f0-Curve (Amit Gal )


Subject: Re: Harmonic Analysis on f0-Curve
From:    Amit Gal  <hobbit(at)PLUTO.MSCC.HUJI.AC.IL>
Date:    Sun, 14 Nov 2004 19:25:17 +0200

Hello, unfortunately, I'm unaware of such a software at the moment. I looked for one too sometime ago, and ended up using SFS and an expensive speech analysis software that is installed in my department's lab. however, I'm now supervising a team of last year CS students that build a muicologically oriented audio analysis software as their final project. they are now gathering requirements and features that the software should be able to perform. I, of course, gave them some basic requirements and further features I needed in the past, but they are open to hear any further needs and features that might make the software more useful. again, the orientation is musicological - that is, extraction of musical features from audio signal that will enable analysis of performances. the project is divided to two parts. the main one is the analysis of monophonic performances, and another branch is the analysis of polyphonic signals which is much more complex, and we are not aiming to implement only a reduced set of features to this branch because of that complexity. i join the request of hermann: if you know of any musicologically oriented performance analysis software, I would very much like to know more about it. however, if you don't - please feel free to share what do you think such software should be able to do. the students' intention is to release the software as an open source freeware - hopefully by the end of the academic year. thanks, amit gal. Quoting Hermann Stemberger <hermann(at)STEMBERGER.AT>: > Hello! > > I'm a student (from Salzburg/Austria) searching a windows based program that > computes a pitch out of a singers vocal wav file. > Furthermore I should be able to change that f0-plot for correction. > What I want to do then is a harmonic analysis on voice. So the program > should compute out of that f0-curve the harmonic levels in dB over time > (window size between 256 and 4096). A pure graphical solution is not > accurate enough, i want to get it in a range of +- 3dB on e.g. the first 12 > harmonics. > > And I should also be able to set the frequency range that should be > processed. > > Does anyone know a program that would do that for me? I tried Praat > (installation failed...), SFS and Wavesurfer. Those seem to be very > speech-orientated, for me it should be rather musicology orientated - > primarily a good set of analysis-options on harmonics. > > Is there even a freeware educational program that does it? > > Best regards, > Hermann Stemberger > ------------------------------------------------- This mail sent through IMP: http://horde.org/imp/


This message came from the mail archive
http://www.auditory.org/postings/2004/
maintained by:
DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University