Re: [Audacity-devel] USING AUDACITY FOR FREQUENCY ANALYSIS ON SPATIAL DATA
A free multi-track audio editor and recorder
Brought to you by:
aosiniao
From: Michel DE R. <mde...@va...> - 2008-11-18 10:41:12
|
Sorry Gale, Alan, Gary & Michael, I was on a long business trip in some unfriendly countries and could not address my mail; to make it worse, our internal mail system apparently discards some mail.. The fact is I have never used a forum to discuss technical things before and have no idea on how it works nor who conducts the symphony.. To make things simple, we are struggling with data analysis problems from the fields (technical data related to "topography")and feel there would be a field to explore in expressing these information in terms of "spectrum"; I have found audacity to be extremely versatile yet simple from the user standpoint.. You have to understand the position of engineers inside corporates, we are not as smart as you guys, we need to spend hours and hours to demonstrate our bosses we need to hire or use some money to get things done and we get a tremendous pressure from the sales & marketing to use our brain with little money. So, my first mail: 1- do you think this non-musical data can be imported and displayed by audacity as a spectrum to display frequencies? And of course: 2- if so, how could our company contribute to the audacity project? By no means it is a corporate "team work" : you work, I play golf, we do not want to get something for nothing, but rather, we need help, we do not know how your system works, but there will be enough bananas for the monkeys.. It seems easier for me to get things done using smart & free people like you guys than relying on our internal "think tank" Thank you for your help, and let me know what could be a fair price or contribution in $ for such a project, I have no intention to make you work and say later on, sorry guys, my bosses do not want to pay.. Have a good day, and sorry again for the late answers. Best regards michel de reynal -----Original Message----- From: Gale Andrews [mailto:ga...@au...] Sent: Wednesday, November 05, 2008 01:09 To: Michel DE REYNAL Cc: aud...@li... Subject: Re: USING AUDACITY FOR FREQUENCY ANALYSIS ON SPATIAL DATA Michel, Gary Nelson replied to you on aud...@li... as below. Did you see it? I did not Cc it to you because you appear to be subscribed to audacity-devel. Gale >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> "Michel I imported your file to OpenOffice Calc to see what it looks like. In order to use Audacity, you'd need to normalize your data so that the largest value is is +1 (.wav files have all audio samples in range +1 to -1). You will also want to scale the spatial sample separation to look like a sampling freq like 44100. Scale factor looks to be 294000. See attached excel format file. You ought to be able to normalize and convert to a stereo .wav format and then import into Audacity as a test. I can't offer a specific method to get that done, but programs like MATLAB work with csv and wav files. I am just getting started with the open source version of Matlab called Scilab so can't tell you that program will do the job. But it can be done so it's a matter of doing some digging. I agree that Audacity has great potential for use in your application and others. In the near term, I think the easiest path is as above. -- Gary Nelson Porpoise Research Institute Port Townsend, WA, 98368" | From "Michel DE REYNAL" <mde...@va...> | Mon, 3 Nov 2008 11:33:53 +0100 | Subject: USING AUDACITY FOR FREQUENCY ANALYSIS ON SPATIAL DATA > Hi gale, it is not so much the export thing, than a way of visualizing > data that is made of surimposed sinusoids (extracting the frequencies > through fft): > > Data flow is data flow, let's call a time span a physical distance now > in yards or meters and that's it: > > 1-Time is now called Distance (first field) > 2-Signal is now called Track-1 > 3-Level (DB) is now called Track-2 > > Any delimiter can be used; on the example I sent you it was semicolon > delimited because I thought this would be better (excel can delimit data > fields using anything..) > > What I like in you program is the ability to shift from wave forms > through pitch to sonogram and be able to change the (envelope) to > correct db fading-out. > > From this perspective, inputing a sound or a data file is not very > different if we know where the plate, the egg and the spoon are; syntax > is the only disturbing thing, as I have no idea of the structure of a > wave file and how it is parsed in audacity, three variables are present > (triplet): a time axis, a signal and a level.. there should be not be a > big deal then.. > > Looking forward.. > > Best regards michel > > > > -----Original Message----- > From: Gale Andrews [mailto:ga...@au...] > Sent: Wednesday, October 29, 2008 19:57 > To: Michel DE REYNAL > Cc: aud...@li... > Subject: Re: USING AUDACITY FOR FREQUENCY ANALYSIS ON SPATIAL DATA > > > | From "Michel DE REYNAL" <mde...@va...> > | Tue, 28 Oct 2008 10:17:25 +0100 > | Subject: USING AUDACITY FOR FREQUENCY ANALYSIS ON SPATIAL DATA > > Your software seems so advanced that I wonder if you could use it for > > purposes other than music or sound (that I love too of course..), say, > > to analyze streaming data.. > > > > I have attached a recording of two tracks over a certain distance > > (spatial survey of two variables); as you can see, it is very similar > to > > a sound on its graphic form, so it should not be too complicated to > > allow your program to input ascii data instead of millivolts v.s. > time; > > these two sets of data plotted against a common distance have some > kind > > of relationship (track-2 varies in the same way than track-1, but do > not > > have the same physics behind nor the same dynamic response..) but near > > the end of the survey, track-2 has definitely something going wrong, > but > > there is still some relative information there.. > > > > So, the question is: > > > > * can we load this file into audacity to get frequency / pitch / > > waveform display? > > * Is it possible to adjust the two tracks so we can spatially match > them? > > (sometimes, the common distance reference is not so common..) > > * Can we invoke some kind of adjusting formula so the two curves / > > frequency maps can be surimposed? > > > > This work is a very interesting one for us, this would definitely be a > plus for > > audacity and could help you get some substantial money for your > > organization. > > No, I don't think you can do anything with that file in Audacity at > present, though modifications to do so would I guess be possible. > > Audacity can currently export a single-track, tab-delimited text > file from an audio waveform with Frequency (Hz) plotted against > level (dB). It cannot yet import such a file back in. > > I've Cc'd this to our developers' mailing list in case anyone wishes > to comment in detail. Michel is not currently subscribed, so anyone > replying will need to include him as a Cc. > > > Thanks > > > Gale > > |