|
|
 |
Re: FN-FORUM Flash audio analysis
date posted 25th June 2002 09:01
I don't know if this is of any use mate
http://www.swift-tools.com/
Go to the Swift MP3 section, you may well have seen it already but if not it
generates a SWF with a spectrum analyzer by well analyzing the original MP3.
It may be possible to customise in some way but I have never used it so
can't really say.
Out of interest where did you find the systems for constructing VU meters?
Steve
on 25/6/02 1:51 AM, Ade(funkyfish) at [EMAIL REMOVED] wrote:
> Can the panel answer the following query:
> (many thanks for any who even attempt it)
>
> I am currently designing a site predominantly in Flash (95%), the idea
> being to showcase several groups of experimental musicians and artists.
>
> To cut to the chase, each artist will have a selection of music and/or
> video clips to be played by streaming in a movie clip within the main
> Flash-site. Some of the music clips do not have an associated
> visualisation and to avoid sticking a stationary or animated track title
> or whatever it has been mooted to attempt to construct some form of
> analytic visualisation system akin to winamp or similar. The 50p
> question is how the hell do I analyse the audio stream in Flash? I know
> it can't be done in ActionScript (I'd like to be proved wrong, but I
> don't think so somehow), could you have some sort of server-side script
> or executable generate the data and throw the values to Flash in
> real-time? Or would it be easier to pre-generate the data and construct
> an algorithm (or algorithms) to process it later - given that, by it's
> nature, the music on the site will be static (I just hope they don't
> want a radio stream doing as well)?
>
> It isn't really an option for a human to generate a set of
> visualisations for each track as this (kind of) defeats the object
> (apparently), a compromise is to get the computer to generate them -
> damn I knew I should never have sent them winamp...
>
> This is not about syncing to BPMs or anything, this is about analysing
> the stream to establish frequency groupings. The music does not have a
> "4/4" easily defined rhythm or any set BPM for that matter. We already
> have suggestions for a few algorithms to work on the data, just need a
> good way of generating the data... actually implementing the algorithms
> is another question for another day...
>
> I've seen a few different systems for constructing VU meters but they
> only "analyse" the relative dynamics - although actually quite useful in
> a way...
>
> Any suggestions???..please bear in mind that my programming consists of
> "goodbye cruel world" in BASIC...;-) I have searched the usual suspect
> Flash sites and done a general web trawl for stuff but came up with
> nothing that tells me stuff I haven't already worked out...
>
> !THANK YOU THANK YOU THANK YOU FOR ANY HELP!
>
> ade
>
> P.S. If all else fails they'll get a random test-card pattern...
|
 |
|