On Fri, Apr 24, 2015 at 5:12 AM, Sha Xin Wei <firstname.lastname@example.org> wrote:
Where can we get the best publicly available HMM external for Max
as a general purpose HMM package
Should we extend / modify gf (which we have via IRCAM license )
and can we use it easily for non-audio data? People claim to have tried it on video.
It seems that the real work is the preliminary feature extraction where a lot of interpretation happens
What are examples of code that do this in interesting ways?
Navid Navab wrote
While FTM is somewhat discontinued, this all is being moved to IRCAM's free Mubu package:
download the package and quickly check some of their example patches.
It contains optimized algorithms building on gf, FTM, cataRT, pipo, etc. While mubu is audio-centric, it is not necessarily audio-specific. mubu buffers can work with multiple data modalities and use a variety of correlation methods to move between these layers... This makes up a fairly wholesome platform without the need to move back and forth between gf, FTM, concatenative synthesis instruments, multimodal data handling, analysis, and etc.
As with most current IRCAM releases, it is highly under-documented. Besides gf that is distributed with their package, the mubu.hhmm object might be good place to start for what you are looking for:
also their xmm object might be of interest: