[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
thank you very much for the paper and code links, I'am going to try various
Currently I modified the implementation done by Ning Ma :
I wondered if I could split the filterbank into 2 parts, one high frequency and
one low frequency (downsampled). But that will complicate the outcome a lot,
as I use the filterbank also for onset detection. I'll try it as a last resort,
but it seems a good idea.
I would be really interested in alternatives, but I wonder what do you mean by
better solutions. I am particularly interested in the phase output of the bank
and the time resolution. If you can suggest me a potentially beter solution I
would be really grateful.
I am planning to participate in the MIREX competition so I am going to reveal
more details there.