This work is based on the simple Haar transform which has a very
simple, narrow basis. There is likely significant aliasing here. Have
people tried other more developed bases? Or even filter banks that are not
wavelets like the pyramid transforms for visual data?
________________________________________________
From:"Jamie Bullock" <ja...@jamiebullock.com>
Sent:"A discussion list for music-related DSP"
<music-dsp@music.columbia.edu>
Cc:"music-dsp@music.columbia.edu"
Date:Thu, July 10, 2014 12:37 am
Subject:Re: [music-dsp] Frequency based analysis alternatives?
>
> There is Antoine Schmitt's Wavelet-based method, implementation
available
> under an MIT license:
>
> https://github.com/antoineschmitt/dywapitchtrack
>
> It is optimized for voice, but I have found it generally works well
with
> PNP sounds.
>
> Jamie
>
>> On Jul 9, 2014, at 1:03 PM, "Rohit Agarwal"
<ro...@khitchdee.com> wrote:
>>
>>
>>
>>
>> Most of our modern DSP techniques that we use for the analysis of
sound
>> signals are based on the FFT as a first step. This imposes limits
on
>> time
>> resolution since the FFT window has to be wide. For most natural
sound
>> apps this is no hindrance as the rate of events is commonly slow.
Speech
>> recognition is such an example. Even in the music space for the
most
>> part
>> the required time res for most common apps is not that great so
FFT
>> suffices.
>> What are the alternatives to the FFT? Have wavelets been
>> used for real world solutions? If an app needs much higher
time
>> resolution
>> and there are limits on sampling frequency, what kind of time
domain
>> techniques are well known?
>> --
>> dupswapdrop -- the music-dsp mailing list and website:
>> subscription info, FAQ, source code archive, list archive, book
reviews,
>> dsp links
>> http://music.columbia.edu/cmc/music-dsp
>> http://music.columbia.edu/mailman/listinfo/music-dsp
> --
> dupswapdrop -- the music-dsp mailing list and website:
> subscription info, FAQ, source code archive, list archive, book
reviews,
> dsp links
> http://music.columbia.edu/cmc/music-dsp
> http://music.columbia.edu/mailman/listinfo/music-dsp
>
--
dupswapdrop -- the music-dsp mailing list and website:
subscription info, FAQ, source code archive, list archive, book reviews, dsp
links
http://music.columbia.edu/cmc/music-dsp
http://music.columbia.edu/mailman/listinfo/music-dsp
