---------------------------- Original Message ----------------------------

Subject: Re: [music-dsp] Antialias question (Kevin Chi)

From: "Kevin Chi" <s...@finecutbodies.com>

Date: Fri, June 1, 2018 2:50 pm

To: music-dsp@music.columbia.edu

--------------------------------------------------------------------------



> Thanks Frank,

>

> I use cubic interpolation to interpolate between samples and it seems a

> bit better than linear for me. I am more worried

> about the downsampling, when the playhead is going faster than the

> original samplerate
or the output pointer ("playhead") is moving faster than the input pointer 
("recordhead").� the latter is always moving at a rate of one sample 
displacement per sample period.� the former, the output pointer has both 
integer and fractional
parts.� the integer part of the output pointer tells you which samples you are 
going to combine, and the fractional part tells you how you will be combining 
them.

>� and that's

> when the high frequencies of the original material start to fold back at

> Nyquist to cause aliasing.

>

> My concern is if it would be a constant pitching up (downsample) rate

> then it would be "easy" to apply an

> antialias filter based on the resampling rate. But as the resampling

> rate can change by every step (as the speed of the

> playhead is modulating), you should apply different antialias filters

> for every step. Or at least the filter needs to use

> different coefficients every time it's applied... That's what I am

> unsure of what is the best solution to solve.
�
what you can do, but it's a little expensive, is when your output pointer 
stepsize is larger than one, cut that step size in half and compute two 
samples.� lowpass filter that stream of samples and pick out every other
sample, discarding the samples in between.
�
--

r b-j� � � � � � � � � � � � �r...@audioimagination.com



"Imagination is more important than knowledge."

�
�
�
�
_______________________________________________
dupswapdrop: music-dsp mailing list
music-dsp@music.columbia.edu
https://lists.columbia.edu/mailman/listinfo/music-dsp

Reply via email to