[hpsdr] Audio resampler

[Alan Hopper]

Warren,
my resampler is definitely low fidelity and is a basic linear interpolation
between audio samples, however I've never noticed an audio problem even in
prolonged listening to fm broadcast stations through it.  I did try going
even simpler and just dropping or repeating individual samples, this did
sound horrible.   For IQ data you obviously need something better although
I have a deep hatred of using vac for anything like this.  A moving average
of 10secs is a good starting point and will work.
I look forward to your improved resampler and may well incorporate it in my
code once it is open source.
73 Alan M0NNB

On Sat, Jun 10, 2017 at 5:43 PM, Warren C. Pratt <warren at wpratt.com> wrote:

> ***** High Performance Software Defined Radio Discussion List *****
>
> Hi Alan,
>
> This is very helpful.  Just a couple quick comments/questions:
>
> * I will need to write a "high-fidelity" resampler that will produce an
> accurate result on a wideband I-Q data stream.  You said you used 'linear
> interpolation'.  Were you referring to linearly interpolating between
> samples, or, linearly interpolating between filter coefficients?
>
> * The PortAudio problem you ran into is interesting.  I seem to remember
> that someone else ran into a problem with not letting PortAudio choose the
> buffer size.
>
> * I was thinking of using a simple (slow) moving average for filtering the
> jittery data.  Good to know that works for you.
>
> * There are several great ideas and some identified pitfalls in your
> notes.  Thanks for sharing your experiences!
>
> 73,
> Warren  NR0V
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help: http://lists.openhpsdr.org/lis
> tinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20170610/d7761d47/attachment.htm>