[LAD] seeking "fresh" way to process/shape human whisper in real-time

Jörn Nettingsmeier nettings at folkwang-hochschule.de
Thu Dec 23 12:05:38 UTC 2010

On 12/22/2010 08:37 PM, ico at vt.edu wrote:
> Hi all,
> I've been battling a kind of a dsp-writer's-block as of late. Namely, I am
> dealing with a project where (at least as of right now) I would like to explore
> human whisper and its percussive/rhythmic power. This would take place in an
> ensemble of "voices." I am also looking to combine whisper with some sort of
> DSP. Obviously vocoder comes as one of the obvious choices but it sounds IMHO
> cliche and as a result I would like to avoid it as much as possible (unless I
> can somehow come up with a cool spin on it which I haven't yet). I also tried
> amp mod, additive, filtering, etc., but none of these struck me as something
> interesting. I do think delays will be fine in terms of "punctuating" the
> overall pattern but I think this should take place at the end of the DSP chain.
> Granular synthesis is also a consideration but I've done so much of it over the
> past years I am hoping to do something different.
> So, as of right now I have:
> 1) whisper
> 2) ???
> 3) delays
> 4) profit! :-)
> Given the mental constipation I have been battling particularly over the past
> couple of days, I wanted to turn to you my fellow LA* enthusiasts for some
> thoughts/ideas/inspiration. Your help would be most appreciated and I will
> gladly credit your ideas in the final piece.

spectral delay (freqtweak), if only to explore the details of the source

and maybe low-pass filtering with extreme compression, to look for
drone-like sounds where you woulnd't expect them?

More information about the Linux-audio-dev mailing list