[vorbis-dev] Transient coding: AAC vs. Vorbis

[Sebastian Gesemann]

On Wed, 4 Aug 2004, Stephen So wrote:

> Sebastian Gesemann wrote:
>
> >(3) MDCT+TNS
> >    a) just a single MDCT like already done in Vorbis I
> >    b) linear predictive coding of the MDCT samples whereas the
>
> Just curious, has TNS been patented yet?

Don't know. TNS is nothing else than linear predictive coding
like it's done in Speex - just on the MDCT samples instead of
time samples.

Although it's so similar, the idea of applying this technique
in the frequency domain might be protected. (?)

AFAIK there's also a tool called SNS (spatial noise shaping)
which is pretty much the same for image coding (like JPEG, to
minimize the gibbs phenomena)


Ghis!
Sebastian

--
PGP-Key-ID (long): 572B1778A4CA0707