Vorbis dev - Aug 2005 - Reg. vorbis for real-time audio

Hi,

 From the vorbis decoder specification, it is clear that the decoder 
needs to have all  the codebooks before decoding can actually begin.
I will appreciate if someone can clear the following questions:

1. I guess the codebooks are derived from the actual input data. 
Probably the encoder may be making two passes through the input. The 
first pass finds out the frequency of different symbols and the second 
pass just replaces these symbols with the codewords.
If the above is true:
I wonder how vorbis encoder can be used for real-time data? Will it use 
some known codebooks or will it use someform of adaptive coding?

Any help appreciated.
Shashank

Shashank Khanvilkar wrote:
> Hi,
>
> From the vorbis decoder specification, it is clear that the decoder
> needs to have all  the codebooks before decoding can actually begin.
> I will appreciate if someone can clear the following questions:
>
> 1. I guess the codebooks are derived from the actual input data.
> Probably the encoder may be making two passes through the input. The
> first pass finds out the frequency of different symbols and the second
> pass just replaces these symbols with the codewords.
> If the above is true:
> I wonder how vorbis encoder can be used for real-time data? Will it
> use some known codebooks or will it use someform of adaptive coding?
>
> Any help appreciated.
> Shashank 
Last I checked oggenc does not try to invent a codebook from the source
audio but picks its codebook depending on the level of compression
that's desired. So two files encoded with oggenc -q3 will have the same
codebook. This is distinct from the huffman coding that's applied to the
data afterwards.

Clever people then try to invent better sounding codebooks.

- Daniel Holth

On Fri, Aug 26, 2005 at 01:35:21PM -0500, Shashank Khanvilkar wrote:
> 1. I guess the codebooks are derived from the actual input data. 
> Probably the encoder may be making two passes through the input. The 
> first pass finds out the frequency of different symbols and the second 
> pass just replaces these symbols with the codewords.
It's certainly possible to write a two-pass encoder, but the one in
libvorbis works in a single pass. The codebooks are tuned for general
audio across many test files during development, and fixed per release.
The encoder "chooses" a set of codebooks based on the target 
quality/bitrate and then just makes choices of which available mode
to use for each block on the fly, I think with a 3 block look-ahead.

So it's not adaptive in the usual sense in that the codebooks are 
fixed at the start. But emphasis on codebooks plural lets the 
encoder tailor mode choices from that fixed set to match particular
input data and so encode in a single pass.

Note that the look-ahead, and the generally long block sizes, and
the latency of the Ogg container itself make vorbis a poor choice
for interactive audio streams. But for "live" broadcast where a
few seconds delay is in fact normal, it works great.

Hope that helps,
 -r

Hi,

is there any work done on "Tarkin" ? Is there a mailing list or
homepage
somewhere ?

--HPS